Vous êtes sur la page 1sur 383

SYMPOSIUM INTERNATIONAL

GiODYNAMIQUE ANDINE

15 - 17 mai 1990
Grenoble, France

Rhum& des communications

&Lp.+i$t
..::::,...
:::
.,.,
::..
..:::;.
.:::,:.t.,:f:::
.::::::.
;;::::
.,:I!.
:...:
.: ,.
i
. :
.*

1.

.*
: ,,

*:
. .

69%:.
; .:.

Organisateurs
INSTITUTFRANCAIS DE RECHERCHESCIENTIFItlUE UNIVERSITt JOSEPH FOURIER
POUR CE D&ELOPPEMENT EN COOPtRATION GRENOBLE.FRANCE
ORSTOM. PARIS

Sponsors

Comitk National Frangais du PICG MINISTtRE DES AFFAIREStTRANGERES Ville de Grenoble


Paris

Editions de IORSTOM
INSTINT FRANGAlS DE RECHERCHE SCIENTIFIQUE POUR LE DhELOPPEMENT EN COOPl%ATION
CollectionCOLLOQUES et StMINAIRES
PARIS 1990
Collaborations

La coordination pour Iorganisation de ce premier Symposium


international sur la Geodynamique Andine a ete assuree par Gerard Laubacher
(ORSTOM), Richard A. Olivier (lnstitut Laue-Langevin, Grenoble), Nicole
Vatin-Perignon (Institut de Geologic, Grenoble) avec Iappui de Louis Dorbath
(ORSTOM-IPG de Strasbourg), Armando Cisternas (IPG de strasbourg) et Jean
Delfaud (universite de Pau).

Corganisation du Symposium a beneficie dun large soutien financier et


materiel de IORSTOM (D.S.T., Departement T.O.A. et C.S.l), de Wniversite et de
la Ville de Grenoble, sans lequel le Symposium naurait et6 possible.

De plus, des credits speciaux, mis a la disposition du Comite


dorganisation par IORSTOM, le Ministere de Affaires Etrangeres et le Comite
National Frangais du PICG (UNESCO), ont permis de participer aux frais de
voyages de plusieurs chercheurs sud-americains inscrits au Symposium.

Enfin, nous remeicions toutes les autres personnes ou institutions qui,


dune faGon ou dune autre, auront permis le bon dkroulement de ce Symposium.

La loi du 11 mars 1957 nautorisant, aux termes des ah&as 2 et 3 de Iarticle 41, dune part,
que les -copies ou reproductions strictement her&es A Iusage priv13du copiste et non des-
tinbes A we utilisation collecthen et, dautre part, que les analyses et les courtes citations
dans un but dexemple et dillustration, etoute reprhntation ou reproduction intbgrale, ou par-
tielle, faite sans le consentement de Iauteur ou de ses ayants droit ayants cause, est illiciten>
(alin~aler de Iarticle 40).
Cette repkentation ou reproduction, par quelque pro&% que ce soit, constituerait done une
contrefa$on sanction&e par les articles 425 et suivants du Code penal.

ISSN : 0767-2896 0 ORSTOM 1990


ISBN : 2-7099-0993-6
SOMMAIRE

GEOPHYSWE paw

Fuenzalida A.. Etude de la subduction au niveau du Chili central. 3

Comte D., Pardo M., Dorbath L., Dorbath C., Haessler H., Rivera L.,
Cisternas A., Ponce L.. Crustal seismicity and subduction morphology around
Antofagasta, Chile: preliminary results from a microearthquake survey. 5

Ramirez L.D.. Estimation of some focal parameters of an historical chilean


earthquake. 7

James D.E., Snoke J.A.. Seismic studies of deep slab morphology beneath
Central Peru. 13

Lindo R., Dorbath C., Cisternas A., Dorbath L.. Microsismicite de la


subduction au Perou Central. 17

Rodriguez L., Tavera J.. Determinaci6n con alta resolution de la geometria


de la zona Wadati-Benioff en la parte central del Peru y determination de efuerzos 19

Arellano M.A., Haesster H.. Etude de la subduction de la plaque Cocos au niveau


du Nicaragua. 21

Araneda M., Avendano MS.. Modelos gravimetricos en el sector del volcan


Villarica associados a la megaestructura Liquine-Ofqui. 23

lntrocaso A., Pacino M.C., Fraga H.. Gravity, isostasy and andean crustal
shortening between 30% and 35% latitudes. 29

Barrientos S.E.. Postseismic land movements in South-Central Chile. 33

Ritz M., Herail G., Bondoux F., Semperd T.. Magnetotelluric inves-
tigations in the northern Altiplano of Bolivia. 37

Schwarz G., Chong D. G., Kriiger D., Martinez E., Massow V., Rath V.,
Viramonte J.. Crustal electrical resistivity structure in the southern Central
Andes. 41

Wigger P.J., Araneda M., Giese P., Heinson W.D., Rower P.,Schmltz M.,
Viramonte J.. Crustal structure of northern Chile and NW- Argentina. 45

Tetleria J.L.. Bolivian global geosciences transects project. 47

Ballesteros J., Berrocaso M., Catalan M., Cruz F., Estrada R.,Lujan A.,
Munoz J., Sanchez det Toro J.,Sastre J.C., Soto R., Viramonte J..
Spanish-argentinian geodynamic GPS net Antartic project. 49

Golombek M.P., Dixon T.H., Stein S., Gordon R., Sacks S.. A global
positioning system experiment in the Andes. 53
B

MacFadden B.J.. Paleomagnetism of late Cenozoic andean basins and comments


on the bolivian orocline hypothesis 57

Hartley A., Jolley E., Turner P.. Paleomagnetic and structural constaints on
Mesozoic-Recent thrust sheet rotation in the Precordillera of northern Chile. 61

Rey D., Turner P., Hartley A.. Tertiary ignimbrites from northern Chile:
anomalous magnetization explained by self-reversal and tectonic rotalion. 63

Kissel C., Lai C., Mitouard P., Macedo O., Roperch P., Surmont..
Paleomagnetic study of Upper Crelaceous and Tertiary formations from the Central
Andes (Southern Ecuador and Peru). 67

Ropperch P., Carlier G.. Paleomagnetism of Jurassic rocks and Cretaceous


plutonic rocks from coastal southern Peru. Importance of rotations in the Arica
deflection. 71

NEOTECTONIQUE

Dewey J.F., Lambs S.H.. Andean displacement and strain partitioning of the
Nazca-South America slip vector during Ihe lasl 5Ma. 77

Mercler J.L., Sebrler M., Lavenu A., Cabrera J., Bellier O.,
Dumont J.F., Machare J.. Changes in the tectonic regime above a subduclion
zone of andean type: the Andes of Peru and Bolivia during the Plio-Pleistocene. 79

Buddln T.S., Stimpson LG., Willams G.B.. Neotectonics within the


continental fore-arc of northern Chile. 63

Flint S., Turner P., Jolley E.. Neotectonic controls on fan-delta sedimentation,
coastal northern Chile: a response lo aseismic ridge subduction. 65

Flint S., Jolley E., Turner P., Williams G.D., Buddin T.. Subduction
of aseismic ridges at the andean margin: a major facture in the sedimentological
and structural evolution of fore-arc basins 69

Machare J., Orttieb L.. Recent vertical motions and the subduction of the Nazca
ridge, Central coast of Peru. 91

Ortlieb L., Machare J.. Qualernaty marines terraces on the Peruvian coast and
recent vertical motions. 95

Strecker M., Bloom A., Mallzia D.. Neotectonic activity in the northern
Sierras Pampeanas, Argentina. 99

H&ail G., Fornarl M., Miranda V., Vlscara G.. Geodynamics of the bolivian
Andes and formation of the alluvial gold deposits. 103

Dumont J.F., Garcia F.. Neotectonics of the andean foredeep (Maranon basin) in
northeastern Peru. 107

Bellier O., Sebrier M.. Late Cenozoic normal and strike-slip faulting in north
Peruvian Western Cordillera: an example of alternate extensional and compres-
sional tectonic regimes in High Andes. 111

Lavenu A., Winter Th., Avouac J.Ph.. Premiers r&ullats des Eludes de failles
actives dans les Andes dEquateur. 115
C

Tibaldl A., Ferrari L.. A wedge model for the Quaternary tectonics of the Andes
of Ecuador. 119

Bles J.L., Marin W., Paris G., Sauret B., Vergara H.. Neotectonique du
sud-ouest de la Colombie: resuitats microtectoniques et application a letude de Ialea
sismique sur le site de Popayan. 123

TECTONlQUE

Storey B.C., Alabaster T.. Late Jurassic ridge-trench collisions and develop-
ment of Gulf of California-type basins along the Andean Cordillera. 127

Solianl E.Jr., Bonhomme M.G.. Nouvelles donnees montrant une remise a zero
g&n&ale de Ihorloge isotopique K-Ar dans la partie nord de la baie de IAmiraute.
tie du Roi George, Antartique. 131

Pincheira M., Thtele R., Fontbote L.. Tectonic transpression along the
southern segment of the Atacama fault-zone, Chile. 133

Omarini R.H., Goetze H.G.. Antofagasta-Salta-Chaco plains transect, Central Andes. 137

Charrier R., Thiele IT., Arcos R., Malbran F., Tectonic evolution of the
Chilean main Cordillera between 33 and 35 south latitude. 141

Reynolds J.H., Jordan T.E., Tabbutt K., RB G., Vilas J., Bercowski F.,
Milana J.P., Beer J.A.,Damanti J.F., Ramos V.A.. Chronology of foreland
deformation in the Precordillera-Sierras Pampeanas region of Argentina
(28-33 S). 145

Scheuber E., Reutter K.J.. Magmatic arc tectonics and their appearence at
different structural levels: examples from northern Chile. 147

Lucassen F., Franz G.. Geology and metamorphic evolution of igneous rocks:
Coastal Cordillera of N-Chile between 2325 and 2420 S. 151

Scanlan P.M., Turner P.. Sedimentological tectonic and paleomagnetic implica-


tions of the Middle Jurassic of Caieta Camarones and Quebrada Chiza, northern Chile:
an oceanic back-arc basin margin. 155

Laubacher G.. Mise en evidence de structures tangentielles andines dans le sud


de la Bolivie (region de Tica Tica, departemento de Potosi) par Spot. 159

Baby P., Sempere T., Oiler J., Blanc0 J., Zubieta D., Herail G.. Evidence
for major shortening on the eastern edge of the Bolivian Altiplano: the Calazaya nappe. 163

Sempere T., Herail G., Oiler J., Baby P., Barrios L., Marocco R.. The
Altiplano: a province of intermontane foreland basins related to crustal shortening
in the Bolivian Orocline area. 167

Farrar E., Clark A.H., Heinrich S;M.. The age of the Zongo pluton and the
tectonothermal evolution of the Zongo-San-Gaban zone in the Cordillera Real, Bolivia. 171

Reinhardt M., Kley J.. Tectonic and geothermal history of the Subandean Ranges
and the eastern Cordillera of southern Bolivia. 175
D

H&ail G., Baby P., Lopez M.,Oller J., Lopez O., Salinas R., Sempere T.,
Beccar G.,Toledo H.. Structure and kinematic evolution of Subandean Thrust
System of Bolivia. 179

Sheffels B.. Structure of the central Bolivian Andes. Implications for orogeny
in Andean-type margins. 183

Soler P., Sebrler M.. Nazca slab retreat versus compressional deformation in
the Central Andes since late Oligocene times. 187

Munoz C., Fontboto L.. Relationships between the Cochas-Gran Bretana reverse
fault and the Azulcocha Zn-As-(Au) ore deposit, Central Peru. 191

Atherton MP., Petford N.. Deformation and granitold emplacement along


a major crustal lineament: the Cordillera Blanca, Peru. 195

Lltherland M., Aspden J.A.. Evidence for pre-Cretaceous collision in the


Ecuadorian Andes. 199

Van Thournout F., Quevedo L.. Allochtonous terranes in northwestern Ecuador. 203

Aguirre L.. Metamorphism of the Celica formation, SW Ecuador: geotectonic


implications. 207

Lavenu A., Noblet C., Winter Th.. Neogene stress pattern in southern Ecuador. 211

Cobbold P., Richard P., Ulloa C.. Cenozoic thrusting and wrenching in the
Cordillera Oriental, Colombia: Field data and experimental insights. 215

van der Wlel A.M.. Uplift age of the Garzon Massif (eastern Cordillera, S.Colombia)
in relation to the infill of the adjacent S. Neiva basin. 217

Toussaint J.F., Restrepo J.J.. Cronologia de las acreciones de terrenos aloctonos


en los Andes colombianos. 219

BASSINS

Rosenfeld U.. Aspects of Mesozoic basin development in southern South America. 225

Bahlburg K.. The Ordovician Puna basin of NW Argentina and N Chile: from back-arc
to foreland basin. 227

Williams G.D., Turner P., Flint S., Stimpson I., Jolley E.J., Hartley A.J..
Andean basin dynamics in northern Chile. 231

Bell C.M., Suarez M.. Triassic depositional basins in northern Chile. 233

Oiler J., Sempere T.. A fluvio-eolian sequence of probable middle triassic-jurassic


age in both Andean and Subandean Bolivia. 237

Martinez C., Vargas E., Laubachdr G.. Evolution tectono-sedimentaire dans


le C&ace du synclinal dOtavi-San Lucas (Bolivie centre-sud). 241

Bogdanic T., Diibel R.. Cretaceous and early Tertiary in northern Chile
between 21 and 23%. 245
E

Charrier R., Reutter K.J.. The Purilactis Group of Northern Chile: link
between arc and back-arc during Late Cretaceous and Paleogene. 249

Wilkes E., Giirler K.. Evolution of the Cordillera de la Sal, Northern Chile. 253

Hillebrandt (von) A., Prinr P., Wilke H.G.. Subsidencia y grado de sedimen-
tacidn en la cuenca mesozoica trh arco del Norte de Chile. 257

Rosas S., Fontbote L.. Sedimentology of the Cercapuquio and Chaucha formations
(Central-Peru). 261

Batty M., Carlotto V., Jacay J., Jaillard E.. The Kimmeridgian (?) -Early
Valanginian tectonic event on the Peruvian Margin. 265

Jaillard E.. Mesozoic extension and crustal thickening in the Peruvian Andes. 269

Marocco Ft., Delfaud J.. Las cuencas continentales de 10s Andes Centrales.
Relaciones con la evoluci6n geodinamica andina. 273

Martinez C., SBguret M.. Les bassins tertiaires de IAltiplano sont-ils des
bassins flexuraux intrachaine? 277

Williams G.D., Stimpson I.G., Daly M.C.. Forearc basin evolution and plate
kinematics: the Northern Andean forearc. 281

Bonhomme M.G., Lavenu A., Noblet C., Dugas F., Eguez A., Vlvler G..
Nouvelles datations WAr sur des roches volcaniques tertiaires et quaternaires
des bassins continentaux intracordill&ains d Equateur. 283

Marocco R., Lavenu A., Noblet C.. La cuenca intramontana en compresi6n


de Vilcabamba (sur del Ecuador). An&sis tectosedimentario. 285

MAGMATISME

Wever H.E., Storey B.C.. Jurassic bimodal magmatism in the Palmer Land,
Antartic peninsula: geochemistry and tectonic setting. 291

Parada M.A., Orslni J.B., Ardilla R., Guerra R., Munizaga F., Berg K..
The plutons rocks of the southern Gerlache strait, Antartica: geochronoloQy,
geochemistry and mineralogy. 293

Kay S., Ardolino A.A., Cortes J.M., Franchi M., Ramos V.A..
Tectonic and geochemical significance of Tertionary Patagonian basalts (40-5OS),
Argentina. 297

Killian R.. The Austral Andean Volcanic Zone (South Patagonia). 301

Hervb F., Pankhurst R.J., Cembrano J., Munizaga F.. Magmatism and
tectonics in the Andes of Chiloe (42-45), Chile. 305

Thlele R., Beccar I., Levi B., Nystrbm J.O., Vergara M.. Tertiary andean
volcanism in a caldera-Qraben setting. 309

M6nard J.J.. Ore deposit genesis in the Chilean iron belt (Atacama area, Chile). 313
F

Viramonte J.G., Petrinovic LA.. Cryptic and partially buried calderas along
a strike-slip fault system in the Central Andes. 317

Schmitt-Rtegraf C., Ptchler H.. Magma genesis and evolution of Central


Andean Cenozoic volcanic rocks in the light of REE and isotopic data. 321

Cartter G., Carlotto V.. Evidence for the origin of a shoshonitic suite by
mixing of peraluminous and ultrapotassic magmas: the Oroscocha and Ouimsachata
Quarternary volcanoes, Sicuani province, Southern Peru. 325

Bonhomme M., Carlier G.. Relation entre magmatisme et mineralisations


dans le batholite dAndahuylas-Yauri (sud-Perou): donnees geochronologiques. 329

Salas A.G., Vatln-Perlgnon N., Poupeau G.. El sillar de Arequipa (Peru):


caracterfsticas de los deprjsitos de flujos piropl&ticos de la quebrada Aiias Huayco. 333

Bigazzl G., Ftores P.R., Pereyra P.A, Poupeau G., Sabil N., Salas G.S.,
Vatin-Perignon N., Villa I.. Datations par TF et K-Ar des verres obsidiennes
(macusanites) de Chilcuno Chico et de Samilia (province de Puno, SE du Perou);
caracterisation geochimique de la nouvelle variete. 337

Cheilletr A., Clark A.H., Farrar E., Arroyo G.P., MacArthur J.D.,
Pichavant M.. Stratigraphy and geochronology of the Macusani ignimbrite field:
chronometer of the Mio-Pliocene geodynamic evolution of the Andes of SE Peru. 341

Andriesoen P.A.M., Reutter K.J.. K-Ar and fission track dating: thermal
histortes and tectonics of igneous rocks in Chile and Argentina. 345

Fontbote L., Calvez J.Y., Pincheira M., Wolf F.. Geotectonic position and
metal sources of stratabound ores in the Central Andes: lead isotopic constraints. 347

Soler P., Rotach-Toulhoat N.. Sr-Nd-Pb isotope constraints on the genesis


of Cenozoic magmatism along a transect of the central Peruvian Andes. 351

Schmitt-Riegraf C..Magma genesis and evolution of Cenozoic volcanic rocks


in the northern Andes of southern Colombia in the light of REE and isotopic data. 355

Breitkreuz C.. Late Carboniferous to Triassic magmatism in the Central and


Southern Andes: the change from accretionary lo an erosive plate margin
mirrors the Pangea history. 359

Schiin C., Miller H.. The evolution of the Lower Paleozoic trondhjemite/granite
suites SW of Cachi, NW-Argentina. 363

Pellitero E., Saavedra J., Rossi J., Toselli A.. Granitoides de Cerro Toro
(sistema de Famatina, Argentina): un ejemplo de 10s procesos de interaction entre
magmas felsicos y maficos en el basamento andino. 367

Acenorota G., Toselli A., Saavedra J., Ross1 J., Peltitero E.. El volcanis-
mo Ordovicico de Las Planchadas (Andes Centrales NW de Argentina) en relation a su
medio geol&$co. 371

Durand F., TosetLi A., Rossi., Saavedra J., Pettitero E.. El complejo
mafico-felsico de Anguinan (La Rioja, Argentina) y su significaci6n dentro del
plutonismo de Paleozoico Inferior andino. 375

Schalamuk I., Arrospide A., Toselli A., Saavedra A., Fernandez R.,
Echeveste H.. El pluton mineralizado de Mazan (La Rioja, NW de Argentina)
y su relacidn con otros granitos de paragenesis estannffera y afines del basamento. 379
G

Kroonenberg S.B.. Geochemistry of Garzon granuliles in the Colombian Andes :


evidence for Proterozoic calcalkaline magmatism. 383

Huaman R.D, Guillande FL. Estudio del peligro volcanico utilizando m6todos
de teledetecci6n: el voldn Nevado Sabancaya (Arequipa, Peru). 387

Thouret J.C., Murcia A., Salinas R., Parra E., Cepeda H.,
Cantagrel J.M.. Stratigraphy and Ouaternary eruptive history of the
Ruiz-Tolima volcanic massif (Colombia). Implications for assessement of
volcanic hazards. 391

Vandelmeulebrouck J., Thouret J.C.. Utilisation de la teledetection pour


Mtude des produits. processus et Bcoulements volcano-glaciaires du Nevado
del Ruir (Colombia). 395

Thouret J.C., Cantagrel J.M., Cepeda H., Murcia A.. Stratigraphy,


Quaternary eruptive history and hazard mapping at Nevadc del Tolima, Central
Cordillera, Colombia. 397

Thouret J.C., Van der Hammen T., Salomons B., Juvignd E.. Late
Qualernary geology, slratigraphy and paleoecology of the last glaciation in the
Colombian Central Cordillera. 399

GoemansF., Vatin-Perignon N., Oliver R.A., Vfvier G., Briqueu L.,


Salas G.A.. Crustal contamination at the Nevado Solimana, Western Cordillera
of the Central Andes, Southern Peru. 403
GEOPHYSIQUE
3

ETUDE DE LA SUBDUCTION AU NIVEAU DU CHILI


CENTRAL

A. Fuenzalida
Institut de Physique du Globe de Strasbourg
5, rue RenC Descartes, 67084 STRASBOURG

La subduction de la Plaque de Nazca sous le continent


SudamCricain est CtudiCe 2 partir des donnees sismiques
precises. Une experience de terrain sest dCroulCe pendant les
mois de Septembre et Octobre 1986, avec 10 stations portables
qui ont CtC installees pour completer la couverture du reseau
permanent. En ajoutant linformation des catalogues SISRA
(196581) on determine la gtometrie de la subduction.
La tectonique observee est expliquee par les deux types de
subduction qui existent ?t cet endroit, subduction normale au
sud de 33S et subhorizontale au nord de 33s. En surface, on
observe une transition graduelle dune morphologie de vallees
transversales a une de depression centrale, du nord au sud.
On observe Cgalement une absence de volcanisme
quaternaire quand langle de plongee de la lithosphere subduite
est subhorizontal. Lactivite volcanique serait lide B larrivbe de
la lithosphere normalement subduite a une profondeur de
130km.
La distribution en profondeur de la sismicit6 montre que la
transition entre les deux types de subduction observes au nord
et au sud de 33S est graduelle. Les variations de langle de
plongement de la plaque de Nazca se presentent plutdt sous la
Cordillbre et en Argentine que sous le Chili. Les contraintes
montrent un caractere compressif pour les seismes
interplaques, mais il devient extensif dorientation est-ouest
pour les seismes intermtdiaires au dela de 70 km de
profondeur.
5

CRUSTAL SEISMICITY AND SUBDUCTION MORPHOLOGY AROUND ANTOFOGASTA,


CHILE : PRELIMINARY RESULTS FROM A MICROEARTHQUAKE SURVEY.

COMTE, D. (1,3), PARDO, M. (1,3), DORBATH, L. (2), DORBATH, C. (2),


HAESSLER, H. (2), RIVERA, L. (2) CISTERNAS, A. (2) and PONCE, L. (3).

(1) Departamento de Geologia y Geofisica, Universidad de Chile, Casilla 2777, Santiago, Chile.

(2) lnstitut de Physique du Globe de Strasbourg, 5 rue Descartes, 67064 Strasbourg cedex,
France.

(3) lnstituto de Geoflsica, Universidad National Autbnoma de Mexico, Mexico D.F., 04510 Mexico.

During October 1988, 13 analog and 16 digital seismographs were installed in


northern Chile between 22.5 S and 24.5 S and 68.5 and 70.5 W. The purposes were
to monitor the microseismic activity probably associated to the Atacama fault system and
to ascribe the morphology of the subduction zone near the southern edge of the great 1877
earthquake (Mw = 8.8).

The analysis of the analog records provide a total of 553 events (2.0 c ML c 5.0;
rms < 0.5). A set of 49 are shallow depth microearthquakes ( 5 30 km ) which might
be associated to the Atacama fault system : 18 events have epicenters near the Coastal
Scarp southward of Antofogasta. 6 events are distribuited in the Mejillones Peninsula, and
17 events are located nearby the numerous branches of the Salar del Carmen fault
northward of Antofogasta. Otherwise, to the southwest of Antofogasta, where not any fault
branches are recognized, neither a shallow microearthquake was recorded. The actual
distribution of the analog seismograph network did not provide enough resolution to
determine focal mechanisms of such low magnitude earthquakes.

A cluster of shallow events located around 24 S approximately 500 km inland


from the trench, is also observed in a region of active volcanoes.

About 450 events have depths varying from 31 km to 250 km; they permit to
delineate the morphology of the subducting plate. North of 24 S the seismicity define a
plate that dips 30 to the east up to 200 km in depth: south of 24O S and deeper than 70
km the seismicity suggests a change to a more subhorizontal subduction that reach 100
km. After that, an absence of seismicity is well defined, but around 500 km inland,
intermediate depth cluster activity is observed: this activity is associated to the well
known place of the 1950 (Mw P 8.2) normal fault earthquake.
7

ESTIMATIONOF SOME FOCAL PARAMETERSOF AN HISTORICALCHILEAN EARTHQUAKE

RAMIREZ LEON DAVID

UNIVERSIDAD DE SANTIAGO DE CHILE


FACULTAD DE CIENCIA
MATUCANA 28-D
CASILLA 5659 - CORREO 2 FAX 6811422 - TELEX 441674-USACH-CZ

Summary: The objetive of this paper is to present the


determination of some focale parameters such as: Sismlc
magnitude, length of rupture, sismlc moment, etc. which
correspond to an historical earthquake the destroyed the city
of Santiago, capital of Chile, on Ray 13th of 1647. In order
to get our purpose we will use the Information relative to the
macrosismics parameters, meaning for that quantitative8 non
instrumentals, observation which has been extracted from the
current historical biblography the next step was to use this
Information, and a rigorous analysis was applied to some
empiric Information between the focal and Macroeiemicce
parameters which has been gothen from a set of Chilean
earthquakes that happened In the present century and for which
the known instrumental information Is confident.

IBTRODUCCIOI?:

Eete es uno de 10s desastres eismicoe que ha dejado huellae


muy profundae en 10s recuerdos de nueetro Paie, Se hizo
celebre en toda Sudamerica y podemos aeegurar que caracteriza
en cierto modo la hletoria de Chile durante 10s tiempoe de la
colonia. Los documentos origlnalee que se refleren a eeta
catdstrofe son numerosoe, sin embargo todos elloe contienen
muchoe aspectoe eociales y religiosoe que dlficilmente pueden
eer utlllzados en la determlnacibn de algunos Par&metros
Focalee que correepondieron a eete terremoto. Practlcamente
todos 10s documentoe investlgadoe concuerdan en ubicar a este
elemo un dia lunes 13 de mayo entre lae 22:30 y lae 22~45
horae en momentoe que la mayoria de 10s habltantee de la
cludad de Santiago capital de Chile se apreetaba a descanear
de su jornada dlarla (Amunategul, 1882).
Beta cludad fue la que eufrl6 10s mayoree dafioe, debiendo
lamentar la muerte de cientoe de vfctimae entre Eepafiolee,
Crlollos e Indlgenae. El territorio de Chile flJ4
practlcamente destruldo entre 10s rios Choapa y Maule que
8

cruzan ciudades ublcadas al Norte y al Sur de Santiago


respectlvamente.
La Qnica lnformacl6n cuantitativa y de caracter oientifico
que podemos extraer de la bibliografia consultada es la
referente a 10s Par&metros Macrosismicos, es decir
observaciones cuantitativas no Instrumentales, a partir de las
cuales lntentaremos estimar la magnitud de algunos Par&metros
Focales de este sismo.
Nuestra experiencia respect0 a la revisi6n de cartas,
manuscritos y numerosos testimonies ineditos, nos indica que
informaclones tales coma: Descripclbn de dafios, distancia
hasta donde fue sentido el slsmo, duraci6n de1 terremoto
principal, duracl6n de1 period0 de replicas, efectos en el
terreno tales coma fracturas, emanaci6n de gases y aguas
subterraneas, oscllaci6n de1 terreno, actlvidad volcanica, pre
y post terremoto, drea de la zona de r&plicas, etc., son sin
lugar a dudas 10s pocos antecedentes que nos pueden acercar
hacia nuestro objetivo. Morla (1983>, Medina (1928), Gay
(1983,),,Gongora y Marmolejo (1862) . Para ello utilizaremos
relaciones empiricas entre Par&metros Focales y Par&metros
Macroslsmlcos para sismos ocurrldos en Chile en el presente
rigl-o, obtenidos de Ramirez David (1988) y de 10s cuales se
tiene alguna informaci6n instrumental .confiable. La idea es
aprovechar estas relacione6 empiricas entre ambos tlpos de
par&metros, con el objet0 de determinar algunos Par&metros
Focales de1 terremoto ocurrido el afio 1647. Algunas de estas
expresiones empiricas que se lndican en las flguras son
similares a las obtenldas por otros autores para diferentes
regiones de1 mundo, con algunos valores sugeridos por la
teoria..

GEOLOGICAL SETTING

de1 Terremntp: Existen muohas verslones sobre la


duraci6n de1 terremoto principal. Asi las estlmaclones de
tiempo van desde las versiones mas frias, hasta la5 muy
exageradas. Sin embargo, las versiones mas frecuentes y que
estlmamos mas probables, dan una estimaci6n de tlempo
equivalente a la demora en rezar 3 o 4 credos, esto significa
poco mas de unos tres minutes, Vicufia Mackenna <1869), Medina
<1928), Montesus de Ballore (1916) y Gay (1983).
SegSn casi todas las
referenciaa que se refieren a este aspect0 10s limites de
perceptibllidad sensible6 al hombre media se ubicarian en las
cludades de Valdivla por el Sur, Provincias de1 Cuyo Argentino
(Mendoza-San Juan) por el Este y la ciudad de1 Cuzco por el
Horte, Barros Arana (1884), Montesus de Ballore (1916)
Amunategul (1882) y Riquelme Venegas (1905).
e- d-1 T.rrr Exicte
bastante acuerdo en 10s documentos revisados respect0 de la
zona que sufri6 mayor destruccibn ocasionada por el terremoto.
9

As1 practicamente todas las referencias mencionan la zona


comprendlda entre 10s rios Maule y Choapa. Barros Arana
(1884>, Montesus de Ballore (1916), Medina (1928) Y Gay
(1983).
Frecuencias Como consecuencia de este gran terremoto
se registraron numerosos replicas que fueron estlmadas por
algunos testigos presenclales. E&as replicas se prolongaron
por mas de un ano, con unos 60 slsmos importantes en 23 dias y
unos 300 en un ano. Los ruldos subterraneos continuaron por
mas de un mes, VicuAa Mackenna (1869) y Montesus de Ballore
(1916) .
Was Efw-+.ns: Se formaron grletas en el suelo de la regi6n
eplcentral, con emanaci6n de aguas turbias, barro diluido e
impregnado de gases mefiticos. De algunos montes se
desprendleron peflascos y se secaron manantiales en donde el
agua era abundante. El movimlento de tierra fu& ta1 que las
personas no podian sostenerse en pie. Barros Arana (I884),
Montesus de Ballore (1916>, Amunategui (1882) y Medlna (1928).
v: De acuerdo a 10s datos e informaciones
macrosismlcos fueron estlmados en 10s sigulentes valores.
1. Tlempo de Percepcibn Media de1 sismo principal
t* = 2.5 - 3.0(minutos)
2. Distancia de mAxIma perceptibilldad
L+ < 1700 C Km 3
3. Tiempo total de duraci6n de las replicas
tw >l afio
4. La zona epicentral que contiene un forma aproxlmada al
Largo de Falla L es equivalente 0 menor a una zona cuya
longitud de ruptura es de1 orden de 10s 500 <Km), en vlrtud
que algunas caracteristicas tales coma: comportamiento de1
terreno y frecuencia de Ins replicas fueron muy oimilares a
las de otro slsmo Contemporaneo cuya longitud de ruptura
esta determinado con mejor precisl6n.

Utfllzando las relaciones empiricas que se lndican en las


f lguras adjuntas y que coma ya hemos dicho se, refleren a
varies sismos Importantes ocurridos en Chile en el preaente
siglo, Ver Ramirez David (1988); las mejores estimaclones
instrumentales que proponemos para el terremoto hlst6rico de
1647 son las siguientes:
Fecha: lunes 13 de mayo de 1647
Iiora: entre las lo:30 y las lo:45 horas (PM)
Epicentre: Latitud aprox. 339(S), Longitud aprox. 72Q-7lQ(W)
Ms= 8.5 en la zona de saturaci6n
Mw = 8.4
L = 400 (Km)
MO = 228 x 10"' (C.G.S.1
10

BSBLIOGRAFIA:
- AMUNATEGUI MIGUEL LUIS, <1882), El terremoto de1 13 de mayo
de 1647; Rafael Jover (Editor), Santiago de Chile, NQ total
de paginas 616.

- BARROS ARANA, DIEGO (18841, Historia General de Chile;


Rafael Jover (Editor), Santiago de Chile, Coleccl6n
oompleta de 16 volirmenes.

- GAY CLAUDIO <1983), Catalogos de cartas manuscrltos y


documentos hist6rlcos. Editorial Nascimiento, Santiago de
Chile, Archive National, Biblioteca National Santiago de
Chile.

- GONGORA y MARHOLEJO (1862). Coleccl6n de Historiadores de


Chile y Documentos relativos a la Hlstorla National; Tomo
II, Imprenta de1 Ferrocarril, Santiago de Chile: NQ total
de p&glnas 345.

- KEDINA, JOSE TORIBIO, (1928); Manuscritos, Documentas


Inbditos para la Hlstoria de Chile (1501-1900). Imprenta
Universitaria. Santiago de Chile, Coleccibn de 4 tomos,
Sala :Medina, Biblloteca Naclonal Santiago de Chile.

- RONTEGSUS DE BALLORE FERNANDO <1911-1916) * Historia


Sismica de 10s Andes Xerldionales al Sur de1 Paralelo XVI:
6 Vmlbmenes. Sociedad Imprenta Lltografia Barcelona.
Santiago, Valparaiso.

- WORLA VICUXA CARLOS (1983>, Cat&logo de cartas manuscrltos


y Dacumentos Hlst6ricos. Editorial Nascimiento, Santiago
de Chile, Archive National, Blblioteca National, Santiago
de Chile.

- RAMIREZ LEON DAVID. (1988>, Estimaci6n de Algunos


Pa&metros Focales de Grandes Terremotos Hist6ricos
Chllenos. Hemoria para optar al grado de Hagister en
Ciencias. Facultad de Clenclas Fisicas y NatemBtlcas,
Universidad de Chile . NQ total de paginas 463.

- RIQUELME VENEGAS DANIEL, 'x1905). El Terremoto de1 Sefior de


Kayo, Imprenta Cervantes, Santiago de Chile, nQ total de
paginas 140.

- VICUESA MACKERNA BENJAMIN (18691, Historla de Santiago,


Imprenta de1 Mercurio, Valparaiso, NQ total de paginas 832.
._.. _-__*__.-
.I .l_,_
_..^_

11

.
.
.
/
./
.
W-

lq P-2151054 Y.
r.011
*. 14

,... M-,W-U.m,,.

Vl,_..

..I..l.. I.-. .1 1.1.41 . . I. .I.,..w,. a. m.q ..,,.,,, .,


*a*.
..l. , I. B.,..* . . ..-r. IL,. .Iww. .*-. er,.w.
P.*. ..,..,a. .Ivr. .I ,.,..II_ ..I .1- . . .C-..a- I. I...1 . . c

S.--I. . . ..I... .I ho-o.. , . ..C..d8.... ..,.I, n.. ..


1.. . . . . I.. I-l..a.. .m ,. ,I,... ._I,.

I.
. l

/ .

.
.
.
log I 0 -2.06 +0.53 M*
I- l 0.64
/ .
N-9

:
_ .._ __.~~__I._ I__.__,. .-__I

12

.
.
/ .
lop 1*042*0.80 108 I.

r 0.7)
N -12

VI,...

-t..t- .-a.. .8 I.,#.- h ,. .*.*.m,. _.I_

..1,.111,.. rn.8 , ., ,.,.,1. ..I I,._ . ...LI .,.I.

m.I..ln. .L .I I.,..,,_ ..I al..,. .-..,a- I. . . L


. . ..r-tn mati.-. L..lC .w,.,.. ., -.
..rr_l. ..-.I.,. .I L- . . . . , .I ,.,.-a... .. I. I.-q.,., . .

CI.r.4 .. .I,... I.._,.. c-l,.. .L._._...

I . _..__., _ __A

iL
. -

.C
..a..*.. . . ...*.. _.. ., -._ _ . .._.. -

C......ld ..a, , c- _ a. -...a . --.


. . . ..I ..__,_-_ a*.-. -.-a.
.I_. Id._ __. a. ,_. L... --..

..l..r~~.m.C.-.h.,.s. _..---.-.
e- d
-..

13

SEISMIC STUDIES OF DEEP SLAB MORPHOLOGY


BENEATH CENTRAL PERU

DAVID E. JAMES and J. ARTHUR SNOKE

;: iunegie Institution of Washington


Department of Terrestrial Magnetism
5241 Broad Branch Rd., N.W.
Washington, D.C. 20015
**
Dept. of Geological Sciences
Virginia Tech
Blackaburg, VA 24061

Resume

A lar e-amplitude secondary P-wave recorded on the broad-band seismograph at


Cueco, B eru, is interpreted to be an underside wide-angle reflection from the upper
surface of the atecply dippin Nasca plate. Possible reflection points are in the depth
range 150 to 400 km, where tf le slab is wholly aseismic, indicating that the descending
Nazca plate exiete in and is probably continous through that zone.

Key Words: Andes, aseismic slab, subduction zones, secondary seismic arrivals

Introduction

Two prominent features characterize subduction of the Nazca plate beneath western
South American: (1) alternating regions of normal and flat subduction; arid (2)
the absence of earthquakes everywhere between depths of 300 and 500 km. Central
Peru is characterized by both flat subduction and an abnormally large aseismic ga
(150-525 km). An important question is whether or not the slab is continuous throug
the aseismic gap. Despite the fact that high frequency seismic wavea from nearby
R
deep focus earthquakes are propagated through at least some aseismic region3 of the
Andes, thie alone is not sufficient to reyrcirc! slab continuity, and the issue remaius
unresolved see James and Snoke, 1990, for discussion).
Wortel (1 B$4) proposed for central Peruthat the transition from older subducting
lithosphere (age > 70 Ma) to youngsr and relatively hotter lithosphere gives rise to
a force system in which the deeper, more dense, slab detaches from the anon~dously
buoyant young Rlab. The buoyant younger slab forms a Aatn subduction limb, while
the more dense lower )ortion sinks into the deeper mantle. On the other band,
Schneider and Sacks (1 !!89) concluded that the descending Nazcn plate is continuous
through the aseismic zone and that the absence of earthquakcs can be explained M
due to a brittle-to-ductile transition at depth.
._l____-_.,_ I^ .,. ^.

14

Here we present direct seismic evidence for the existence and, inferentially, the
continuity of the Nazca plate through the saseismic zone beneath eastern Peru. This
result is baaed on the analysis of nn anorna~ous large-amplitude P-wave Hrrival in the
P codas of Peru-Brazil deep focus earthquakes recorded on the Carnegie broad-band
station at Cuzco, Peru (CUS) (see Figure 1). We shall show that the anomalous
arrival, which occurs approximately 1.5 second3 after direct. P and is the largest
arrival in the P-wevetrain, is best explained w an underside wide-angle reflection
from the u per surface of the slab somewhere in the depth range 130 to 400 km, a
region in w$1ch the slab is wholly aseismic.
_,

-78 -I@ -74 -22 -70

LONQITUOE( (de(l)

Fig. 1. Map of central Peru showin Benioff zone contours where defined by seismic-
ity (shallow contours not complete), B eru-Brazil deep focus earthquakes, and aseismic
zone within which slab reflections axe rrhown to occur.

Data

The primary data for this study are broad-band recordings at Cuscu, Peru, of
four deep focus Peru-Brazil earthquakes with epiccntral distances of c 4.0 from
the station. An exam le of velocity-corrected seismo rams and yurticle motion from
one of the Peru-Brazi P deeps is shown in Figure 2. A e anomalous arrival which we
andyze in this paper (PV) occurs approximately 1.5 aec after the direct P arrival Pd).
P, has a number of important clmracteristics, some of which can be secu clear \y in
Fi ure 2:
1. 6,t II the largest amplitude arrival in the P-wavetrain, with amplitude substantially
rcatcs that that of Pd (note radial component).
2. tst is a clearly defined phwe with sharp onset and rectilinear part& motion.
3. It is a P-wave.
4. It iS OppOSite in polarity b Pd.
15

5, It differs substantially in bachzimuth and emergence angle from Pd and hence ham
a significantly different propagation path.
P, is clearly observed on all records of deep focwe events from the Peru-Brazil region
at e icentral distances less than 4, and the waveform characterist.ics are strikingly
simr-Pof between events. Revond about 4 results are less consistent and beyond about
6. P, is not seen at all. .
b

Figure 2. Particle motion for Yecondary phase, Pr. (a): Radial to transverse
components. (b): Radial to vertical components. Vertical l&s on seismic traces
define time window over which particle motion ie determined. Directional conventions
F, B, L R, U, and D stand for front, back, left, right, up, and down, respectively,
for an observer facing toward the station from the epicenter.

Analysis

We anal ze the nature of the anomalous arrival, P,., chiefl by examining particle
motion. J easured particle motions for Pd (direct arrival) con x rnr that the Ilrst arrival
ropagates along the direct path as predicted b a radially symmetric earth model.
i5 n the other hand, particle motion analysis oYPI shows that the angle of arrival
at CUS is very much out of the plane of direct propagating phases. In particular,
the bachazimuth of P, is about 18 greater than that of the direct arrival Fig.
2)and the vertical-radial particle motion shown in Fig. 2 yields an apparent ang\ e of
emer encc of e 49. When the effect of interference between Pd and Pr is removed
by su%tracting the llrst arrival from the vertical and radial components (see James
and Snoke, 1990, for detail*), the apparent angle of emergence for PI is found to be
b 34 f 8.
The results of the article motion analysis aud forward modelling show that the
anomaIous P, arriv BPcannot he due to effects either of the crust beneath CUS or
source or near-source complexity. The most Iausible explanation for the phsase is
that it has pro agated along a relatively hig R Q path and has been reflected at a
major velocity ! iscontinuity with a ngcrtiua velocity contra& that produced n polarity
reversal.
_ ______- _..I .I - .-.

16

Corrclufiouo

We conclude that P, is B wide-angle reflection from the boundary between the


descending plate and the overlying mantle wedge. Thus, Pr is a P-wave that leaves
the source travelling u ward in the dab and is subsequently reflected at wide-angle
from the underside oft Re upper elab boundary. For a specified depth of the reflection
point, forward modelling vcs the latitude and longitude of the refection and the
attitude (strike and dip) oP the reflector at that point. hlodelling results show that
the reflection point must be at a depth between 150 km and 400 km and that the
slab must be very steeply dipping at the point of reflection. Inferred depths to the
rubducting slab are shown contoured in Fig. 3a, and families of calculated reflection
ointe that fit the data for a variety of slab velocity models are shown in Figure 3b.
5 he large amplitude of Pr seem to re uire e large velocity contrast between slab
and overlying mantIe of at ieaat 5 to 10%o.

smarw4 mtl

Figure 3. (a) Sch ematic map showing contours of descending Nazca plate in the
aseismic regions beneath central Peru aa inferred from this study. Intermediate seis-
mkity determined tram temporary central Peru ortable array shown for reference.
b) Modelled reflection points, with dip angle in $ rcated projected onto cross-section
A-A. Crosses are event locations shown in (a). Heavy, line denotes approximate
upper rtufacc of descending Nazca plate inferred from scismicity.

Referencelr
James, D. E., and J. A. Snoke, Seismic evidence for continuity of the deep slab
beneath central and erwtern Peru, .I. Oeophyu. Rw, 95, in press, 1990.
Schneider, ,I. F. and I. S. Sacks, Subduction of the Nazca plate beneath central
Peru from local earthquakes, Jour. Gcophy,. Re,., 95, in press, 1990,
Wortel, M. J. R., Spatial snd temporal variationa in the Andean subduction zone,
J, Geof. Sot. London, 141,783-791, 1984.

.
17

MICROSISMICITE DE LA SUBDUCTION AU PEROU CENTRAL.

AUTEURS: R. Lindo12, C. Dorbath213, A. Cisternas et L. Dorbath213

1 lnstituto Geofisico del Peru, Lima, Peru.


2 lnstitut de Physique du Globe de Strasbourg, France.
3 ORSTOM, France.

Resume

Une experience de surveillance sismique a et6 efectue dans la


region cdtiere du Perou, entre Lima et Pisco, et la Cordillere
Occidentale. II donne une perspective assez Claire de la croOte et
de la region de la subduction. Les resultats correspondent a 8
semaines de campagne. La zone de Wadati-Benioff ne montre pas de
discontinuites fortes. Dans la partie la plus proche de la fosse la
plaque subduct6 fait un angle denviron 17 puis elle devient
presque horizontale. Dans cette region nous trouvons des
mecanismes au foyer en faille inverse; plus loin il faut signaler
la presence de mecanismes de type normal et de decrochement. On
observe aussi de la sismicite crustale des deux c&es de la
Cordillere Occidentale.
19

DETERMINACION CON ALTA RESOLUCION DE LA GKOMERIA DE LA ZONA


WADATI-BENIOFF EN LA PARTE CENTRAL DEL PERU Y DETERMINACION
DE ESFIJERZOS

*: Direction de Servicios Tecnicos


Institute Geofisico de1 Peru
Apartado 3747 Lima 100, PERU.

Resumen

Siguiendo a Hasegawa y Sacks (1981), Boyd (1984))


Deza(1984), Schneider Sacks (1987), (1988); la geometria de
la zona Wadatti-Benioff en la parte central de1 Peru es
determinada con alta resolucibn usando 2 conjuntos de datos
sismicos. Ocho adios, (1980-1988), registrados por el eistema
de detection sismica en tiempo real (AUTOSEIS) conectado a
una red sismica telemetrica (I.G.P.) y cuatro meses de
informacibn sismica registrada en la campafia de Carnegie
Institution of Waehington-IGP en 1985. Varioe algoritmos de
localizacibn de terremotos han sido probados anteriormente
comparando 10s errores producidos por cada uno de ellos con
el fin de determinar cual de estos genera soluciones ma8
estables. Diferencias apreciables fueron encontradas entre
10s metodos considerados.para un mismo conjunto de datos.
Los event06 slsmicos han sido recalculados usando el metodo
de "Multiple Master Event". El conjunto de datos utilisados
incluye mas de 2000 eventos, de 10s cuales aproximadamente
200 han sido escogidos coma eventos maestros (sismoe con la8
mejores eoluciones). Los eventoe son recalculados utilizando
las correcciones por estacion derivados de 10s residuales de
10s sismoe, maestros. De acuerdo a 10s avancee logrados con
nuestros datos hasta la fecha (el Area de estudio comprende
aproximadamente 800 kma desde la fosa oceanica) se delineard
la zona Wadatti-Bennioff tomando una aproximacion
relativamente simple de la tendencia de la superficie
hypocentral, la que se construira deE;de una aproximacion
polinomica partiendo de las ubicaciones hypocentrales
generando una representaci6n tridimensional de la misma. Asi
miemo 6e analizara 10s mecanismos focales (simples y
. __lr___ -_.._ .__ . .._-a I --..

20

compuestos) a partir de la polaridad de las fases P para


eventos ocurridos en la region de estudio; con el proposito
da identificar 10s patronee dominantee de esfuerzos
relacionados a la geometria a determinar y asi poder conocer
la posible contoreiin de la zona de Wadatti-Benioff en eeta
regi6n en forma ma6 preciaa de lo que se ha hecho hasta la
actualidad.
21

ETUDE DE LA SUBDUCTION DE LA PLAQUE COCOS


AU NIVEAU DU NICARAGUA

Marco Antonio ARELLANO et Henri HAESSLER


Institut de Physique du Globe de Strasbourg
5, rue Rene Descartes, 67084 STRASBOURG CEDEX

La sismicite du Nicaragua est principalement due a la


subduction de la plaque Cocos sous la plaque Caraibes. On a
effect& une analyse de la sismicite dans le but de preciser
les caracteristiques de la subduction dans cette region.
Plusieurs annees de fonctionnement dun reseau
sismologique local ont permis lacquisition dun nombre
important de donnees. Un grand nombre de seismes ont
ainsi CtC localises avec precision de 1975 B 1981. Nous
avons relocalise les evenements de 1981 afin de verifier la
qualite des localisations effect&es en routine. Celles-ci
apparaissent trbs correctes et le catalogue de cette periode
(1975-1981) apparait comme la meilleure source de
donnCes.

Les caracteristiques mises en Cvidence par lanalyse


de la sismicite permettent de comparer la subduction au
Nicaragua B celles observees en dautres regions du globe.
Une analyse fine de la distribution spatiale des seismes
avec relocalisation des foyers met en evidence les traits
plus specifiques de la subduction au niveau du Nicaragua :
pendage tres Cleve et existence de plusieurs lacunes de
sismicite.
23

MODEL05 CRAVIHETRICOS EN EL SECTOR DEL VOLCAN VILLARRICA ASOCIADOS


A LA
MEGAESTRUCTURA LIQUIRE-OFQUI

ARANEDA, M., AVENDARO, M.S.

Departamento de Geologia y Geofisica


Universidad de Chile
Casilla 2777, Santiago-CHILE

Resumen

Se presentan resultados de un estudio gravimgtrico en el &ea de1 Volcen


Villarrica y sus alrededores. La modelacion de 10s perfiles de gravedad,
ubicados en la parte norte de la estructura mencionada estaria confirmando
la existencia de una anomalia negativa de1 orden de 8 miligales en una
extension proyectada de 3 Kmen el sentido W-E. TambiGn se observa, hacia
el Este de1 perfil 2 un bajo gravimetrico de menor magnitud que podria
asociarse a un fallamiento paralelo al anterior. Estos nuevos anteceden-
tes estarian confirmando la zona de debilitamiento LiquiRe-Ofqui. Si
proyect&amos las anomaliasencontradas hacia el Sur, podriamos sugerir que
el Vol&n Villarrica se encontraria en el zona de debilitamiento presenta-
da por una enomalia negativa.

Abstract

Results of a gravimetric study carried out in the Villarrica Volcano


region are presented. In the northern part of the volcano, on a E-W
profile, a low value of the Bouguer anomaly (8 mgl) extending over 3 Km
was found. Towards the East, another - smaller amplitude - low value was
located. These two anomalies are interpreted as two subparallel faults
evidencing the presence of a zone of weaknesscorresponding to the Liquifie-
Ofqui fault system. The southward projection of the anomalies coincides
with the Villarrica volcano, hence the volcano would lie on the zone of
weakness.
24

Introduccih

La Megafalla Liquiiie-Ofqui se encuentra ubicada en la parte occidental de


la Cordillera de Los Andes, aproximadamente entre la latitud 38.5O S,
sector Lago Caburga y 47.0 S, sector Golfo de Penas. Su longitud de mas
de 1.000 Km se presenta principalmente por una rama de orientation N IO0 E
con desplazamiento transcurrente dextral M. Her& (1976). El sistema de
falla LiquiRe-Ofqui se caracteriza por fallas lineares, fuentes termales,
afloramientos de rotas de falla , cones volcanicos, especialmente monoge-
n&icos y grandes depresiones slineadas a lo largo de la traza principal
F. Her& (1984). La estructura fue reconocida localmente-por Klon (1960)
y Saint Armand (1961) quienes la designaron acertadamente coma Zona de
debilidad estructural y Falla Reloncavi, respectivamente. Posterior-
mente, fue reconocida en el tramo norte por F. Her& (1974) y Moreno y
Parada (1974, 1976) denominandola Falla LiquiRe-Reloncavi. Finalmente, se
reconocid su presencia en Chiloe Continental,, extrapolandose hasta el
Istmo de Ofqui F. Her& et al. (1979), asignandoie el nombre de zona de
falla LiquiRe-Ofqui.

Algunos antecedentes geologicos de granitoides dispuestos en lado W y E


de1 Lago Caburga hacen suponer que la mayor estructura continuaria hacia
el norte de Liquiiie; de la misma forma se supone que un tercio de la parte
norte estaria constituida por una sola rama principal rectilinea y rumbo
IO0 E Thiele et al. (1986). El primer supuesto se puede sustentar por
10s modelos preliminares aportsdos por la interpretation gravimetrica, no
asi el segundo, ya que se puede observar en la Figura Ic la existencia de
un fallamiento paralelo al principal de magnitud ligeramente inferior. En
la parte central, a la latitud de Chaiten aparecen dos ramas principales
bien individualizadas, una occidental que pasa cerca de la costa y otra
que lo hate unos 20 Km al interior. La parte austrsl es la menos conocida
y parece estar representada por una serie de fallaa en echelon o en cola
de caballo con concavidad hacia el NW Thiele et al. (1986).

El preaente estudio tiene coma objetivo ,analizar la megaestructura en


sector Lago Caburga y parte norte de1 volcan Villarrica mediante observa-
ciones gravim&ricas. Estos antecedentes corresponden a una parte de un
estudio m&a general que se extiende entre 38OlO y 3Y030 de latitud sur.

La megaestructura Liquiile-Ofqui y el metodo de gravedsd

Debido a qua dicha estructura presenta una notable anomalia negativa en


la8 zona de Cayutue, Ralun y Pocoihuen-Cochamo Araneda en Thiele et al.
(1985)~ Araneda y Avendsfio (IYES), en un ancho promedio de 1.5 Km, determi-
nado m&s bien por la longitud de 10s perfiles que por el ancho de la
anomalia y con el conocimiento que se tiene sobre las formaciones compo-
nentes abundancia de rota fracturada y brechizada yacentes en la zona
deprimida, hacen que el metodo aplicado sea el m&s adecuado para pesquisar
la estructura a lo largo de su zona de falla.

La zona estudiada se encuentra al norte de1 volcan Villarrica, en el area


se han tornado m&a de 300 estaciones grsvimetricas, densificada en la zona
de inter&. Se utiliza un gravimetro La Coste Romberg modelo G-411 y para
la altura, al+.imetros cuyss bases fueron pilares de nivelacion de primer
orden de1 IGM.
.^ _ ._ . . _ .._

FIGURA 1a

mgal
mgal PERIlL I SECTGR CAGURGA m9al
PERFIL 2 SECTOR CASURGA VILLARRICt
197
AtiOMALIA DE GOUIilER
!9S

II

81

7s

77

75

73
RESIDUAL
- OBSERVAOD
---CALJLADO

RESIDUAL
8L - OBSERVADO
--CALCULADD

FIGURA 1b FIGURA lc
26

De 10s datos e interpretation

Los datos tuvieron las correcciones normales incluyendo la topografica.


Para el modelaje bidimensional se consideraron las siguientes densidades:
2.2 gr/cm3 para la rota de falla,, la cual fue considerada homogenea en
todo su volumen. Lo mismo ocurrio para las formaciones adyacentes de la
rota de falla = 2.7 gr/cmf esta rocas es&n compuestas por granitoides, a
ellas se les hicieron determinaciones de densidad, las que dieron un valor
medio dado anteriormente, Por lo tanto, el contrsste de densidad fue de
AY=D. 5 gr/cm3 . Este contraste fue el que se utilizo en el modelaje de la
estructura presentsdo en la Figura lb y Ic. La interpretation considero
perfiles amarrsdos en 10s intrusivos ubicados en las zonas mostradas en la
Figura la. En la Figura lb se puede observer una anomalia de -8 miliga-
les hacia el W juste en la parte sur de1 Lago Caburga que podria conside-
rarse coma la zona de debilitamiento LiquiRe-Ofqui, aunque la anomalia
podria interpretarse corn0 un gran espesor de sedimentos glaciares.
Pensamos que afinando el modelo con nuevos antecedentes e interpreta-
ci& con densidades variables podria sustentarse con m& profundidad la
correspondencia de la megaestructura. Al este de1 perfil, Figura lb (E.
Lago Caburga) se observa una anomalia menor que se puede atribuir a un
valle de menor magnitud que tendria la misma longitud transversal de la
parte sur de1 Lago Caburga.

En la Figura Ic se observsn dos minimos, el primer0 corresponderia a una


zona de fallamiento paralels,a la estructura ,principal que se ubicaria
hacie el E. La ubicacion de1 ultimo bajo gravimetrico atribuible a la zona
de falla se encuentra casi perfectamente alineado con la anomalia mayor
presentada en el perfil 1 . Su amplitud y forma tienen una gran simili-
tud, que en este analisis preliminar corresponderian a la falla Liquiiie-
Ofqui en el sector Caburga. Si el alineamiento descrito se prolonga hacia
el sur, pssaria por el vole& Villarrica hacienda suponer que esta
estructura estaria dentro de la zona de falls.

Agradecimientos

El presente Proyecto ha sido patrocinado parcialmente por FONDECYT,


mediante el Proyecto 274/07.

Bibligrafia

Klohn, C. 1960. Uns zona de inestabilidad estructural con fracturas


profundas en Los Andes de1 Sur de Chile, reactivadas
en el terremoto de 22 de mayo de 1960. Informe
inedito, Inst. Invest. Geol., Santiago.

Her&, M. 1976. Estudio Geologic0 de la Falla LiquiRe-Reloncavi en el


area de Liquifie; antecedentes de un movimiento
trsnscurrente . Actas ler Congr. Geol. Chileno,
Santiago, l(B): 39 - 56.

Araneda, M. y Avendaiio M., 1985. Estudio gravimetrico de la Falla


LiquiRe-Ofqui, en 1~ sectores Cerro Cayatue, Bahia
Ralk y Cochamo-Punta Pocoihuen. Actas 4O Congr.
Geol. Chilefio, Antofagasta, I(2): 1 - 16.
27

Herve, F., 1984. Rejuvenecimientode edades ra,diometricasen la zona de


falla Liquifie-Ofquien Aysen.Comunicaciones,34:107-
115.

Saint Amand, P. 1961. Los terremotos de Mayo-Chile, 1960. Tech. Article


14 Michelsen Lab. U.S. Naval Ordenance Test Station,
China Lake, California.

Her& F., Fuenzalida, J., Araya: F. y Solano A.,1979. Edades radiometri-


cas y tectonicas neogenas en el sector costero de
Chiloe Continental, X Region. Actas 2O Congr. Geol.
Chileno, Arica I(F):l-18.

Moreno, H. y Parada,M, 1974. Geologia de1 area de Liquihe, Neltume y Lago


Pirihuico, Proy. Hidroelec. Neltume, Informe inedito,
Inst. de Invest. Geol., Santiago.

Moreno, H. y Parada, M., 1976. Esquema geologico de la Cordillera de Los


Andes entre 10s paralelos 39OO' y 4130' S. Actas
ler Congr. Geol. Chileno, Santiago I(A): 213-226.

Thiela, R., Her&, F., Parada, M.A. y Godoy, E., 1986. La Megafalla
LiquiRe-Ofquien el fiord0 Reloncavi (41030'), Chile.
Comunicaciones37; 31-47.
29

GRAVITY, ISOSTASY AND ANDEAH (XlX!AL BEXWEDI 30% AND 35%


SHOKC'ENING

Introcaso Antonio (+I - PacinoMarti Cristina(+)- FragaH&or (+)

(+)~acultadde cs. !&a&as e IngenierTa(U.N.R.) y aXICET


Avda. PellegriniNo 250 - 2000 lXXARI0(Sta.Fe) - APGEmlXA

Cuatm seccionesg-ravim&ricas AndinasEW ubicadasen 30S,32's,33% y


35% revelan:(i)Axptable equilibrioisostitim (Airy);(ii)Que las
prdxblesreparticionesdemasasheterog&ms subcorticales mpremtan
resultadosgravi~&tricosobservables;(iii)-10s de acortamiento
pue-
den justificaxtotalmntelafonnaci6ndelos Andesyde las SierrasPam
PeaMs.
Keywords: gravity, isostasy,Andesand ,aS. Pampeams building.
Introduction
Rasedon12 gravimetric sectionsin east-westdirection properly
distributedupontheSouth&ne.ricanaustralsectorbetween 20S and 44%
latitudes, and mostlyextendedfran the PacificOceanto the Atlantic
oCean,gravitycharts ofFreeAiranamlies,&uguerancmalies and
isostaticanonaliesfor Argentina-Chilewereprepared.
Usinggravimtric inversionofthosesections,crustalmdelswere
prepared(In tmcaso-Pacino,in prep.),and a chart of &%ohofor the Andes
ofArgentina-Chilewasdonebasedonthosen&els.Achartofisostatic
geoidfor the Audesaxnpleteour study.Addingthe followingpublished
charts:(i)the satellital geoid; (ii)the 3 Km contourAndeanelevatiw
and (iii)the contoursof depthto the centralpart of the Wadati-Benioff
seismiczone,we have done a ccmparative analysiswhichwe will later
canment.Fran this general work, fourg.ravimtricAndmnsectionslocated
at 30'S,32'5,33% and 35% latitudes,allowedus to anal&e with n-ore
details the Andeanstructure's behaviourin this transitional zone,yet
studiedinadeeperwayusing seismological data,by the InstituteSism-
ldgicoZorxda(I.S.Z.), the InstitutoNationalde Prevmci6n Slsnica
(INPRES)and by sm researchers fran Cornell'sUniversity, for example,
betweenothers,Smalleyand Isacks (1987).
Ihe analizedsegmentgoes acrossthe Cordillera de la Uxta, the Valle
centrdland the Cordillera de 10s Andes in Chile,and the CordilleraPr+
cipalandtheLlanura Chaco-Pampeana inArgentina.Thethreenorthernmst
sectionsalso go acrossthe WesternSierrasPaspeanas.
Theseranges,which are similarin many ways to the U.S.A.Laxamide
uplifts,show activecanpression at present,pinted out by many focal
30

mechanisms, by,eitheraltitudeand gravitytemporarychanges,and by the


recognition of reversefaultingfields.Studiesof deep reflectionalso
mark the listriccharacterof the crustalfaulting.Curiously, we have
foundsignificative positivegravityvaluesover theseranges.!Iheanalized
gravimetric sectionsfrcm N to S present:extensionsof abut 3200 km,
2000 km, 1900 km, 1450 km; maximundephts in the trenchof 6400m, 6200m,
5900m and 5475m; maximunmean altitudes(andfrommeasurement) over the
Andeanbuildiq of 5800 (4790)m,4200 (2150)m,4200 (3265)mand 3500
(2280)mrespectively: maximunFree Air alym.of -175n-Gal, -180 mGa1,-193
nGal and -130mGa1 over the trench,and maximunRougueranomaliesof -340
mGa1, -320mGa1, -300mGa1 and -225 mGa1 over the Andeanbelt. (Intrccaso,
1980;Pacinoe Introcaso, 1988;Fraga e Introcaso,1988).
Results.Brief discussion
A relationship betweenBougueranomaliesand altitudesin 1" x lo obtained
for Argentinaand the Andeanbelt,modifiesthe relationship defined20
years ago with lessvalues,and approximates the relationship in lo x lo
obtainedfor U.S.A.,being consideredas in isostaticeguilibrim.
There also exista rgnarkable correlationbetweeneach of the following
charts:topographic one, of Bouguer,of iscstaticcorrection, of the Moho
and the Geoid (whichpresets, becauseof its esence,a widthbiggerthan
the others). This correlation clearlyindicatesthat,at leastin a great
part, thereexist isostaticcrustalcompensation in the Airy Systen.
Apparently, a correlation betweenthe slopechartof the SubductedNazca
Plateand the precedingcharts,does not exist.Below 33O.S latitude,and
as we had alreadypointedout for the 24% parallel(Introcaso y Pacino,
1988),the probablegravimetric effects(pceitive one for the subducted
plateand negativeone for the wedgeof hot astenospheric materials,
sourceof Quaternaryvolcanisn),if they exist,couldbe mostlycancelled.
Above 33OS latitudein the flattensubductionzone,we should'oonsider
almostonly the probablepositiveeffectof the plate lookingfor its
canpensation beneaththe OceanicPlate,since it is not reflectedin the
results,or we shouldthinkthat the effectof the plate is not
significative.
The most cbviousconclusionis that the Bouguerananaly,both in the
studiedsectorand in the wholeArgentina-Chile, is mainlycontrolled by
Mohorovicic discontinuity.
The fourgravimetric sectionswere regionalizated by the upJards-co_?
tinuation method (Pacinoe Introcaso,1987).Studiesof delay time
amnalies in four east-weststationsin WesternArgentina,and of deep
seismicity(crustand uppermantle in the Chileansector,and intermedim
crust in Argentina)cunpletethe preparation of our models.The inversions
allowedus to defineantircotsbeyondthe Chile Trenchand the bottomof
the crustwith the followingmaximundepths:71 km at 30%, 65.5 at 32'+S,
65 at 33% and 57 at 35%. The crustalrootswere involvedin the Andean
shorteniqmodels,as we will latersee. Recently,shortenings of 40-45km
for the CordilleraPrincipal(Ranos,V. personalcanunic.)and 95 km for
the Precordillera (StateOil ArgentineCompany)were recognized.
With thismodifications, the whole crustalshorteningfor the Argentine
sector,includingthe Pampeansegmentwould be of 150 to 160 km. Taking
intoaccountthe Chileansector,we ought to admita probablecrustal
shorteniq of about180-190km for the totalextensionof the profile.For
the SierrasPampeanas, we have recentlyanalizeda compressional rotation
liftiq mechanism,which explainsonly in part the high gravitythatwas
Air An. fvBougusAn.
(Cl

GC= 2.9 gr/cc


c
I

FIG.l.Gravityresultsan one of the


four studiedsections:30%
(A) Itinerary: CC (CbIdilleXadela
Costa);Cp (QrdilleraPrincipal);
CF (OxdilleraFmntal);P @recordi-
llera);BB (Bermejo Basin);SVF (Sie-
ma de Valle F&till; SS (Sierrade
SaEogasta).
(B)Free AirandBouguer anaxalies,
togmgza~yandOceanicB&kxn.
(C)CrustalgravityMel bwx. depth
70 km), 51 (topographicmass)
h bxxklMss).Sho&eningofabout
200 km. C: Crust;OL: OceanicLithos-
phere;CL: CmtinentalLithosphere;
IM: UthosphericMantle.
(D)Studiedsqnent's location.CX ti
right:loo-125Km depthWadati-
Benioffzonecontours(fran
Emauw-Isacks, 1987).
32

obmwed. Iheotherpartcouldbedue todenseanmalousmsses, located


atmdiumanduppercrust.
WiththecrustalgravimetricRodels,~~ve evaluatedthecrustalroot
areaqwhich- Eddedtothe excedentmassesrespectively- allow&us to
find ratesof shortming of about 200 km, 180 Ianand 90 Ianfor the 3os,
32'S (and33'~)and 35"s sections,consistentwith the last studies,and
to explainthe whole Cenozoicbuildingof the Andeancordillera.
Magmaticintrusicncouldbeapartial factorfor the Ardeanbuilding
elevati.on.Ncnrrericalcdlculusrevelavery littlecontributiondur'
Cenozoic.In fact,a rmsonable msgmaticintrusionof 900 to 1500' SW,
Ian
would maJcethe crust (whichis 200 km wide) thickerin 4,5- 7,s lcn,
justifying only a 10% to 20% of the actualAndeanaltitudes.

Conclusions
Classicalcrustalmdelsshow:antixcotstithe ChileTrenchsector,
sighificative crustalthickening with rootsof 38 km (in30'S),32 km (in
32'S and 33"s)and 24 km (in 35'S)amsistentwith the topographic
excesses(reasoMbleisostatic equilibrium in theAi.rySystem).The
addition of all the assumedsubcrustal effectsis not significatively
reflectedintheobservedgravimetrical results.
Shortehingmdelsexplain:(i)theWeste.rhSierrasPampeanas building,by
rotationand liftingthroughlistricfaults.Thismechanisnjustifies
only a littlepart of the high observedpositivegravityamnalies. the
rest can be explainedby significative massesof high density,settedat
the intermedimtouppercrust; (ii)the studied Andean segment elevation
(that involvescrustalrootsgravimetricallydefine.d), withno needof
consideringothermechanisns as, for instance,mgmaticintrusion.

acis paparwas supportedby Facultadde CienciasExactase IngeaierL (u-


n.iversidadNacionalde IQsario);I.P.G.K.(econdmic
help to Gravity
Studies)and ConsejoNationalde Investigaciones Cientificasy !&r&as
de Argentina(Grant:PID 30-9330088).

References
Fraga,H.-Introcaso
A. 1988.Modelogravin&ricocorticalen la latitud
35?s.~ctasv Qxqr. Geolog.chileno.'Rx-o11:107-121.
InixocasoA., 1980.Pesultados
gravin&ricosen la bmda latitudinal de
Argentinacentraly pafsesvecinos.Rev. GeoflsicaI.P.G.H.,12:5-25.
IntrocasoA.-PacinoM.C., 1988.GravityAnd= tie1 associated with
subductionnear 24O 25'S latitude.Rev. de Geoflsica22:29-44.
PacinoM.C.-IntrccasoA., 1987.Regionalancnalydeterminationusingthe
upmrds continuationmethod.Bollettinodi Geofisicate6ricaed applicatq
XXIX (14):113-122.
Pacino M.C.-Dkccaso A. 1988.&de10 gravim&ricosabreel sistenade
subducci6nPlacadeNazca-Placa Sudamericana
en la latitud33'5.ActasV
Ccngr.GeolijgicoChileno,!Rxm 11:77-89.
SmalleyR.-IsacksB., 1987.A high resolutionlocalnetworkof the Nazca
PlateWadati- BenioffZonetier westerhArgentina.J.G.R.,92 (B13):
13903-13912.
. _ ..l..___. ._ .._

33

POSTSEISHIC LC\ND tlOVEtlENT5 IN SOUTH-CENTRAL CHILE

Serqio E. Barrientos

Dept. of Geology and Geophysics


University of Chile
Casilla 2777
Santiago, Chile
BITNET address: SBfiRRIENWCHCECVM

Resume

Tide gage records at Puerto Montt, referenced to Talcahuano, indicate a


large (1 m) postseismic uplift of the region. Field observations in 1989
carried out at the same locations of previous measurements in 1968
CPlafker, 19701 are consistent with trde gage records. The postseismic
elevation changes are modeled as the product of a propagating creep on
the down dip extension of the fault. For a Xi0 E fault dip, minimum root
mean square error indicate a creep velocity of 3-5 km/yr and slip
amplitude of 4-6 m.

Resunen

Reqlstros dp mareoqrafos WI Puerto Montt, con rpspecto a 10s rpqlstros en


Ta 1cahuano , 1ntllcen II n Ievantamlpnto de al menu5 1 m PII I,$ rry161-1.
Observ.lciones dir-ectas de levantamlento de la costa efectuadas en 19BV en
las mlsmas localidades que fueron medidas en 1968 CPlafker, 19791 son
consistentes con 10s registros de mareas. Los cambios de elevacldn son
modelados coma product0 de un desllramiento asismico en la continuacibn
en profundidad de la falla que acomod6 movimiento cosismico. Para una
falla de manteo de 35OE, la suma de errores al cuadrado es minima cuando
la velocidad y amplitud de deslizamiento varian entre 3-5 km/aZo y 4-6 m
respectivamente.

Introduction

The great Chilean earthquake sequence of May 21-22, 1960, with a moment
magnitude of 9.5, has been the largest event recorded in this century
CKanamori, 1977; Cifuenter, 19893. Remarkable changes in land levels were
observed in a region lOOO-km long by 200-km wide stretching southward
from Arauco Peninsula. Extreme co-seismic sea level changes ranged from
5.7 m of uplift in Guamblin Island to 2.7 m of subsidence in the city of
Valdivia. These values are based on measurements of sea level change made
by G. Plafker who visited the site in 1968, eight years after the
earthquake. The sea level observations are complemented by geodetic
34

measurements along two survey lines carried out by the Institute


Geografico Militar CIGM, 19653: A 150-km leveling line from the coastal
city of ConcepciIn to Parral and a 600-km line extending southward to
Puerto Montt.

Plafker and Savage Cl9703 analyzed the static deformation data and
presented teleseismic surface wave evidence to support their preferred
uniform slip dislocation model that involves between 20 and 40 m of dip
slip on a fault 1000 km long and at least 60 km wide. Plafker Cl9723 re-
analyzed the static deformation and deduced a model involving a fault 120
km wide by 1000 km long dipping 20@ E with 20 m of slip. Barrientos and
Ward Cl9901, inverting the elevation change data for a variable slip
model, inferred that most of the slip is concentrated on a 900 km long by
150 km wide band parallel to the coast. Sever a 1 patches of moment,
isolated from the main body of moment release, are found at GO to 110 km
depth, presumably indicating aseismic slip. Due to the size of the
coseismic displacement, large postseismic readjustments are expected.
This paper investigates the evidence of postseismic movements and
establishes the bounds on creep velocity as indicated by the records of
the Puerto Montt and Talcahuano tide gages.

Data

The Puerto Montt and Talcahuano tide gage records form the basis of our
analysis. The Talcahuano tide gage -located just outside of the 1960
deformation zone- has recorded the sea level almost continuously since
1949. The longest period lackinq information extends for approximately a
year -from September, 19% until October 1954. Seven interruptions of two
months or less also conspire aqainst the continuity of the data. Hnnt!lly
mean values of the sea level are obtained by averaginq hourly readinqr.
Sea level remains stable for most part of the total interval slightly
increasing after 1976. Most of the signal shows a uniform standard
-deviation (s.d.) of approximately 37 cm, some anomalously low s. d.
values are found during periods of incomplete record.

The Puerto Montt tide gage record is less complete. This record can be
divided in four periods: i) from Feb. 1945 to Dec. 1952, ii) from Nov.
1957 to March 1960, iii) from June 1964 to Nov. 1972, and iv) from Apr.
,198O to Dec. 1984. 011 of these periods are referred to a different
reference system except for periods ii) and iii) which include year J960.
The standard deviation of this signal is about 150 cm, about 3.5 times
the s.d. of the Talcahuano tide gage, reflecting the different
environmental conditions of marine currents.

The trends along all periods in the Puerto tlontt tida gage show different
tendencies. Prior to 1960, both periods of observation show a stable sea
level. This situation changed sometime between 1960 and 1964 evidenced by
the continuous decrease of the sea level between 1964 and 1972. The sea
level decrease is not apparent during the interval 1980-1984.

To eliminate the eustatic sea level changes -and also high frequency
signals exhibited in the records- it is desirable to analyze the
difference of the signals, i.e. Puerto Montt minus Talcahuano amplitudes.
35

This procedure was


suggested by wyss 15o
tr9763 when analyrinq PUERTO MONTT
the possible variations s
of sea level as 0100 ,-
/.-- .- P+
premonitory phenomena - ...-
of large earthquakes. t
This procedure gives a 2 x) I@
better estimation of 3 /
/-
the coastal uplift rate
at Puerto Montt. Figure a
+ o .-~$_.......~
1 shows the difference 2
of the signals recorded 8
at both tide gages. The -50 1. . . , . . .. . . . ,. . .. ... . . ..(
1045 1955 1055 l&S ~085
dashed line joins the
TIME (YR)
different observation
samples and it is Fip. 1. Elevation change at the Puerto Hontt tide page as a
interpretative for the function of tire. Thedashedline, which is the interpretation of
1953-1950 and 1973-1980 coastal uplift, indicates a postseisric change of lore than I 1.
periods. A straight
horizontal line was
assumed for the pre-1960 epoch. We have assumed that for the period 1960-
1964 no elevation change took place during the co-seismic stage, hence
all movement is postseismic (another possibility is that 18 cm of uplift
took place co-seismically and then a continuous uplift rate -4.5 cm/yr-
coincident with observations for the 1964-1972 interval). Lastly, the
uplift rate between 1973 and 1980 is shown as the continuous second-
degree curve that is coincident with the slopes at the two ends (4.5
cm/yr and 2.2 cm/yr) where data is available.

Discussion

The post-seismic movements can be interpreted in terms of two qenerally


accepted models. Visco-elastic readjustments of astenospheric material
and retarded slip along the fault. In subduction zones, visco-elastic
models have been proposed by Nur and Mavko 119743, Thatcher and Rundle
119793 to explain post-seismic deformation landward of oceanic trenches
mainly associated with Japanese earthquakes. Fitch and Scholz Cl9713
modeled the post-seismic deformation following the 1946 Nankaido (Japan)
earthquake with additional forward slip on the down-dip extension of the
fault plane. Kasahara Cl9753 has discussed the possibility of aseismic
faulting following the 1973 Nemuro-Oki earthquake (Japan) based upon
level ing 1 ines and tide gage records. Along the ideas stated by Kasahara
Cl9753 is the possiblity of retarded faulting due to aseismic propSgation
of the fault.

The coastal uplift at Puerto Montt has been interpreted as the product of
creep propagating along the down-dip extension of the fault which
ruptured during the 1960 sequence, although viscoelastic rebound cannot
be ruled out. Since Puerto Montt lies in the center of the lOOO-km long
rupture region, edge effects can be neglected allowing for a two-
dimensional fault representation. The propagating pulse is modeled as
several line sources activated according to a prescribed velocity. The
36

fault model inputs are: dip angle, velocity, and starting point of the
propagating pulse. For a given combination of. these parameters, slip
amplitude is inverted through a linear regression. The data used in the
inversion correspond to the dashed line in Figure 1 (with no co-seismic
movement) sampled every 0.5 yr. Fixing the dip angle at 35OE CKadinsky-
Cade, 19853 minimum root mean square error (RMSE) produce velocities and
slip amplitudes varying between 3-5 km/yr and 4-6 m respectively.

These observations were corroborated during the austral summer of 1989,


when re-determinations of sea level at several locations in the Puerto
tlontt area were consistently lower (up to 50 cm) than those observed by
Plafker in 1968 at the same locations. Values of this amount are
predicted by the model.

References

Barrientos, S. E. and S. N. Ward, The 1960 Chile earthquake:


Inversion for slip distribution from surface deformation,
submitted for publication in mphvs. J. 1%
Cifuentes, I., 1989. The 1960 Chilean earthquakes, J. Geophvs.
Res., 94, 665-b00.
Fitch, T. and C. Scholr, 1971. Mechanism of underthrusting in
southwest Japan: a model of converqent plate interactions, JA
Geophys. Res.. 80, 1444-1447.
IGM, 1965. Anuario N09, Instituto Geografico Militar, 1960-1965.
Santiago, Chile.
Kadinsky-Cade, K. A., 1985. Seismotectonics of the Chile marqin
and the 1977 Caucete earthquake of western Arqentina, Ph. 0.
Thesis, Cornell University.
Kanamori, H., 1977. The energy release in great earthquakes, J,
Geoohys.~Res. 82, 2981-2487.
Kasahara. K., 1975. Aseismic faultinq following the 1973 Nemuro-
Oki iarthquake, Hokkaido, Japan ia possibiiity), Peaeoph.!&
127-139.
Nur. fl. and G. Mavko. 1974. Post-seismic visco-elastic rebound,
Science, 183, 264-206.
Plafker. G. and J. C. Savage, 1970. Mechanism of the Chilean
eaithquakes of May 21-and 22, 1960. Geol. Sot. Am. Bull..Gl,
1001-1030.
Plafker, G., 1972. Alaskan earthquake of 1964 and Chilean
earthquake of 1960: Implications for arc tectonics. JL
Geophvs. Res. 77, 901-925.
Thatcher, W. and J. Rundle. 1979. A model for the earthquake
cycle in the underthrust zone, J. Geoohvs. Res. 84, 5540-
5556.
wyss, M., 1976. Local changes of sea level before large
earthquakes in South America. Bull. Seis. Sot. Clm.. 66, 903-
914.
37

MAGNETOTELLURIC INVESTIGATIONS IN THE NORTHERN ALTIPIANO OF BOLIVIA

RITZ, M:, HERAIL, G., BONDOUX, F., SEMPERE, T.

* OFSTOM,Dakar, Senegal
** OBSTOM, La Paz, Bolivia

*** ORSTOM, Port Vila, Vanuatu

**** ORSTOM, Santa Cruz, Bolivia

The Altiplano is characteristicof the structural and orographic


device of the central Andes (1, 2). It is a high plateau having
elevations of 3500-4000 m over an area about 200 km wide and 1500 km
long. Although the plateau seems clearly associatedwith the subduction
of the Nazca plate beneath western South America, its structure remains
ill-knownbecause of the lack of geophysicaldata. During 1989 August and
September,magnetotelluric(MT) measurementsat periods from 10 to 5000 s
were made at nine locations in the Bolivian Altiplano (Fig. 1) along a
200 km SW-NE profile, approximatelyperpendicularto the dominant surface
structures. A four channel digital MT system was used in this survey and
the horizontal components of the electric and magnetic fields. were
measured at each site. The MT traverse was designed primarily to provide
detailed informationconcerning the structure, thickness and extent of
the sedimentaryprovinces and examine the use of the method in a region
in which not much is known. The field tapes were analyzed following a
procedure described by Vozoff (3.1. Tensor apparent resistivities and
phases were rotated to their principal directions corresponding to the
geoelectricalstrike and perpendicularto it.
For most sites rotation angles were well defined and the average
electrical strike obtained from the impedance principal directions is N
140%, roughly parallel to the dominant trends of the study area. The MT
responses appear to fall into three distinct groups which are consistent
with separating the sites according to locality, i.e., the western group
with sites 1, 2, 3, and 4, the central group with sites 5, 6, and 7, and
the eastern group with sites 8 and 9. Typical sounding curves are shown
in Figure 2, which gives the MT responses in the parallel-to-strike(TE
mode) and perpendicular-to-strike(TM mode) directions,observed at sites
2, 7, and 9. All of the curves from these sites reveal some two- and
three-dimensional characteristics. The data for site 2 show a very
pronounced parallel split between the TM and TE amplitude curves, which
is caused by small, local, near-surface inhomogeneities (4l.This
- .,--...-.~--_ ,._ _, _.__ __1--_

38

, 20 km A.0
0

Figure 1. MT site locations superimposedon geology of the northern


Altiplano in Bolivia. 1: Paleozoic: 2: Mio-Pliocene sediments; 3: Plio-
Quaternary volcanic overburden; 6: Quaternary sediments; 5: major
volcanoes: 6: location of MT sites. A: Intra-AndeanBoundary Fault, B:
San Andres Fault; C: Sivinca Fault: D: Coniri thrust.

Figure 2. MT responses for three typical sites with error bars. Tensor
apparent resistivity and phase plots for the TE and TM directions are
shown by the dots and crosses, respectively.The solid curves through the
observed MT data are theoreticalcurves computed from the 2-D model given
in Figure 3.
39

distortion is seen only on the resistivity data, and the phase data are
unaffected as indicated in Figure 2. The MT responses for sites 7 and 9
appear to be little affected by the static shift effects of near-surface
inhomogeneitiesand the geological structuresbeneath the sites along the
eastern half of the profile (sites 5-91 can be described as two-
dimensional (2-D).
Near-surface distortionsare present at the four western sites and to
determine the major structuresresponsible for our observations, we used
the phase of the determinantof the MT impedance tensor (51 which is
unaffected by static shift, and truly represents the phase of the
impedance tensor itself. However, an interpretationbased completelyon
phase gives no informationregarding absolute resistivities. Thus, we
used the phase in a qualitativesense only for most of our sites, and
undertook a 1-D interpretation(6) based on the determinantparametersof
the impedance tensor for some sites consideredto be free of static shift
effects. The layered electricalmodels generated in this way provide a
starting point for 2-D modelling (7) across the whole traverse.
Results of 2-D modelling are shown in Figure 3. This model was
iterated until a reasonable fit to phase data for westernmost sites was
achieved. The geoelectricalmodel shows the presence of thick conductive
layers with resistivitiesin the range of 1 to 8 ohm-m except at sites 8
and 9 on the east end of the traverse where the resistivity is 30 ohm-m.
The base of the sequence of low-resistivitysediments lies at a maximum
depth of 5 km and contains depressionsand uplifts to within 200-700 m of
the surface. Below the sedimentary units, the crustal units are
characterized by resistivitiesless than 200 ohm-m. On the northeast
portion of the traverse, MT data give evidence for NW-SE faults which can
be associatedwith known structural features namely, Coniri thrust (near
sites 8 and 9) and San Andres fault (west of site 6). A significant
lateral discontinuity can be detected again west of site 2. This
discontinuityis an important element to locate the Intra-AndeanBoundary
Fault (FLIA), which has been individualisedby tectonostratigraphic data
(8, 9). At depths in excess of 13 Ionthere is a general trend toward
lower resistivities, the transition from 200 ohm-m to less than 8 ohm-m
occuring in the depth range 63 to 52 km. Schwarz et al. (101 indicateda
similar general decrease in resistivityat about 10 km beneath the
Altiplano in southern Bolivia. Finally, the geoelectricalmodel suggests
that three units are present below the Altiplano: very low resistivity
units in the upper 5 km, high resistivityunits in the middle crust, and
low resistivityunits in the lower crust or upper mantle.

References

(11 A. Lavenu, Thesis, ORSTOM, TDM 28, 820 p., 1986.


(2) B.L. Isacks, J. Geophs. Res., 93, 3211-3231, 1988.
(3) K. Vozoff, Geophysics,37, 98-161, 1972.
(C) M.N. Berdichevsky, V.I.Dimitriev, h Adam, A., ed., Geoelectric
geothermal studies, AkademiaiKiado, p. 165-221.
(51 R.P. Ranganayaki,Geophysics,89, 1730-1788, 1988.
-,
I-~_ __-_,

40

(6) D.L.B Jupp, K. Vozoff, Geophys. J. R. Astron. Sot., 62, 957-976,


1975.
(7) P.E. Wannamaker, J.A. Stodt, L. Rijo, Geophys. J.R. Astron. Sot., 88,
277-296,1907.
(8) Th. Sempirb, G. Herail, J. Oller, V Cong. Geol. Chili, Tl, A, 127-
162, 1988.
(9) Th. Sempere, G. H&ail, J. Oiler, P. Baby, L. Barrios, R. Marocco,
Int. Symp. Andean geodynamic,this issue, 1990.
(10) G. Schwarz, V. Haak, E. Martinez, J. Bannister, J. Geophy., 55, 169-
178, 1988.

SW 4 5 6 7 6 9 NE
sitrr
1 * 30

6 200

200
--I

Figure 3. Two-dimensional resistivitymodel computed to model the MT


traverse along the Bolivian Altiplano. Numbers indicate resistivitiesin
ohm meters.
41

CRUSTAL ELECTRICAL RESISTIVITY STRUCTURE IN THE SOUTHERN


CENTRAL ANDES

Gerhard Schwarz+, Guillermo Chong D.*,Detlef Kruger+, Eloy


MartinezO, Winfrid Massow+, Volker Rath+, Jose Viramonte-

+)Institut fiirGeophysik, Freie Universitat Berlin,


Rheinbabenallee 49, D 1000 Berlin 33
*)Degfo:fGeologia, Universidad de1 Norte, Antofagasta, Chile
Ge encia de Exploration, YPFB, Santa Cruz, Bolivia
Depto. Geologia, UNSa, 4400 Salta, Argentina

Since 1982 we have done long period magnetotelluric (T: 40 -


20,000 s) and geomagntetic deep soundings at more than 120
sites in northern Chile, southern Bolivia and northwestern
Argentina (Fig. 1). Two profiles of more than 700 km length
each between 21' and 25' southern latitude are crossing all
strut-tural units from the Pacific coast in the West towards
the Andean foreland in the East. Schwarz et al. (1984, 1986)
described measuring procedures as well as data analysis.

The main structural features of the Andean orogene could be


seperated by their electrical resistivity structure quite
well. The active foreland is characterised by a very low re-
sistive cover, although three-dimensional in its structure,
overlying a moderately resistive crust and upper mantle. In
the Subandean Ranges we observed a very similar situation,
but crustal resistivities never exeed 100 ohm m.
The Eastern Cordillera has high resistivities in its eastern
part, correlating with the uprising Precambrian basement. In
the western part of the Eastern Cordillera, west of Tarija in
Bolivia we measured extreme low values of only some ohm m at
shallow depth. Further towards south, in Argentina, the sit-
uation is not that clear: West of S. Antonio de 10s Cobres
low resistive structures were found again at crustal depth.
Their connection with the structures observed further north
is not yet clear, due to a gap in measuring sites at that
latitude. But we may assume that the very low resistive zone
is related to a thick sedimentary cover and young magmatic
processes with subsequent hydrothermal circulation in the up-
per crust. These processes formed the so-called tin belt,
which stretches from southern Peru towards northwestern
Argentina.
42

The Bolivian Altiplano and the Argentine Puna can be


seperated very well by their different crustal resistivity
structure. The Altiplano - two-dimensional in its structure -
has a well conducting cover, a moderately resistive middle
crust, and at a depth below 40 km a very low resistive zone
(I - 2 ohm m), of which the lower bound is unknown. The crust
of the Puna is moderately conductive too, but the low
resitivity layer found in the North is missing - or found at
much greater depth, e.g. at about 80 km.
The most striking feature on the traverse are very low resis-
tive upper crustal structures in the Western Cordillera of
northern Chile - striking approximately SSE - NNW and bending

Fig. 1: Geomagnetic deep sounding data from the Southern


Central Andes together with distribution of volcanoes:
Induction arrows (period T: 1000 s) as the reversed path
finders for zones of high electrical conductivity in the
crust, pointing away from anomalous zones. A zone of high
electrical conductivity (conductance of 20,000 - 30,000
S) beneath the Western Cordillera (the volcanic belt) of
northern Chile and southwestern Bolivia is striking under
NNW-SSE, running into northwestern Argentina. The eastern
border of this large scale anomaly is found in the
Eastern Cordillera of southern Bolivia and northwestern
Argentina. High electrical conductivity should be related
here to thick sedimantary layers and hydrothermal CirCU-
lation in the upper crust.
43

further towards W at about 22' southern latitude. Total con-


ductance (the thickness-resistivity ratio) is reaching values
between 20,000 to 30,000 Siemens, pointing towards the ex-
treme physical state of the crust. These values may originate
from partially melted acidic rocks - stuck in the crust, as
well as from uprising fluids penetrating the continental
crust due to dehydration processes at depth of the descending
Nazca plate. The very low resistive zone of at least 20 km
thickness correlates with a zone of reduced seismic veloci-
ties (Wigger 1988), attenuation and/or scattering of seismic
waves (Chinn et al. 1980), as well as with an extended low in
residual gravity (Gotze et al. 1988). Local induction anoma-
lies were found as well within the preandean depression zone,
e.g. the Salar de Atacama
The Pre-Cordillera and the Longitudinal Valley show again a
moderately, but not very well structured crust. Geomagnetic
deep sounding data suggest deeply fractured zones within the
Coastal Cordillera, striking about E-W.
We summarize that more or less all data observed indicate a
moderately to low resistive Andean crust, with even lower re-
sistivities beneath the Western Cordillera, the Altiplano and
the western ranges of the Eastern Cordillera. This shows that
the process of subduction of the Pacific Plate beneath wes-
tern South America affects the whole Andean crust from the
coast towards the foreland.

References

Chinn, D.S., Isacks, B.L., Barazangi, M. (1980): High fre-


quency seismic wave propagation in western South America
along the continetal margin, in the Nazca plate and across
the Altiplano. Geophys. J.R. astron. Sot., m, 209-244
Gdtze, H.J., Schmidt, S., Strunk, S. (1988): Central Andean
Gravity Field and its Relation to Crustal Structures. In:
Bahlburg, H., Breitkreuz, Ch., Giese, P. (eds.): The Southern
Central Andes. Lecture Notes in Earth Sciences, Springer, u,
199-208
Schwarz, G., Haak, V., Martinez, E., Bannister, J. (1984):
The electrical conductivity of the Andean crust in northern
Chile and southern Bolivia as inferred from magnetotelluric
measurements. J. Geophys., s, 169-178
Schwarz, G., Martinez, E., Bannister, J. (1986): Untersuchun-
gen zur elektrischen Leitfahigkeit in den Zentralen Anden.
Berliner geowiss. Abh., (A) a, 49-72
Wigger, P.J. (1988): Seismicity and crustal structure of the
Central Andes.In: Bahlburg, H., Breitkreuz, Ch., Giese, P.
(eds.): The Southern Central Andes. Lecture Notes in Earth
Sciences, Springer, u, 209-229
._..

45

CRUSTAL STRUCTURE OF NORTHERN CHILE AND NW-ARGENTINA

PETER, Ji WIGGER? MA EL ARANEDA', PETE GIESE? WOLF-DIETE


NY , MICHAEL SCHMITZ!? and JOSE VIRAMONTE ?
HEINSOHN , PETER RCiWER

1) Inst. fiirGeophysikalische Wissenschaften, FUB,


Rheinbabenallee 49, D 1000 BERLIN 33
2) Depto. de Geofisica, U'de Chile,
Casilla 2377, Santiago de Chile
3) Universidad National de Salta,
Buenos Aires 177, 4400 Salta, R. Argentina

Abstract
Seismic refraction investigations have been carried out in
northern Chile and north-western Argentina in several field
campaigns during the years 1982, 1984 and 1987. The net of
mostly reversed profiles covers the main morpho-structural
units of the Central Andes: Coastal Cordillera, Longitudinal
Valley, Pre-Cordillera and Pre-Andean Depression, Western
Cordillera and Puna up to the Eastern Cordillera (see fig.).
Field work was carried out in a broad Argentine-chilean-
German cooperation, logistically supported by the Chilean
Navy (sea blasts) and YPF (drilling for blasts at the Puna)
and mainly sponsored by the German Research Organisation
(DFG).
x.. w
!r i: Y w Y
, r,

+i
t t
T
+ & ! h: * ..e.. .,,. I
h
) ( ;; . /
:._j L;. /r .- BOLIVIA .a.
0 X
> /. ._.... I
c, I I
m.

0
1 L
r
,.+
A RG EN

CENTRAL ANDES
GEOTRAVERSE
- _
I-.m...r

46

The field data of analogues and PCM type have been converted
to digital data and were processed in Berlin. The presented
results obtained by one- and two-dimensional model
calculations are based on 17 record sections with recording
distances up to 450 km.

Crustal structure of the Andean belts varies strongly in


north and south directions expressing the tectonic and
magmatic evolution of the convergent continental margin.
The structure beneath the Coastal Cordillera can be described
down to 45 km. P-Wave velocities of 5.8-6.0 km/s are observed
at the surface and already in 7-10 km a value of 6.7 km/s has
to be stated continued by a further positive gradient down to
a discontinuity at a depth of 20 km where 7.2 km/s is
reached. The mean velocity for this upper 20 km crustal part
amounts to 6.6-6.7 km/s. Deeper structure down to 38-43 km
displays at least two low velocity zones (LVZ) characterized
by velocities between 6.2 - 6.4 km/s interconnected by a high
velocity structure (7.6) at about 30 km depth. The velocity
increases at the discontinuity at 38-43 km depth to 8.3 km/s
and is proved by clear prograde phases.
A rapid change in structure as well as in crustal thickness
is evident to eastern direction. The crustal segment of
strongest variation belongs to the Longitudinal Valley and
Pre-Cordillera where thickness reaches about 60 km. Contrary
to the Coastal Cordillera under the Pre- and Western
Cordillera P-wave velocity at the southern W-E profile
increases slightly and reaches 6.1 km/s at about 30 km depth.
The range between 30 and 60 km is characterised by a change
of high and low velocities which reach a value of 8.1 km at
the Moho. The mean velocity for the total crust as well as
for the levels O-30 km and 30-60 km is 5.9 km/s.
In the area of the geothermal anomaly of El Tatio between
Pre-Cordillera and Altiplano there are strong indications for
a LVZ at depths greater than 10 km.

Crustal structure of the Puna's crust is similar to the


Western Cordillera. The top of the thick lower crust is found
in 25-30 km, below there exists a LVZ and the Moho depth is
58-65 km. The mean velocity has a value of 5.9 km/s.

Maximum crustal thickness of more than 70 km is expected


under the western Cordillera at latitude of about 23'South.
Crustal thickening of the Central Andes as well as Structure
and state at the up to 40 km thick lower crust will be
discussed in view of accretion and stacking. Different models
regarding tectonic and accretional processes for the
structure beneath the Coastal Cordillera will be presented.
47

BOLIVIAN GLOBAL GEOSCIENCES TRANSECTS PROJECT.

TELLERIA, J.L.() et al.

l Centro de lnvestigaciones Geologicas, Casilla 12198, Universidad Mayor de San Andres,


La Paz, Bolivia

The Bolivian Global Geosciences Transects Project (CC-7, ICL) involved three
transects executed in the area of Bolivia (SA4, SAl and SA2, from south to north).
Theese threecorridors are connected with the corresponding transects of the
sourrounding countries (Argentina, Brazil, Peru and Chile).

The transects represents compilations, in maps and cross sections at l:l.OOO


000 scale, of existing geological and geophysical data, along selected corridors
crossing cricial structures for a better understanding of the nature and evolution of
the lithosphere.

Each transect covers an area of 100 km wide, with a variable longitude


according to the different drafts (560, 1200 and 880 km).

Theese three transects cut, from west to east, the following major geological
domains : l- OCCIDENTAL ANDES, a Volcanic Arc. 2 - ALTIPLANO, a T- Cretaceous
sedimentary basin. 3- EASTERN ANDES and SUBANDEAN RANGES, a wide Paleozoic-
Mesozoic sedimentary basin, folded and thrust faulted. 4 - CHACO PLAINS and
LLANOS, a thin T-Q sedimentary basin. 5- BRAZILIAN SHIELD, a Precambrian
cratonic terrane.

The geological strip maps shows the direct correlation betbeen the major
geological terranes with the geophysical maps and profiles of gravimetric anomalies.
The interpretatives CROSS SECTIONS shows the main geotectonic features, geological
strata and MOHO depths.

The, Moho depths calculated, thicken toward the Central part of the Altiplano,
reaching 66 km and Bouger anomalies about to more than -400 milligals.

All the results are presented in a typical GGT display (one square meter for
each transect).
.__x -. -w.. ^__ - ._I*__._,., --P...
49

SPANISH-CIRGENfINIfiN GEODYNMIC GPS NET CVJTCIRTIC PROJECT

J. Ballesterost, M.Berrocasot*, M. Catalan, F.Cruzl


R. Estradatl$, A. Lujant, J. Mui(on*t*, J. Sanchez de1 ToroSaX
J.C. SastreS$, R. SotoS1 b J.G. Viramonte 6

I Instituto GeogrPfico de EspaXa


$L Real Observatorio de la Armada, San Fernando, Cadir, Espala.
1$$ Servicio Geografico de1 Ejercito, EspaXa.
# Uni versidad Naci onal de Sal ta-CONICET- Instituto Rnt&-tic0
fkgentino, Buenos Aires, 177 - 4400 Salta, Argentina.

Se presenta una red geodbsica GPS antdrtica que une las Shetland
de1 Sur , Peni nsul a Antartica y el continente Sudamericano. Se
prevee usar la red con medidas sistematicas de alta presici6n con
fines -geodinamicos para medida de1 movimiento relative en tiempo
real, de1 centre de expan6ic)n de1 Bransfield, Arco de Scotia etc.

Key words: GPS, Geodinamics, Qntartica, Geodetic net, Spain,


Argentina.

Introduction

During 87-88; 88-89 and 09-90 antarti c 6ummer a Spani 6h-


Argentinian geodet i c-geodynami c antartic project built up and
measured a high precision geodetic GPS net, that 1 inked South
Shetland 16land6, CIntartic Peninsula and South America.
The monument 6, the geometry of the net and the hardware and
software used as well as systematic future measurements will
enable its usage for geodynamic purposes.

The Rntutic GPS net

Fig. 1 tl3 and Bl,shows the present Spanish-Argentinian GPS


Antartic net at present and it6 future development.
Double frequency GPS 4000 SLD Trimble Navigation receiver6
and Trimvectm software were used. This fact has made it possible
to reach good precision in lb6olute positionament (Table 1) and
high precision in relative positionament between two 6tat i on
(Table 2).
Latitude, Longitude and ephemeris values are referred to the
World Geodetic System (WGS-B4).
During 87-88 southern rummer GPS and Transit absolute
measurements were made in 3 station6 (Estrada 1989): 2 in
YALVINAS
ISLAND Q BP

-- ---- - ---mm _

a.0 --__
Q --.
Ip -

&P REFERENCES
0 6okm
lOPS Srotion B001:
we Drokr Porrogo
lPloned GPS Storlon

g. 1 Key maps show an Argcntinian-Spain Antortic GPS Net.


- II_._ -s. - IX ., _
..-- __?..._

51

Livingston Island. (Punta Polaca and Base Antartica Eepaiiola,


Juan Carlos I1 and 1 in Deception Island. (Destacamento Naval
Decepci 6n, Argentina); while in 88-89, four high precision GPS
stations were built and measured (Ballesteros et al. 1989):
Livingston Island: Base Qntdrtica E6pPRO1 a (B&E-GPS 1002)
Deception Island: Fumarole Bay (GPS 20021
Antartic Peninsula: Base Almirante Brown, Argentina (GPS 3001)
Greenwich Island: Base Arturo Pratt, Chile (GPS 40011
During 89-90 summer we hope to link the CIntartic GPS net with
a point built in Unhuaia City, Argentina, South fimerica and check
the GPS Antartic net again.
We hope that future systematic measurements of the net,
besides future Navstar constell at i on, hardware and software GPS
system improvements will enable the net to be used for geodynamic
purposes, specially for the control of the absolute movement in
real time of the Bransfield rift spreading center.

Ref l rences

Estradr Nbrida R. 1989. Resumen de resultados de 10s trabajos


topogr&fico-geodCsicoo realizados en la campaXa antartica
1987-80. Servicio Geogr&f ice de1 Ejercito, Espaiia.

Ballesteroc,J., Berrocamo,tl. .Catrlan.tl., Cruz,F.. Estrada,R.,


Luj&,Ai, th18oz,J.; Skhez d&l ho, J., Sastrr,J.C.
Soto, R., y Viramonte,J.G. 1969 Las Campaiias geodesicas
1987-88; 1988-89 en 1 ar Shet 1 and de1 Eur. 11 I Symposium de
estudi OS Antbrticos, Gredos, 3-5 Octubre 1989, Spain.
Reeumenes

TABLE 1

TRIOlHENSlDNAL RBSOLUTE POCISIONING (6PS - 1988-89)

STRT ION LAT. LONG. LILT.

lOOl-PUNTA POLKA (LIVINGSTON Is .I b25943.37021 S 60'23'40.IPbOB' W 33.410


1002-B&E (LIVINGSTON It .I 623946.95058 S 602320.21120 Y 30.668
2001~IRGENTINE St.(DECEPTION Is. ) b2SB30.336Sl S 60415S.513B5 Y 20.037
2002-FUIMROLE BI\Y (DECEPTION Is. ) 625741.13540 S b042S9.7011b Y 19.103
3001~LL. BROYN St.(PARAISO BAY) No drtr
4001-R. PRAT. St. (GREENYICH Is.) 622846.35658 S 593947.61355 Y 23.590

TRRNSIT (YGS - a4) 1987-88

1002-BAE (LIVINGSTON Is.) 623946.68 5 b0231a.70 Y 14.88


2001~LR6ENTINE St. (DECPTION Is.) 625828.33 S b0422.46 w 224.50
52

TMLE 2

Barr line 3: ME-6PS 1002 to Al.Brown-GPS 3001


28 Jan-89 Lat Long &It
Station 1 ,b2' 39' 46.78564" S 60 23' 20.50016 W 37.241
Station 2 : Triplr Float Fixed
Latitudr 64 5345.75426 S b4' 5345.7bSb3 S 64 53'45.76539" S
Longitude 62 52'13.13249' W 62 5213.08720 Y 62 S2'13 .09208" Y
Alti tudr 48.127 47.672 47.677

dx -213721.495 -213721.198 -213721.251


dy 138298.094 138298.621 138298.584
dz -10990s.559 -109985.296 -109985.297
dh 10.886 10.431 10.436
tlax. incr. FLT-FIX....S.I cm
Recomndcd Sol. Triple.

Barr line b: BAE GPS 1002 to Fumrolc Bay 6PS 2002


a February -89
Station 1 62 39'46.79102" S 60232O.bS322 W 42.062
Station 2
Latitude 625741.02229 S 62'57'41.02080' S 62'57'41.02090" S
Longitudv 60'42'59.99319" W b04239.99b7Sa Y 604259.99704 S
AItitudi 33.025 l 33.031 l 33.043 n

dx -29094.644 l -29094.667 l -29094.669 8


dy L7552.562 a 17552.49? l 17SS2.495 I
dz -191aa.343 a -15188.327 l -15188.339 I
dh -9.037 R -9.031 (I -9.019 a
Hrx. Incr. FL1 - FIX . .. 1.2 cm
Recoronded sol. . . . TRIPLE

Sam Line 9: ME GPS 1002 to fwturo Prat B. 6PS 4001


20 February -89
Station 1 62'39'46.91010' 5 602320.43663 Y 38.301
Station 2
Latitude 622046.20692 S 622046.20076* S 622846 . 28648 S
Longitude 59'39'47.87320" W S93947.00bSSY S 593947;89036 Y
Altitude 29.444 29.471 29.481

dx 41309.059 I 41389.879 a 41389.857 l

dv 2919.514 a 2919.450 II 2919.373 l


dz 9429.326 I 9429.276 8 9429.295 n
dh -8.857 l -8.830 I) -8.820 l
flax. Incr. FLT - FIX .. . 7.7 cm
Recorrnded sol. ,.. TRIPLE
53

A GLOBAL POSITIONING SYSTEM EXPERIMENT IN THE ANDES

M. P. GO&X&&~, T. H. Dixonl, S. Stein2, R. Gordon2, and S. Sacks3


l~arth and S ace Sciences Division, Jet Propulsion Laboratory, Caltech, Pasadena. California,
9 1109 USA, s Department of Geological Sciences, Northwestern University, Evanston, Illinois,
~0&~3 IJJt 3Camegie Institution, Department of Terrestrial Magnetism, Washington. D.C.,
*

Abstract:
A geodetic experiment is being planned to measure the convergence between the Nazca
and the South American plates and its relationship to shortening and uplift within the Andes. The
experiment will employ Global Positioning System (GPS) receivers operated simultaneously at
Juan Fernandez and San Felix Islands, the Andean mainland, and 10 or more globally disaibuted
tracking sites. The Andean portion of the experiment will consist of networks bounding the
major subdivisions of the Andes in both a north-south and east-west sense. This coverage
should allow increased understanding of the interaction between the four major segments of the
subducting oceanic slab with tectonic elements of the overlying continental plate.

Introduction:
NASAs Jet Propulsion Laboratory in conjunction with Northwestern University and the
Carnegie Institution is currently in the early stages of planning for a space-based geodetic
experiment in the southeast Pacific and Andes beginning in 1991. using the Global Positioning
System (GPS). This experiment, known as SNAPP (South America-Nazca Plate Project) is
designed to monitor subduction and key aspects of Andean tectonics. GPS is a highly accurate
space-based geodetic technique that, for the first time. offers the flexibility and portability for
economically measuring a relatively dense geodetic network extending several thousand
kilometers at centimeter accuracy. The experiment being planned involves multiple
measurements of baselines between south Pacific islands and the mainland of western South
America, a dense network across the Andes, and a globally distributed tracking network.
Multiple GPS campaigns made a few years apart will allow direct measurement of the rate of
subduction across the trench as well as the rate of horizontal shortening and vertical uplift across
individual tectonic provinces of the Andes and relationships to differences in the subduction of
the Nazca oceanic plate. In this abstract. we briefly describe the scientific rationale for the
Andean part of this experiment and some aspects of the GPS geodetic technique.

Nnzca Plate Subduction and Segmentation:


Subduction of the Nazca plate beneath South America occurs along four major segments
defined by the seismicity of the Benioff zone (e.g., l-5). Two segments, one beneath Peru (2S
to 15%) and the other beneath middle Chile and Argentina from 27s to 33% have very shallow
dips of 5-100 to the east. The other two segments, one between the flat-plate segments described
above and the other south of 33% dip at about 300 to the east. The rate of convergence between
the Nazca and South American plates has been estimated at 8.3 cm&r from global plate motion
models, 18.7-12.8 cm/yr from seismic slip, and 6.5 cm&r from space geodetic measurements
(6). Understanding the discrepancies between the different rate estimates and how convergence
is apportioned between subduction and deformation in the overlying South American plate is a
major goal of SNAPP.

Andean Tectonics:
A direct nlationship exists between the segmentation of the subducted Nazcaglate and
characteristics of the overlying segments of the Andes (7). The region above the 30 -dipping
slab in southern Peru, northern Chile and Bolivia (18S to 27S) contains a longitudinal valley
54

along the coast that is locally under extension (8). A well developed volcanic arc is composed of
Quatemary andesitic stratovolcanoes that make up the crest and the western edge of the high
plateaus. The Altiplano-Puna plateaus are internally drainiig basins that experienced significant
uplift at the end of the Neogene. The cause of the plateaus is not known, however Isacks (9) has
recently suggested thlckemng of a lower crustal weak zone since the middle Miocene. To the
east of the plateaus lies the hinterland and foreland, which are made up of the Eastern Cordillera
and the Subandean belt. The Subandean belt is a thin-skinned fold and thrust belt, that is
analogous to the thii-skinned fold and thrust belt of the Canadian Rocky Mountains, and has
been active from the Miocene to the present. Earthquake focal mechanisms show this region is
still under corn ssion as do Quatemary reverse faults in the Subandean to the north (8).
Reverse faults r unding the Subandean belt root beneath the Eastern Cordillera, which is
composed of Paleozoic and older strata deformed by thrust faults and high angle reverse faults
that were mostl active during the Miocene.
Along L Chil&rgentina flat plate segment (27s to 33OS)the fomarc rises steadily from
the am without a fotwrc basin or evidence for extensional deformation. Quatemary volcanism is
absent and there are only minor older volcanics of Miocene age (10). To the east of the forearc
and the crest of the Andes lie the Frontal Cord&m and Precordillera, which are thin-skinned
fold and thrust belts. Evidence suggests the Frontal Cordillera was active in the Miocene,
whereas thcPrecordillera has been active in PlioPleistocene time, suggestive of an eastward
migrating thrust belt. To the east of this thrust belt are the Pampeanas Ranges, a major system of
foreland crystalline basement uplifts, 2-4 km to locally 6 km high, 25-75 km wide. These
basement uplifts are separated by broad intervening basins (typically wider than the uplifts) at
about 1 km elevation. The ranges have been uplifted along north- to northwest-striking reverse
faults that dip W-60, basedon field obsetvations at the surface and subsurface focal mechanism
solutions. The faults commonly have about 6 km of throw and uplift Precambrian and Paleozoic
crystalline basement over upper Cenozoic strata Geologic relations indicate PliocenePleistocene
deformation; compressional earthquakes indicates that the deformation is continuing today. A
case has been made for the Pampeanas Ranges being analogous to the Laramide foreland
basement uplifts of the Rocky Mountains in western North America.
The segment of the Andes above the flat subducting slab in Peru and southern Ecuador
(2OSto 18OS)is generally similar to the Chile-Argentina flat plate segment. No forearc basin is
reported. The volcanic am was last active in the middle Miocene, with little or no volcanism
since about 5 Ma when flat plate subduction began. Hinterland and foreland folding is reported
during the Neogene and into the Quaternary. Foreland basement uplifts are found to the east of
the fold belt. These uplifts have many of the same characteristics as the Pampeanas Range, but
are less well developed, perhaps due to the younger age of the flat slab in Peru and the implied
smaller shortening.
The,southem 30O-dipping segment extends south of 33S to at least 38s. This segment
has an extensional forearc basin. the Central Valley. The Princiual Cordillera is a folded and
faulted belt deformed from the r&ldle Tertiary to the Miocene. The volcanic arc is well defined
by Quaternary basaltic andesite stratovolcanoes. Unlike the 3Odipping segment to the north the
deformation m this segment of the Andes is confined to a narrOw zone. To the east of this area
exists an enotmous field of Miocene to present basal&
A few other relationships deserve mention. Earthquakes within the upper plate are most
frequent and largest in magnitude above the flat plate subduction zones. The forearc and plateaus
are for the most part quiet seismically (7). The forearc is locally under extension above the 30-
dipping segments (8) and under compression above the flat plate segments. The hinterland and
foreland are consistently under compression along the entire length of the Andes (8, 11). The
eastern extent of the deformation in the foreland corresponds to the eastern extent of the Benioff
zone in all the segments discussed above. Jordan et al. (1983) suggest that the preexisting
structure and geometry of the upper plate may have had an influence on the location and
geometry of segments in the subducting plate.
Estimates of the total shortening across the Andes are limited and rates of deformation are
uncertain. Geologic reconstructions suggest roughly 20-3096 of shortening (60-200 km) across
55

the Subandean fold and thrust belt. depending on latitude (12. 13). Shortening across the
Pampeanas is estimated at 10-20 km (14) and about 50 km across the Precordillera at that latitude
for a total of about 70 km across this segment. Megard (15) has estimated a little over a hundred
kilometers of shortening over part of the Andes in Peru. If this deformation has been occurring
over about the past 10 Ma (7). then total rates of shortening across the Andes are of order 1
cm/yr. Backs (9) has constructed a model for the deformation and uplift of the central Andes that
lii the horizontal shortening in the foreland thrust belts with the thickening and vertical uplift
of the high plateaus. This model predicts from one to a few hundred kilometers of shortening
over about 15 Ma, which also suggests a rate of shortening across the Andes of order 1 cm/yr.
Isacks model also predicts decmased shortening away from the Bolivia orocline (the bend in the
Andes at about 20%) and concomitant clockwise rotation of the arc south of the bend and
counterclockwise rotation of the arc north of the bend, both of which appear supported by
paleomagnetic data (16-18). Uplift of the Andes over the past 10 Ma has been rapid,
approaching 1 c* (19,20).

GPS Geodesy:
The precision and accuracy of GPS geodetic measurements for regional tectonics
applications depends on the location of the experiment. baseline length (receiver spacing),
experiment design and analytical technique. Numerous GPS experiments have been conducted
in southern California, where very long baseline interferometry (VLBI) measurements are
available for comparison. These data, as well as GPS data from various other experiments allow
us to predict the liiely performance of GPS in the Andes.
The repeatability of a GPS measurement is one measure of precision, and for southern
California measurements, has proven to be about 5-10 mm plus 1-2 parts in lo* of baseline
length for the horizontal components (21-23). The accuracy of these measurements is roughly
comparable to their repeatability based on comparisons to VLBI data. The precision and
accuracy of the vertical component is somewhat worse than the horizontal components, in the
range 3-6 cm, and is not correlated with baseline length.
In other regions, the precision and accuracy of GPS may degrade, depending on the level
of and analytical treatment for two major GPS errors sources. uncertainties in the positions of the
GPS satellites, and signal propa atlon effects due to the troposphere. Preliminary results from
GPS experiments in the Gulf of &l*ifomia (23-25), the northern Caribbean (26) and the northern
Andes and Central America (27-29) suggest that these errors can be minimired such that the total
uncertainty in the horizontal components of GPS measurements in these regions is no worse than
10 mm plus 2-3 parts in 108 of baseline length.
The only impediment to achieving similar results in the central and southern Andes would
be increased levels of error associated with satellite orbit uncertainties, caused by greater distance
from GPS tracking (fiducial) sites. Typically, GPS &ta is acquired at these fiducial sites
simultaneously with the experiment in the region of interest. The positions of the fiducial sites
are known u priori from independent measurement, for example by VLBI, thus data from these
sites provide independent information on satellite trajectories. We have performed a covariance
analysis using synthetic GPS data and proposed site locations, as well as realistic satellite
geometries for the early 1990s, to determine quantitatively the level of orbit-nlated errors for the
SNAPP experiment. These studies suggest that accuracy equivalent to that achieved in
experiments in the Caribbean and northern South America can be achieved throughout the Andes,
providing several tracking sites in the southern hemisphere (e.g., Australia and New Zealand) are
employed.
GPS receivers will be glaced in three latitudinal bands across gmvinces within the three
segments defined south of 18 S (Fig. 1). In the region above the 30 -Qppmg slab in southern
Peru, Chile and Bolivia receivers will be placed at the east and west ends of the forearc, on either
side of the Altiplano, to the east of the Subandean belt, and perhaps between the Subandean and
Eastern Cordillera. In the region above the flat slab segment in Chile and Argentina receivers
will be placed: on the coast, on the crest of the mountains, between the Precordillera and
..,-
.l__l . . .._.._ . _^. _,.. ..^

56

Pampeanas and east of the Pampeanas. In the region above the southern 30-dipping slab
meivers will be placed: on the coast. near the arc, and east of the Principal Cordillcra. Our frost
qerimcnt is tentatively planned for 1991, with rcuccupations in 1993 and 1995.

References: 1 Stauder, W., 1973, J. Gcophys. Res. 78.5033~5061. 2 Stauder, W.. 1975, J.
Geophys. Res., 80, 1053-1064, 3 Barazangi, M. & B. Isacks, 1979, Geophys. J. Royal Ast.
Sot. 57.537-555; 4 Isacks, B. & M. Barazangi, 1977, in Island
back Amer. Geophys.Un., 99-114; 5 Hasegawa, A. & L Sacks, 1981, J. Gcophys.
Res. 86, 4971-880; 6 Stein et al., 1986, Geophys. Res. Lett. 13.713-716; 7 Jordan et al.,
1983, Geol. Sot. Amer. Bull. 94.341-361; 8 Sebrier, M.et al., 1985. Tectonics 4,739-780; 9
Isacks, B., 1988, J. Geophys. Res. 93, 3211-3231; 10 Kay. S., et al., 1988, J. South Amer.
Earth Sci. 1.21-38; 11 Suarez, G. P. Molnar & B. Burchfiel, 1983. J. Geophys. Res., 88,
10,403-10,429; 12 Allmendin er, R.et al., 1983, Tectonics 2, l-6; 13 Sheffels, B., B.
Burchfiel & P. Molnar, 1986, E8 S Trans. Amer. Geophys. Un. 44.1241; 14 Jordan, T. & R.
Allmendinger, 1986, Amer. J. Sci. 286,737-764; 15 Megard, F.,1984, J. Gcol. Sot. London
141,893~900,16 Beck, M.. R. Drake & R. Butler, 1986, Geology 4.132-126; 17 Beck, M.,
1988. J. South Amer. Earth Sci. 1, 39-52; 18 Kono, M., K. Heki & Y. Hamano. 1985, J.
Geodyn., 2,193-209,19 Crough, T., 1983, Earth Planet. Sci. I.&t. 64396397; 20 Benjamin
et al., 1987, Geology 15.680-683; 21 Blewitt, G.. 1989, J. Geophys. Res., 94, 10187-10203;
22 Davis, J., W. Prescott. J. Svarc, K. Wendt. 1989, J. Geophys. Res., 94, 1363513650,
1989; 23 Tralli, D. & T. Dixon, 1988, Geophys. Res. Lett. 15. 353-356; 24 Tralli, D., T.
Dixon, & S. Stephens, 1988, J. Geophys. Res. 93, 65456557 25 Dixon. T., et al., 1990,
AAPG Mem 47, In Press; 26 Dixon et al., 1990, J. Geophs. Res., in press; 27 Kellogg. J. T.
Dixon & R Neilan. 1989, EOS:Trans. Am Geophys. U. 70, 649-656; 28 Dixon. T. & S.
Komreich Wolf, 1990, Geophys. Res. Len., In Press; 29 Komreich Wolf, S., T. Dixon & J.
Freymueller, 1990, Geophys. Res. l&t., In Press;

Figure 1. Map of
geodetic network
20S
planned for the
SNAPP experiment.
Dashed lines are the
baselinesto be
measured. Plate
velocity is shown
for reference.
IOS
PALE~MAGNETI~M 0~ LATE Cenozoic &DEAN BASINS
AND COMMENTS ON THE BOLIVIAN OROCLINE HYPOTHESIS

Bruce J. MacFadden, Institute for the Study of Continents (INSTOC) and Department of
Geological Sciences, Cornell University, Ithaca, New York 14853-1504 USA (Permanent Address:.
Florida Museum of Natural History, University of Florida, Gainesville, Florida 32611 USA)

Paleomagnetic data are presented for late Cenozoic high-elevation basins in the Cordillera
Oriental of Bolivia. These results suggest negligible rotation at 17OS lat. and significant (17.8 f
5.1) clockwise rotation at 22 S. These rotations may be interpreted as (1) local block rotations,
or (2) the preferred hypothesis of Neogene oroclinal bending of the Andes.

Nuevos datos paleomagneticos son presentados para las cuencas de alta correspondientes al
Cenozoic0 tardfo en la Cordillera Oriental de Bolivia. Estos resultados indican una
insignificante rota&n a 10s 170 S y una significante rota&n (17.8 f 5.1) en la direcci6n de las
agujas de1 reloj a 10s 22 S. Probablemente, estas rotaci6nes representen (1) rotaci6nes locales de
bloques estructurales, o (2) en la hip6tesis preferida, deformaci6n tectonica del oroclinal
Boliviano durante el Ne6geno.

Key Words: Paleornagnetism, Cenozoic, Andes, Basins, Tectonics, Qrocline

Introduction

The high-elevation sedimentary basins of the Cordillera Oriental of Bolivia provide an


excellent opportunity to understand the tectonic evolution of the Andes. This paper presents
paleomagnetic data from Quebrada Honda and Micana. In conjunction with previous
paleomagnetic results from Salla, Bolivia (MacFadden et al., 1985) and the Ocros dyke swarm of
Peru (Heki et al., 1985) these data are interpreted relative to possible local rotation versus
oroclinal rotation.

Geological Setting

(1) Quebrada Honda, consisting of a 305-m thick section, is located in the southern
Altiplano at lat. 22 S. Hoffstctter (1977) describccl the presence of Friasian land mammals from
this locality, and its age is middle Miocene, or about 12 Ma (MacFadden et al., 1990). (2) Micana,
consisting of a 205 m thick section, is located in the Cordillera Oriental at lat. 17 S. MacFadden
et al. (1990) have determined that this locality is late Miocene, or about 7 Ma. (3) Salla,
r _... .~_^__ _ _. ._ I_,.--_

58

Table 1. Paleomagnetic data for late Cenozoic localities from Bolivia (taken from MacFadden et
al., 1990). Ns/Nu=number of sites originally sampled/number of sites used for overall mean
directions; RES, Resultant (Fisher, 1953); args, 95% cone of confidence; R, amount of rotation
(Beck et al., 1986); A R, error in calculatton of rotation (Beck et al., 1986).
_----- -___________________-~~-~~~~~~-~~-~-~~~~~~~~~~-___~~~
Locality Lat. Dec. Inc. Ns/Nu RES org5 K R AR
CS) (0) (1 (1 (1 (*I

Salla 17 353.4 -37.4 104/58 53.7 5.4 13.1 6.6 6.8


Micana 17 355.2 -25.8 39/25 24.0 5.9 24.8 -4.8 6.6
C&Honda 22 17.8 -40.7 U/79 74.6 3.9 17.9 17.8 5.1

consisting of a 540 m thick section, is located in the eastern Cordillera at lat. 170 S. The age of
this locality is late Oligocene to early Miocene, or about 28 to 22 Ma (MacFadden et al., 1985;
Naeser et al., 1987). (41The Neogene Ocros dyke swarm is located in central Peru at 13OS. Heki
et al. (19851 present the paleomagnetic results from this locality which are reviewed here in a
context of the tectonic development of the Andes.

Paleomagnetic Procedures and Analyses

In all three basins listed above, paleomagnetic sites, or horizons (each consisting of three
or more separately oriented samples), were collected throughout the fine-grained deposits and
were spaced vertically at intervals averaging about 6 m. This sampling regime, which spans at
least 1 myr in each of the sedimentary basins, therefore averages out any possible instantaneous
effects of non-dipole components of the ancient geomagnetic field in calculating a mean direction
for the each stratigraphic section.
Isothermal remanent magnetization experiments indicate that the magnetic mineralogy is
complex and could result from carriers where saturation occurs in low fields or others where
saturation does not occur in applied fields up to 3 Tesla (MacFadden et al., 19901. Given this
potentially complex magnetization for each site, individual samples were subjected to either
stepwise (usually in 4 to 10 increments, or morel treatments using either standard alternating
field or thermal demagnetization. Therefore, all sites were treated by both techniques to better
isolate characteristic magnetizations. Paleomagnetic samples were measured in a cryogenic
magnetometer located in a low-field shielded room at the University of Florida.
Using these laboratory procedures, linear trajectories were isolated for many samples and
principal component analysis (Kirschvink, 19801 was used to calculate the least-squares, best-
fitting line for determining the characteristic, stable direction, presumably primary in origin.
Within sites the individual sample directions were then combined to determine a mean direction
for that stratigraphic horizon using Fisher statistics (Fisher, 1953). In order to calculate the
overall formational mean directions, only sites whose R-statistic (Resultant in Table 1 here1
satisfied Watsons significance point criteria (Irving, 1964) were used. Because the sediments arc
essentially flat-lying, fold tests to determine stability of remanence were impossible. However,
the presence of site means with antipodal normal and reversed populations suggests ancient
magnetizations rather than spurious secondary overprints (MacFadden et al., 19901.
The R k AR values for the three sites studied here indicate different amounts of rotation
relative to the late Tertiary reference pole for stable South America. Given the associated
errors, Salla and Micana have negligible, whereas Quebrada Honda indicates significant (17.8 f
5.1) clockwise rotation since time of deposition (no stratigraphic trends in declination anomalies
are apparent from any of the three localities).
Figure. 1. Map of central part of South America showing general extent of the Andes (shaded)
and the sites described in the text. The arrows within the half-circles represent the amount of
rotation (R) and the black cones encircling the arrows represent AR, the estimated error.

Results and Interpretation

Of the possible structural/tectonic interpretations of rotated terranes (Beck, 1989) two


seem plausible. (1) The data above indicate local, small-block rotation. If this is the case, then
the scale on which rotated blocks must have occurred would have been greater than 6 km2,
indicating some meso- to large scale tectonic process. (2) These data support the hypothesis of
oroclinal bending of the Andean orogen. Carey (1955) proposed the concept of an orocline to
describe large scale tectonic bending of mountain ranges such as the Andes, from which the term
Bolivian Orocline was proposed (1958). In plan view, the Andes are curved with about 45 of
flexure along an axis aligned SW-NE inland along the Arica deflection.
Numerous studies have analyzed the paleomagnetism of volcanic and sedimentary rocks of
Mesozoic age (summarized in Beck, 1988; Isacks, 1988) on either limb of the supposed Bolivian
Orocline. Isacks (1988) interpret these data to suggest oroclinal bending, but given the available
data, the age of bending is only constrained as post-Cretaceous. If the principal phase of
oroclinal bending is related to subduction of the Nazca plate and Andean uplift (Isacks, 19881,
then this tectonic deformation should have occurred very recently, i. e., during the late Cenozoic.
However, there have been few paleomagnetic studies of young rocks in relevant areas. In
addition to the data presented in Table 1 here, Heki et al. (19851 reported 14.5 f 5.5 of
counterclockwise rotation for the Neogene Ocros dyke swarm at 13 S lat. in Peru. Thus, these
four data sets would suggest counterclockwise rotation at 13OS, negligible rotation at 17 S, and
clockwise rotation at 22OS during the late Cenozoic (Fig. 1). The magnitude and sense of these
data corroborate the hypothesis of about lo-209 of late Cenozoic oroclinal bending on both the
north and south limbs of the presumed orocline, with little bending near the axis (hacks, 1988;
Figs. I,2 here). This, however, does not account for all the bending necessary to account for the 45
bend at the Arica deflection. As Beck (1988) has suggested (although he does not advocate the
oroclinal model), there may have also been an earlier phase of structural deformation.
60

iii
QUESRADA
p 20
HONDA
9 SOUTHERN LIMB
Y 10
3
i= 0
f
s
iz -lo
a -20

5
-30 J
Figure 2. Predicted values (shaded) of oroclinal rotation (Isacks, 1988) and the late Cenozoic
paleomagnetic data discussed here. The bars on either side of the calculated rotation (R)
represent the estimated error, AR.

I prefer the hypothesis of Neogene oroclinal bending because the paleomagnetic data
presented here are consistent with Isacks (1988) model. However, these data could also
represent local block rotations. Additional paleomagnetic data points are needed from other
rocks in the Andes to further test these two alternate hypotheses
Acknowledgmenb. This research was supported by US National Science Foundation grants
DEB 7945861, EAR 8318903,and EAR 8716207.

Beck, hi. E., Jr. 1988. Analysis of late Jurassic-Recent paleomagnetic data from active plate
margins of South America. J. S. Amer. Earth Sci. 1:39-52.
Beck, M. E, Jr. 1989. Block rotations in continental crust: examples from western North America:
_In: Kissel, C., & C. Laj feds.). Paleomagnetic Rotations and Continental Deformation.
Khrwer Acad. Pub. Dordrecht. p.l-16.
Carey, S. W. 1955. The orocline concept in geotectonics. Proc. Royal Sot. Tasmania. 89255258.
Carey, S. W. 1958. The tectonic approach to continental drift. In: Carey, S. W. ted.). Continental
Drift-A Symposium: Hobart, Tasmania, Univ. Tasmania Press. p. 178-355.
Fisher, R. A. 1953. Dispersion on a sphere. Proc. Royal Sot., London. 217:2954X
Isacks, B. 1988. Uplift of the central Andean Plateau and bending of the Bolivian Orocline. J.
Geophys. Res. 93:3211-3231.
Heki, K., Hamano, Y., Kono, M., and U. Tadahide. 1985. Paleomagnetism of Neogene 0x0s dyke
swarm, the Peruvian Andes: implication for the Bolivian orocline. Geophys. J. Royal
Astron. Sot. 80527-534.
Hoffstetter, R., 1977, Un gisement de mammiferes miocene a Quebrada Honda (Sud Bolivien).
Comptes-rendus Acad. Sci., Paris D:1517-1529.
Irving, E. 1964. Paleomagnetism and its applications to geological and geophysical problems.
New York, Wiley. 399 p.
Kirschvink, J. L. 1980. The least-squares line and plane and the analysis of paleomagnetic data.
Geophys. J. Royal Astron. Sot. 62:699-719.
MacFadden, B. J. et al. 1990. Late Cenozoic paleomagnetism and chronology of Andean basins of
Bolivia, with comments on possible oroclinal bending. J. Geol. in press.
MacFadden, B. J., et al. 1985. Magnetic polarity stratigraphy and mammalian fauna of the
Deseadan (late Oligocene-early Miocene) Salla beds of northern Bolivia. J. Geol. 93:223-250.
61

PALAEOMAGNETIC AND Sl'RUC'J'UHAL CONSTRAlN'l':; ON MESO%c',lC-


RECENT THRJJST SHEET ROTATION IN TIiE PRECORDILLERA OF
NORTHERN CHlLE

ian HarLJ._e_x1, Elizabeth Jolleya and Peter Turner*.

1 Department of Geo.logy, University of Wales College


Cardiff, PO Box 914, Cardiff, CFI 3YE. UK.

2 School of Earth Sciences, University of Birmingham,


PO Box 363, Birmingham. UK.

Resume

Palaeomagnetic results are reported from two red bed se-


quences of late Cretaceous to Miocene age from the Precor-
dillera (Cordillera de Domeyko) of northern Chile. Com-
parison of isolated primary NRM components from the late
Cretaceous-Palaeocene Purilactis Formation with the APWP
for cratonic South America indicate that significant
post-Palaeocene clockwise rotation (33+/-12.5 ) has taken
place. Resolved primary NRM components from the Oligo-
Miocene Paciencia Group which unconformabky overlies the
Purilactis Formation also indicate significant clockwise
rotation but of considerably smaller magnitude (22+/-14.5
) when compared to the present day pole position.
Detailed structural mapping of the area utilising the
line balanced section technique has revealed the presence
of a series of thrust sheets which have been periodically
active during late Mesozoic to Holocene times. The above
data coupled with stratigraphical information indicate
that following deposition of the Purilactis Formation,
lower Eocene deformation (the Incaic Orogeny) resulted in
the folding and clockwise rotation of the Formation in
response to the eastward propagation of a major thrust
front (the Frontal Domeyko Thrust). Deposition of the
Paciencia Group occurred after the Incaic Orogeny during a
period of relative tectonic quiescence in the Oigocene and
Lower Miocene. Late Miocene reactivation along the Frontal
Domeyko Thrust resulted in uplift folding and clockwise
rotation of both the Paciencia Group and the Purilactis
Formation. Subsequent Quaternary and Holocene deformation
62

involved foreland directed thrust front propagation with


little or no significant rotation of the Purilactis Forna-
tion or Paciencia Group.
This study demonstrates that palaeomagnetically
detected rotation in the Precordillera of northern Chile,
200 km inland from the Peru-Chile trench, is due to thrust
front propagation associated with a compressional tectonic
regime, and not related to oroclinal bending. strike-slip
faulting or fault block rotation.
63

TERTIARY IGNINBRITES FROM NORTHERN CHILE: ANONALOUS


NAGNETIZATION EXPLAINED BY SELF-REVERSAL AND TECTONIC
ROTATION.

D. REX (*), P. TURNER (*), A. J. HARTLEY (**)

(*I School of Earth Sciences, Birmingham University,


Edgbaston, Birmingham B15 2TT

(**I Department of Geology, UCW Cardiff, Cathays Park


Cardiff

RESUMEN

El estudio paleomagnetico de una serie de muestras de


material de tipo ignimbritico procedente de1 norte de Chile
muestra una remanencia anomala. La magnetizacion se explica
a traves de la combination de un doble proceso de self-
reversal y rotation tectonica.
Key Words: Upper-Miocene ignimbrites Chile, self-reversal,
fore-arc rotation,

INTRODUCCION
Upper Miocene ignimbrites of the San Bartolo Group were
sampeled in the San Bartolo Dome area of Northern Chile. The
area lies immediately north of the large Salar de Atacama
basin. The stratigraphy of the area has been described by
Hollingworth and Ruthland (19681, Flint (1985) and Jolley et
al. (1990). The Oligo-Miocene Paciencia Group is overline
unconformably by the Mio-Pliocene San Bartolo Group. This
unit conteins seven ignimbrites rainging in age from 12.6 to
2.3 Ma. The Paciencia and San Bartolo groups were rotated
clockwise by Neogene thrusting (Jolley et al., 1990 and
Hartley et al., this conference). Paleomagnetic sampeling of
all seven ignimbrites was undertaken using field drilling
and hand sampling. Orientation was by sun compass and
magnetic compass. Meassurements of the NRM was made usinng
2-axis cryogenic and spinner magnetometers. Thermal and AF
demagnetization was done for specimens from each site. In
64

addition a variety of rock magnetic measurements were made


including studies of the susceptibility anisotropy.

PALEOMAGNETIC RESULTS

The intensity of the initial NEM ranged from 3541.8 to


50.4 A/m. Fig. la shows the distribution of the initial NEM
directions and the present day field direction (PDFD) for
the area. The most significant feature of these results is
that the sign of inclination is opposite to that of the
PDFD. After AF demagnetization of all the specimens at 50 mT
no significant directional differences were observed. Within
site scatter is very low, but between site scatter is much
higher. Between 4 and 8 specimens from each site were
subjected to thermal or AF demagnetization. During AF
demagnetization there is a rapid loss of in NEM intensity;
normalized decay curves are concave-up with median
destructive field of 5 to 40 mT, typical of multidomain
magnetite (Fig. lb). The NEM is stable to AF
demagnetization; stereographic proyections and orthogonal
plots show esentially a single component of magnetization (a
randomly oriented low-coercivity component is removed
between 0 and 5 mT). Similar results are seen in thermal
demagnetization (Fig. 2). Only minor loss of remanence
occurs between room temperature and 3OO'C (Fig. 2a). Above
580C there is virtually no remanence remaining, and the
specimens are almost completely demagnetized. There are only
minor directional changes observed during thermal
demagnetization. For the most part specimens show a single
normal component of magnetization with NE declination and
shallow to moderate downward inclination or SW declination
with shallow to moderate upward inclination. In the majority
of cases the direction in the NE quadrant show a change in
inclination to a more negative position between 500 and
650C. Although this component is to weak to be precisely
defined, it lies nearer to the PDFD and may be more
representative of the Upper Miocene field direction. In rare
cases both these components are present in the same
specimen. It is interesting to note that in such cases the
400C unblocking temperature is absent (Fig. 2a).
From the demagnetization results the carachteristic
remanence of the Upper Miocene ignimbrites has been
stablished. Three groups are recognized M,, Mn, M3 (Fig. 2c
& 2d). The M, and Mt remanences are caracterized by wide
coercivity (O-100 mT) and blocking temperature (20-580C)
spectra. M, has mean Da24.86" 1=16.56' aos=6.61. Ma has mean
D=207.26", I=-19.37" and aos=17.33. Ms is characterized by
much higher coercivity and has a blocking temperature
spectrum in the range 580-680C. Ms has mean D=16.52", I=
-15.03" and aos=7.64. It should be noticed that remanences
M, and Mn are aproximatly antiparallel whereas remanence MS
,_ .1.._, ._ .._ .,

65

c/
__..-.tti -.___
.
<! n

._ _

I
fig.lb fig. Ic
fig.la

fig. 20 fig.nb fig. 2c


66

lies on the same axis but whith a shallow negative


inclination.

DISCUSSION

The most significant feature of the results is that the


characteristic remanence is anomalous. Northly directed
vectors have downward inclination, whilst southerly directed
vectors have upward inclination. Since the sampling sites
lie 22-S the PDFD and predicted Upper Miocene field
direction showed, have the opposite sign of inclination. For
this reason we believe that the results are best explained
in terms of self-reversal (Neel, 1959). This phenomenon is
well known in dacitic and rhyo-dacitic volcanic rocks like
those described here.
Self-reversal can occur as the result of the presence
of two discrete magnetic constituents, having two different
Curie points (Neel, 1959). During cooling magnetite passes
through its Curie point an is magnetized. At the same time
maghemite is produced as the result of lower temperature
oxidation. At some point the maghemite starts to cristalize
under the influence of the demagnetization field due to the
magnetite. Under these circunstances the maghemite can be
magnetized oposite to the external field.
However, even allowing for self-reversal, the axis of
magnetization sould still coincide with predicted for the
Upper Miocene field direcction of the area. Our prediction
is that this could not be substantially different from the
from the PDFD by 10" . Our results are thus consistent with
tectonic rotation after self-reversal. In fact the spread of
isolated components between the PDFD and their -present
position suggests that self-reversal may have been
contemporaneous with tectonic rotation.

REFERENCES

HOLLINWORTH & RUTLAND, R W R (1968). "Post Cretaceous


Evolution of the San Bartolo Area, Northern Chile".
Geol. Jo=., 6, 49-62.
FLINT, S (1985). "Aluvial Fan and Playa Sedimentation in
an Andean Closed Basin: The Paciencia Group (Mid.
Tertiary), Antofagasta Province, Chile .'IJour. a
Eeol. Sot. Lnr&~, 141, 533-546.
JOLLEY, L et al. (1990). "Sedimentological Response of an
Alluvial System to Neogene Thrust Tectonics, Atacama
desert, Northern Chile" (in press).
NEEZ, L (1951). "L'inversion de l'aimantacion permanent
des roches." Ann. GeoDhvs.,7, 90-102.
.._^^._.LX _.
^), .

67

PALEOMAGNETIC STUDY OF UPPER CRETACEOUS AND TERTIARY


FORMATIONS FROM THE CENTRAL ANDES
(Southern Ecuador and Peru)

C. Kjssel, m*, P. Mitouard, 0. Macedo, P. Roperch and J. Surmon;

lCentre des Faibles Radioactivites, Laboratoire mixte CNRBAXA, Avenuede la Terrasse, 91198 Gif-sur-Yvette Cedex,
France.

l * ORSTOM, 70 Route dAulnay, 93140 Bondy, France.

RIaumi

Le long de la cordillere andine: environ 2 000 6chantillons provenant de 195 sites pal6omagn6tiquea dPge cr&ac4
sup&ieur et c6nozoique. reparks entre le bud de IEquateur et le sud du Perou, ont et6 prelevas. Les don&es
obtenues montrent la presence de rotations horaires au nord de la deflexion da Huancabamba, et antihoraires au cud.
Ces rotations sont en partie r6centes (postoligocene inferieur). La remarquabte coherence de la rotation antihoraire au
Perou plaide en faveur de IhypoMse dune rotation dansemble de la cordillere occidentale.

Kay words : Pateomagnetism, Central Andes, Cretaceoua, Cenozoic.

Abstract

The Central Andes have been considered as a typical marginal orogen. related exclusively to the subduction of the
Pacific plate. They differ from the Northern and Southern Andes by the lack of any ophiolitic suture and by the presence
of some major bends in the cordillera, known as the Huancabamba and Arica deflections. We present new paleomagnetic
results we obtained from upper Cretaceous and Tertiary formations north of the Huancabamba deflection, in southern
Ecuador and northern Peru (Lancones basin), and south of it, in the Cajamarca area, in central and southern Peru, just
north of the Arica deflection.

About 2000 cores from over 195 sites were sampled in these regions, in upper Cretaceous to upper Miocene
sedimentary, volcanic and intrusive formations. The location of the sampled regions is shown in the Figure. North of the
Huancabamba deflection. in the Lancones syndinorfum. Albian to Benonian vokanics (9 sites) and Paleogene intrusive
bodies (11 sites) were sampled. South of the Huancabamba deflection, in the Cajamarca area, 42 sites were sampled in
the black limestones of the Middle Albfn Parfatambo formation, 12 sites in the slightly deformed Paleocene-Lower
Eocene volcanic Llama formation, 7 sites in the overlying undeformed Upper Eocene-Early Oligocene Huambos volcanic
formation, and 5 sites in some Middle Eocene intrusive granodioritic bodies. Southward, we have sampled 55 sites,
mainly in Albian but also in Eocene and Miocene formations, along three E-W transverses (Lima-La Oroya, Mantaro and
Canete valleys), running from the coastal area to the Eastern Cordillera. Those sites are not all studied yet. Finally, the
analysis of 50 additional sites. sampled in southern Peru, is presently in progress.

In most of the studied samples, a stable characteristic component of magnetization is isolated after thennal heating
beyond 200-800C. This stable component is used to calculate a mean direction for each site. In order to interpret our
data in term of deformation, we have referred the mean regional direct&w to the Upper Cretaceoua pole defined by Beck
(1988) for stable South Amerfca, and to a 40 Ma pole that we have calculated averaging Lower Tertiary and Miocene
published poles. The results obtained from each regionare shownontheFigureand the values of R and F (rotation and
latitudiiaf drtft with respect to the stable continent respectively) are reported in the Table.

One can immedfftly notim that the observed rotations are in opposite senses north and south of the Huanrabamba
deflection, independently of the age of the studied formations.
- North of the deflection, the clockwise rotation has occurred partly during the Uppermost Mesozoic-Lowermost
Cenozoic. and partly later than Paleocene. It has been shown (Mourier et al., 1988; Laj et al., 1989) that these rotations
succeed to a previcus one recorded by the pre-Albian basement.
68

paleomagnetic site-mean directions:

*O Normal 00 Cenozoic
*a Reverse A*Cretaceous

LANCONES SYNCLINORIUM \

A DEFLECT10

CAJAYARCA AREA

Accreted ocrrnlc terranes

Coastal 8ro8 of the Hurncsbamba Andes

L.:; y.:i] Integral Andes

LIMA - LA OROYA. YANTARO


81 CANETE VALLEYS
69

- South of the Huancabamba deflection, the results obtained from the Cretaceous formations of the Cajamarca area and
the E-W transverses, consistently show that both regions have undergone an anticlockwise rotation of about 25-2Q
since Upper Cretaceous. This is in agreement with other published results obtained farther East (Kono et al., 1985)
This rotation is only slightly larger than the one obtained from the Tertiary fom-rations of the Cajamarca area. Thus, at but
in this region, most of this anticlockwise rotation has occurred later than Eady0ligocene. Only 7 Tertiary sites cutof 11
have been studied right now along the three E-W transverses. Only three of them show reliable pateomagnetic results
whii are also consistent with the Cretaceous ones.

The results obtained from tertiary formations and reported here thus document an unpredicted pattern of post-
Paleocene clockwise and post-Early Oligocene counterclockwise rotations north and south of the Huancabamba
deflection, respectively (Mitouard et al., 1999). The clockwise rotations north of the Huancabamba deflection are probably
related to the distributed shear regimes documented by geological studies. South of the deflection, on the contrary, no
large-scale strike-slip faulting has been observed. and the remarkable consistency of the results over suoh a large area
indicate that the observed anticlockwise rotations more realistically k-&ate a rotation of the entire Peruvian margin.
This latter rotation, the timing of which needs to be better constrained between Early-Oligocene and Present, could be
related to the uplift of the Central Andean Plateau, in agreement with a recent published model (Isacks, 1988). When
combined with this model, the data obtained here suggest that the amount of shortening in the Central Andes is greater
but of the same order of magnitude than the one documented by geological studies (Sheffels, 19QQ).

Additional data from southern Peru will be discussed and compared to the former results from the Central Andes.

References :
Beck M.E. (1QQQ). -J. S. Am. Earth Sci., 1, 1, pp. 39-52.
Isscks B.L (1939). - J. Geophys. Pea., 93, 84. pp. 3211323f.
Kono M., Hekl, K., and Hamano, Y., (1995) J. Geodyn., 2. 193-209.
LaJ C., Mltouard P., Roperch P., Klssel C., Mourler T. L M6gsrd F. (1999).- Paleomagnetic Rotations
and Continental Deformations. C. KISSEL 6 C. LAI Eds, Kluwer Publ. Co., pp. 489-51 I.
Mltousrd P., Klssel, C., and Laj, C., (19QO) in press, E.P.S.L.
Mourler T., Laj C., Megard F., Roperch P., Mltouard P. & Forfsn Medrano A. (lQ99). - Earth Planet.
Sci. Leg, 88, pp.l82-192.
Sheffela B. M. (1988) unpublished PhD. M.I.T., 165~~.

Albian Tertiary
/ -
r _~
N RfAR FfAF Period N RfAR FfAF

9 +63.5*7 18.3 f 7.6 Wa lo/11 +24.7f 11.6 43f16.6

,.......................... ..,.....,.,..,................................ Huancabamba deflection .. ... .... . .. . .... ... ... .. .. .. .. . .. .. ... ... ..... .... ... ... ..

Cajsrnsrcs area x/42 -28f6 19 f 6.9 Tpa to Early To 20/24 - 19 f 9.7 1.6fl!

E - W Transverses 23144 -2.5 f7.6 15.5 f 8.6 Te-Tm 3111 -13f16.2 8.7f17.2

Table: Results obtained from Cretaceous and Tertiary formations in southern Ecuador and in Peru. The location of the
sampled regions is shown in the figure. R is positive or negative when the rotation is clockwise and anticlockwise
respectively. N: number of sites. Tpa: Paleocene; Te: Eocene; To: Oligocene; Tm: Miocene (16 Ma)
II__.*. I ., . .., -e---s.

71

PALEOMAGNETISM OF JURASSIC VOLCANICS AND CRETACEOUS PLUTONIC


ROCKS FROM COASTAL SOUTHERN PERU.
IMPORTANCE OF ROTATIONS IN THE ARICA DEFLECTION.

vROPEACH AND GABRIEL CARLIER

l
ORSTOM and College of Oceanography, OSU. Cowallis. OR 97331, USA

ORSTOM. lnstitut Frangais de Recherche Scientifique pour le Oeveloppement en Cooperation, 213 rue La Fayette,
75480 Paris, France

Une etude paleomagnetique detaillee (51 sites) des unites volcaniques du jurassique inferieur et de roches intrusives
jurassique a c&ace inferieur de la zone citiere sudperuvienne (de 16% a 18s) a et6 entreprise afin de tester
lhypothese dune deformation oroclinale des Andes centrales. La formation Chocolate (jurassique inf6rieur) a enregistr6
une reaimantation avec une declinaison paleomagnetfque denviron 330 . Laimantation primafre de6 unites vokzaniques
alnsi que Ies resultats sur le batholite dAr6quipa indquent des rotations antihoraires plus importantes denviron 50 8 60.

Key Words: Paleomagnetism. Tectonic rotations, Andes

Introduction

Geological evidences for allochtonous terrane in northern Peru and Western coastal Ecuador have recently been
confirmed by paleomagnetic studies [Mourier et al., 1966: Roperch et al., 19671. On the contrary, the numerous
paleomagnetic studies undertaken in the central Andes have not shown any evidence for the occurrence of allochtonous
terranes [Palmer et al., 1980a.b; Heki et al.. 1984; May and Butler, 19851.However, these studies have suggested some
angular discrepancies between Peru and Chile that led Kono et al. [I9851 to propose a large oroclinal bending of the
Andes. This interpretation has recently been challenged by Beck [1986]. who suggests in situ block rotations as a result
of different directions of convergence between the oceanic plate and the overriding continental crust. On the other hand,
Isacks [I9883 suggests that the high elevation of modern Central Andes and the large width of the altiplano as well as the
thickness of the crust is a consequence of crustal shortening. A model allowing various amount of shortening with a
maximum in the central Andes of between 250 and 425 km indicates the possibility of relative forearc rotation of lS to 25
between Peru and Chile [Isacks. 19881.
We have undertaken a paleomagnetic study of Jurassic volcanism and lower Cretaceous intrusions in coastal southern
Peru where the strike of the andean structures may suggest the most anticlockwise rotation in the hypothesis of a strong
oroclinal bending in the Andes.

Geology and Paleomagnrtlc sampling (Flgurr 1)

The Southern Coastal Peru area, between 14% and 18S, is limited to the east by the over-thrustsystem of the Chincha-
Lluta (Vicente et al., 1979, Vicente, 19851. The outcrops of the Precambrian Arequipa massif are confined west of this
limit. The Chocolate volcanic formation is unconformably deposited on the Precambrian basement. The age of this
formation is not clearly established but it overlies the lower Permian Machani formation [Salinas. 198;l and a lower-
Sinemurian faunal association is recorded at the top of the formation [Vargas. 1970; Vicente. 1981, Vicente et al.. 19821.
Therefore, the age of the Chocolate lormation is limited to the Upper Permian-Hettangian interval. Near Chala. Ilo and La
Yarada. the base of this thick formation (>3000m) begins with conglomerate beds with intercalated acid volcanic horizons.
The upper part is typically caracterized by andesitic volcanic breccias with interbedded marine and continental
sedimentary sequences and basaltic to andesitic flows. At La Yarada. the Chocolate unit is overlain by the Guaneros
Formation, a shallow marine serie that represents eastward lateral variation of the Bajocian Socosani Formation. The
Bathonian to tithonic Yura Group [Jenks. 1948; Benavides. 1962; Vargas. 1970; Vicente et al., 1981; Vicente. 19821are
not recorded in the Southern Coastal Peruvian area and the Jurassic formations are unconlormably overlain by the
.__._.._ _.^ .._,

72

volcanic Toquepala formation of Paleocene


age (Laighlin et al., 1968: Belton and
Lefevre. 1976; Eslrada. 1975). Near
Ccona. where the Jurassic sequence is not
deposited or not preserved, localised
granodiorite plutons are intruded during the
lriasic times (205Ma. Stewart et al., 1974).
Near Ilo, the Chocolate Formation is
intruded by the gabbro-monrotonalite rwks
of the Puma Coles and the Ilo super-units
respectively 196-182 Ma and 150 Ma old
[Mukasa. 1966; McBride, 1977). The main
magmatic aclivity is recorded by Ihe
coastal batholith. In the Arequipa segment,
the early phases of the magmatic activity
are dated at 105 Ma and the last ones at 61
Ma [Mukasa. 19661. Four Super-units are
distinguished: the early gabbros (105101
Ma], the Tiabaya superunits (64.78 Ma.),
the Unga-Yarabamba S-U (70.5-82.1 Ma)
Fig. 1 : simple gsubgic& acdpaleomagnetic sampling map and the Cerro Verde quartz Monronite (81
Ma) (Mukasa. 19863.
_. .. .
1ne pafeomagnetrc samptmg was carnea
out in six different localities hundreds kilometers apart along the southern coast of Peru with 51 paleomagnetic sites and
a total of 470 cores. Near Chala. eleven sites were sampled in the Chocolate Jurassic volcanic formation, one site of red
sandstone interbedded in the vokanics and two sites in a plutonic intrusion a few kilometers north of Chala. Further south,
near the locality of Ccona. two sites were drilled in a plutonic intrusion, possibly of upper triasic age. Several sites were
drilled in different units of the batholith from the Arequipa area. Upper Jurassic and lower cretaceous sediments were also
sampled at four sites north of Arequipa. Five sites of the Chocolate formation were sampled near Ilo on the coast and 3
sites were drilled in nearby intrusive formations. Fourteen sites were collected in the area of La Yarada. west of Tacna.
It was diftttult to recognize if sills and dikes are contemporaneous of the votcanics of the Chocolate formation thal they
intrude.

Paleomrgnetic results

E UP
N

Down #XPEUU)tA (Lu Yarndo)


88PE0305B (Chuln) XSPW106A Wo)
Fig.2 : Examples ol thermal demagnetizafion showing a secondary magnefization befween 150 and 450C.
Open circ/es correspond lo the projecfion of the verlical component, so/id circles correspond lo fhe horizontal
and the shaded area highlights the devialion ol the dectinafion from the present day north.

Because of the existence of hematite whose magnetic properties were clearly recognized during laboratory experiments.
progressive thermal demagnetization proved 10 be most superior in demagnetiring some of the volcanic units than
alternating field techniques and it was generally very efficient in isolating components of magnelizalion with different
unblocking temperatures.

Fig.3: Equal-area projections ot the mean-site direcfions with lheir 95% confidence angle for Ihe secondary
magnelization observed at Chala (a), Jlofb) and La Yarada (c).
-- ,_.-. .-..^ ., _.

73

A magnetic overprint was observed in several volcanic units in Chala. llo and La Yarada and examples of demagnetizatron
diagrams are shown on figure 2. This secondary remanence has unblocking temperature between 150.450 and It is often
the only component of magnetization left in a sample. Mean-site directions are shown in figure 3. Thus magnetrzation was
acquired after deformation of these units because the tilt correction significantly increases the dlsperslon of the data.

A primary component of magnetization with high unblocking temperature above 500C was isolated in volcanics at several
sites near Chala and Ilo. A primary magnetization was also identified in different tntrusives units. Mean-srte results are
shown on figure 4
N Y Y

Fig.4 : Equal-area projections of lhe mean-site directions with their 95% contidence angle for the pnmary magnetisation
observed at Chata (a), Ocona-Arequipa(b) and /to (c).

Both normal and reversed polarities are found and are almost antipodal; this constitutes a positive reversal test which
demonstrates that these paleomagnetic directions do not correspond to anomalous magnetic field behavior. No tilt
correction was applied to the intrusive rocks. Large counterclockwise deviations of the declination is observed at Chala
and in intrusives rocks near Ocona and Areouioa. On the contrarv. no deviation from the oresent dav oeoaraohrc axis is
observed at Ho. but because the secondary magnetiration ha&a westerly declination: this is an &id&de for local
clockwise rotation prior to the acquisition of the secondary magnetization.

Discussion

Relative tectonic rotations of the central Andes with respect to stable South America (SA) can only be assessed by
comparing the observed paleomagnetic direction with the one expected for SA. To do this, the knowledge of the age of the
magnetization and an accurate paleomagnetic apparent polar wander path (APWP) for SA are necessary. The APWP of SA
is poorly defined but it has been within 15 around the present geographic axis since 200Ma. The problem of an accurate
APWP has often been neglected in interpretations of previous paleomagnetic results. Clockwise or counterclockwise
rotations of 10-15 cannot be really demonstrated as long as the uncertainties in the APWP are not reduced. Only direct
comparison of paleomagnetic results from rocks of the same age north and south of the Arica hinge may improve the
accuracy of relative rotation models between Peru and Chile.
The secondary magnetization observed in the Chocolate volcanics is not easily dated. The major thermal event
corresponds to the intrusion of the batholith and the last event of deformation is the Chincha-Lluta overthrust of santonian
age [Vicente. 1995) Thus it seems reasonable to give an upper cretaceous age to this remagnetization. The mean
declinations vary from 325 to 331 which leads to a Conservative estimate of counterclockwise rotations of about 20. This
rotation is consistent with a large shortening in the central Andes.
The primary magnetization indicates that about 50 to 60 counterclockwise rotation occurred from Chala to Arequipa.
Since only about 20 is recorded by the post remagnetization. this means that 30 to 40 counterclockwise rotation had
occurred in cretaceous time, possibly at the time of the Chincha-Lluta overthrust. On the other hand, results from llo
suggest local clockwise rotation.
We are starting a paleomagnetic survey of Bolivia and we hope that additional paleomagnetic data from the inner and
eastern part of the chain will provide a better understanding of the pattern of tectonic rotations in the Andes.
75

NEOTECTONIQUE
77

ANDEAN DISPLACEMENT AND STRAIN PARTITIONING


OF THE NAZCA-SOUTH AMERICA SLIP VECTOR DURING THE LAST 5 MA.

DEWEY, J. F. and LAMB, S. H.

Department of Earth Sciences, University of Oxford, Parks Road, Oxford OX1 3PR
(tel: 0865 272021; fax: 0865 372072)

From 75 mma-l at 6N to 87 mma-l at 47S of Nazca-South America plate


convergence is presently absorbed in the Andes. Along any Andean traverse, the sum of
relative velocities between points must equal the relative plate motion. We have developed a
kinematic synthesis of displacement and strain partitioning in the Andes from 47s to 5N
relevant for the last 5 ma. based upon:

1) relative plate motion deduced from oceanic circuits giving a roughly constant
azimuth between 075 and 080,
2) moment tensor solutions for earlhquaqes since 1980,
3) structural studies of deformed Plio-Pleistocene rocks,
4) topographic/geomorphic studies,
5) paleomagnetic data,
6) geodetic data.

We recognize four neotectonic zones some with subzones and boundary transfer zones
that solve compatibility problems between zones that are partitioned in different ways. These
zones are not coincident with the classic zones defined by the presence or absence of a
volcanic chain and the differences in finite displacements and strains and tectonic form and it
is clear, therefore, that the long term segmentation and finite evolution of the Andes does not
occur In constantly-defined segments In space and time.

In segment 1 (47-39 S), the slip vector is partitioned into roughly orthogonal
Benioff Zone slip with large magnitude/large slip-surface earthquakes and both distributed
dextrai shear giving clockwise rotations of up to 50 and dextral slip on the curved
Liquine-Ofqui Fault giving 5-10 of anticlockwise rotation.

In segment 2 (39-200S), the slip vector Is partitioned into Benioff Zone slip
roughly parallel with the slip vector, Andean crustai shortening and a minor component of
dextrai slip, mainly on the Atacama Fault System. Between 39 and 34S, a cross-strike
dextral transfer zone that deflects the Chile Trench absorbs the shortening contrast between
segments 1 and 2.

In segment 3 (20-6S), the slip vector is partitioned into roughly orthogonal


Benioff Zone slip, crustai shortening, sinistrai trench-parallel faulting and north south
extension. Compatibility between segments 2 and 3 Is maintained by the sinistral
ESE-trending Cochabamba shear zone and N-trending dextral faults.

In segment 4 (SS to 5ON), the slip vector is partitioned into roughly orthogonal
Benioff Zone slip and dextral strike-slip faulting in the fore-arc and volcanic chain.
.. . _._ ,..,._ ., ._.__

79

CHANGES IN THE TECTONIC REGIME ABOVE A SUBDUCTION ZONE OF ANDEAN TYPE:


THE ANDES OF PERU AND BOLIVIA DURING THE PLIOCENE-PLEISTOCENE

Jacques Louis MERCIER l, Michel SEBRIER l, Main LAVENU l*, Juste CABRERA l, l**, Olivier BELLIER l,
Jean-Fran9oieDLUONT l * and Jo06 HAWARE l, l **

l UA (CURS) G&physique et GBodynamique Interne, Universite de Perie-Sud,


Bit. 509, 91 CO5 ORSAy, Frence

** ORSTW, 213 rue Lefayette, 75 010 PARIS, France

+** Inatituto Geofisico de1 Per& Apartado 3747, LIMA 100, Perti

RCsun6

Line4th detaillee de la cin&sstiqw des failles pliocku-quaternaires des Andes du Perou et de


Bolivie mcntre we succession de trois (tats da contrainte. Cette succession se traduit dans Les
Nauter Andes par (1) &me extension N-S au Puaternsire moyen & Recent, (2) uy canpreasion E-U ou N-S
au Guatcrnaire ancien , (5) we extension E-U ou NE-SU au Pliocene. Cc8 variations du regima
tectoniqw sont mires en relation l vec de8 variations dss forces lux Limiter dues probablemsnt & UI
changement de g&aaGtrie da la pique plongeante.

Key words : xndea, faulting, stress patterns, Pliocene, GuaterNry

Introduction

The changes in the tectonic regima in the Peruvian snd Bolivian Andes, analysed in this psper, mainly
concern the period rlt#eqJent to the upper Fiiocena. At that time, the Nigh Andes had almost
reached their present-day elevation above sea-level. The stress patterns are de&cad essentially
from a field study of fault kinemstics and a nuserical inversion of the slip vector data maasured on
the fault planes.

Geological data

The Cuzco fault systam in southern Peru is chosen as an lxurple to itlustrete the used
methodology. In this region, rtriationa on both active and Nolocene faults l re in lgreemant
with . N-S tension. But faults affecting Louer Pleistocene deposits exhibit tw flnritiesof
striaticna. The mer results from the previous N-S tension, the older, involving reverse
motions, result fraa either an E-U, or e N-S carpression. Faults affecting Pliocene forsmticns,
often shw an oldest family of striations resulting from a NE-SU or an E-U trending tension. Thus,
three tectonic regimas are demcnstrated tiich are also supported by regional uvonformities and
sedimento~ogicaldata: (1) a Pliocene extensional regima , (2) a Louer Pleistocene corrpressioM1
regime (3) a Mid. Pteistocene-Present day extensional regima. Similar analyses conducted in the
- ._Il._^r..._ _ .-. . .__

80

Pacific nd rub-Al&an lcular& allow to sketch the succ*ssive Pliocene-auatrrnerystr*ss


ptterns in the Centre1 Andes. The Guaternary end Present-day stress pettern ir characterired by a
Y-S tension in the high Andes end in the Pacific \oulands and by an E-U compression in the
s&-Andean loulende and at the contact betueen the ~azce and South Aoerican plates iFig. 1 ). This
stress pattern is interpreted at a large uavelength f> 100 km) as en effect of canpmrated
topography. this E&L, uppores that the vertical Lithospheric stressgzz incrreses with the
tcpography the crustal thickness end the Low density untlo bsneath and that the Lithosphericstaxi~
horizontal rtrem VHmx trending E-U rcughly perellel to the convergence, is fairly constant. On both
edges of thr Andes, tectonic being conpressionel, ULL is 63 and ullsmxis 01. fn the High Andes,Vzz
becaaesQ(, then the E-U trendingUHsux is VZ and~iimin trending Y-S is 43, lllouing extension to
occur in this direction (Fig. Q). The Pliocsne stress pettrrn uas cherecterizedby a YE-SW or an E-U
trending tensicn in thr nigh Andes, in the Pacific Lowlands end possibly in the sub-Andean
lwlndr trio. 81).This stress pattern uaa clearly differsnt from the Present-day one because the E-U
trending rtross uas OhsUn. This rqired a ueak push or eventually tractional boundary forces acting
onto thr South American Lithosphere. This right result frm a strong slab pull due to l long steep
dipping rtab&tich &creased thr value of theVu stress transmitted to the overriding plate (Fig. CA,
8). lho Lower Pleistocene state of stress ues cqressional (Fig. 2A, 8). As the elevation of the
Andes had not-wrkedly decreased CrinG this period, this rer@red an i~~eaee of the E-U trerding
stress value. This resultsd frar l strong capling betueen the two Lithosphere8posoibly due to a
rupture of l long slab wdor itr OM Wight (Fig. AC). Other spatial changeo In the stress pattern
are rotated .to the particular situation of the foreerc, to the r&dustion of the buoyant Nazc8 ridge
end to the dffferent dips of the slab.

Paper pfosantd at 28th Int. Geol. Cong., Ueshington 0-C. USA, July 9-19, 1989

figura captlw

FiS. 1 himipal stress directions deduced from the enalysis of Guaternary and active faults of Peru
end Bolivia. Divergent lrrous, tensional horizontal stress directions (V3) ; convergent lrrous,
carpressionslhorizontal stress directions (0,) ; fillsd circles attached to the arrow, conputed
directione ; open circles, praphicaliy defined directions ; balloons, focal mechanism of earthqakes
; lottrrs +ifer to authors, Ab : Abe(19721, St : Stauder (197S), Sp : Deuey and Spence (1929). Pe :
Pemington (19811, Su : Suerez (1982) and Suarer 8t al. (1983), C : Chim and Isacks (1983). Mechanirm
CNG is frca Grange et al. (19&b). CcepressionaLstress directions F.A and S.A are obtainwl frm the
inversion of theso teleseiric focal Iwehanisms and LLU from the inversion of s~icroseisnicfocal
swheniu (Grange et at., 1984b) ; larpe lrrou, direction of the Wazca-South Amarica plate
convergme Wnster end Jordan, 1978).

Fig. 2 Principal conprerricnalstress directicns deduced froo kinaoaties of reverse faults of Low
Pleistocene age. A, E-U trending.cospressioneIdirections. B, N-S trending scepressioneldirections.

Fig. 3 Principal tenlonal stress directions de&ced fran kinsmsticr of noml faults of Late MioCeM
- PLiocene age.

Fig. 4 Guslitative scenario to interpret the changes in the Andean state of stresr during the P~~OC*IM
- Pleistocm. VNA and VSW, velocities of the Nazca and South Aoerica plates respectively ; VT.
velocity of the wstuard rtab retrest. lhir rcenarfo is hypothesized for the evolution of the Andes of
southern Peru tiring thr PLiocene-Guaternaryperiod.
...._.l_r.___.r ., _

81

ir

c
IL

- I

-, L

I_
2
.I . ,.._.. __. ..,_ .
. ,_..I
l.l.- _I._
_ .,I * ,.l^,.

a3

NEO-TECTONICS WITHIN TWi CONTINENTAL FOR&-ARC OF NORTHERN CHILE

T.S.Buddin,I.G.Stimpson, G. D.Villlams

Departmnt of Geology, University of Keele. Keele. Staffs. .England Sr5 SBG

The continental fore-arc of northern Chile exhibits fuxtsposition oi

tectonic styles over a relstively narrow stretch of the continental lithosphere.

Detailed fieldwork reveals that the style of deformation varies across the

fore-arc from a broadly extensional Coastal Cordillera with fault-alluvial fan

interactions and wall developed beach terraces indicating an emergent coastline

(confirmed by tide gauge records), to active fore and back-thrusting in the

Sierra do Moreno which lies 135km west of the Pacific and 15km from the active

volcanic arc. Between these areas of active tectonism is a zone of quiescence

some 80-100km wide containing wholly undeformed Tertiary and Quaternary

successions. This is the Central Depression and may well represent an area of

non-emergence of a linked fault system whose activity or otherwise 1s determined

by the rate, angle and obliquity of subduction of the Nazca plate, and whose

emergence at the surface Is controlled by the pro-existing structural framework.

This presentation attempts to show, vla analysis of neo-tectonic structural

expression (aided by detailed aerial photography and Landsat imagery), and using

earthquake focal mechanisms derived using the Pearce Relative Amplitude Moment-

Tensor Program, how the fore-arc Is deforming in the area of interest.

The controls being exerted on the accumulation of clastlc sediments in an

extremely arid environment by the emergent coastline and active extensional

regime at the coast are clearly recognisable with fault scarps cutting the

Quaternary to Recent alluvium on the eastern side of the coastal range. Multiple

beach terraces are developed and fault controlled alluvial fans accumulate on

the Pacific coast.


a5

NEOfECTONlC CONTROLS ON FAN-DELTA SEDIMENTATION, COASTAL


NORTHERN CHILE: A RESPONSE TO ASEISMIC RIDGE SUBDUCTION

STFPHFN FI It&, PETER TURNER and ELIZABETH JOLLEY

l Departmentof Earth Sciences, University of Liverpool, P.O. Box 147, Liverpool, U.K.

School of Earth Sciences, University of Birmingham, P.O. Box 363, Birmingham 815, U.K.

Resumen

La estructura, el desanollo geomorfol&$zoy la respuestasedimentadadel area costera entre


Antofagastaen al sur y Arlca en al none sido investigadospara evaluar la influencia de la
subsldenda de la placa de Nazca sobree bs cambiis del nivel del mar durante el cuaternarioy al
reciente. Los datos obtenidos sugieren que las vadacio?tesregio/nalesde patrones del
leventamiento dei terreno b large de la costa none de Chile estan controlados por la sub&x&n
de una dorsal asef&mtca,la cual producefluctuacimeseustaticasdel nivel del mar.

lntroductlon

In the last decade the Interdisciplinary science of basin analyds has produced advances in our
understanding of the role of base level changes in controlling the stratigraphy and architecture
of basin-fill successions. Both global eustacy and regional tectonic activity are known to have
major affects on basin-fill characteristicsand facies architecture.However, reliable distinction
between the relative importance of tectonics and eustacy in controlling base level and thus
architecture has not generally been made. In this paper we investigate the structure, iandfon
development and sedimentarydynamics of the north Chilean Pacific margin to evaluate the
influence of Nazca plate subduction processes on relative sea level changes over Cuatemary to
Recent times.

Reglonal Geology

The coastline of northern Chile Is dominated by the Coastal Cordillera, a longitudinally continuous
mountain range, over 2000 m high in places (Fig. 1). This range comprisesJurassic volcanic rocks
and minor intercalated sedimentsof the La Negra Formation and has been traditionally
Interpreted as maklng the position of the Jurassic Andean volcantc arc (Coira et. al., 1982). The
La Negra Formationis overlain unconformably,focally by early Cretaceousbasin-fills but mainly by
transgressive marine sandstones and coquina limestones of the Miocene-Pliocene La Portada
Formation(Fern& & Dlbase, 1976).The La Portada Formationis in turn overlain by Quaternary
alluvial farVfandelta deposits.
Morphologically the coastline consists of a series of bays, headlands and distinctive
terrace&Wave-cutplatforms. In this paper we describe the dynamic evolution of the Quaternary
fan systems and integrate the fiekf evidence with published marine geophysical and
oceanographic data to elucidate the Quaternary-Recentrelative sea level history of the Chilean
86

margin and possible dtivlng mechanisms. A major feature in this region is the arldity of the climate.
In the study area only one river, the Rio Loa, flows tthroogh the Coastal Cordiliera into the Pacific
ocean (Flg. 1).

Sedlmentology of the coastal fan/fan-delta systems

We summarise here the first detailed sedimentological studies of these sequences, fully
described by Flint et al. (1989). The fans can conveniently be divided into three main groups on
the basis of catchment size:
(I) Internal fans which have a catchment area comprising
. _ the immediate watershed of the Coastal
i2&Sllera;
(2) External fans, which are much larger and have access to external drainage from within and
occasionally right through the Cordillera (e.g. the Rio Loa fan; Fig. I). Sixty percent of the feeder
canyons (n40) follow the trace of steep, east-west trending normal faults which cut the Coastal
Cordllleti.
(3) Slde cones, which are small, steep scree cones with no true feeder canyon dominating the
supply of sediment to the fan.
Out of 52 maln locations studled between the mouth of the Rio Loa southwards to the
Mejilbnes peninsula (Fig. I) a total of 43 external and 79 internal fan systems were surveyed.
However, many of the external systems are coalesced with subsidiary internal and side cone fans
(Flint et al., 1989).

Chllean coastal dynamics

At Arica (Fig. I) the coastal range is in net extension, characterised by extensional normal faulting
and subsidence, in common with much of the Chilean margin. South of Arlca uplift is recorded by
spectacular canyon cutting, marine terrace development and incision of alluvial fan surfaces; uplift
reaches a maximum south of Iqulque. The northern boundary between the regions in net
subsidence and net uplift is marked by north-facing neotectonic normal fault scams (Mortimer,
1972; Mortimer & Sarle, 1972;).
Recently published tide gauge records for the west coast of South America for the 30
year period between 1940 and 1970 (Aubrey et al., 1988) provide an independent assessment
of uplift at the Andean margin. Plotting of the uplift pattern and tide gauge readings on a map with
Paclflc bathymetric data reveals a good correlation between areas undergoing uplift and the
intersection of aselsmlc ridges with the South American plate edge (Fig. 2). Thus the uplift
(measurable over a 30 year period) in specific areas of Equador and Peru Is coincident with the
Carnegie and Nazca ridges and the Antofagasta-lquique sector coincides with the ongoing
subduction of an unnamed ridge (Fig. 2). Plate reconstructions indicate that aseismic ridge
subduction has been an Important component of the Tertiary Nazca plate history (Cross 8 Pilger,
1982; Pilger, 1984) and several authors have suggested that subduction of such ridges may, on a
large scale and over perbcis of several million years, affect arc volcanism and structural evolution
(Nur & Ben Avraham, 1981; Cross & Pilger, op cit.).
The role of aseismlc ridge subduction in producing recent and ongofng local
distributed uplift was first suggested by Aubrey et al. (op cit.). Our data further support the thesis
that ridge irregularities or a series of seamounts/ridges may have been responsible for ddving the
high frequency pattern of relative sea level changes throughout the Late Teniary/Quaternary in
northern Chile.
The subsidence and extensional tectonic regime in inter-ridge areas such as Arlca (Fig. 1)
Is well documented (Katz, 1971), despite continuous subduction at the Andean margin since
Jurassic times (Colra et al., 1982). This forearc extension may be a result of subduction roll-back
(Hartley et al., 1988), typical for the Yordilleran margin type of Aubouin (I989) and is marked in
the study area by oblique NE-SW and N-S trending normal faults in the Coastal Cordillera (Fig. I)
and many neotectonic fault scarps in the Central Depression.
~-Ylh- ! Lm I
Figure 2: Relative land levels (Aubrey et. al.. 1988; + values = land rising.
Figure 1: The Andean margin of northern Chile showing main morphotectonic
divisions (inset) & locations of tie studied fan S~S~IIIS (numbered). Note the - vahtes = subsidence, in mm&r) show that the Andun margin is currendy in
exlension/subsidencc except for uw coincident with ridges. where rctive
intense pattern of exlensional normal faulling in the Coascrl Cordilka.
uplift i recorded ova. ravnt 30 year period.
w----m

88

Concluslon

Cur data suggest that regionally variable patterns of Quaternary coastal uplift along the north
Chilean coast are controlled by the subduction of an aseismic ridge, which overprints the effects
of eustatic sea level fluctuations. Thus, subduction of oceanic plate heterogeneities may provide
a mechanism for producing cyclidty in sedimentary sequences at a frequency equal to or higher
than glade-eustacy in fore-arc and possibly back-arc sedimentary basins. These sequences will
be neither of global extent nor global synchroneity.

References

AUBOUIN, J. (1989) Some aspects of the tectonics of subduction zones. In. Subducfion zones:
the Kaiko project. (ed. by J.P. Cadet & S. Uteda) Tectonophysics 180,1-21.

AUBREY, D.G., EMERY, K.O. 8 UCHUPI, E. (1988) Changing coastal levels of South America and
the Caribbean regfon from tide-gauge records. Tectonophyskx 154,289-284.

COIRA, B., DAVIDSON, J., MPODOZIS, C. & RAfvtOS, V. (1982) Tectonic and magmatic evolution
of the Andes of nolthem Argentina and Chile. Earth. Sci. Rev. 18,303332.

CROSS, T. A. & PILGER, R.H. Jr. (1982) Control of subduction geometry, location of magmatic
arcs and tectonics of arcs and back-arc regions. Bull. Geol. Sot. Am. 93,545582.

FERRIAS, F. & DIBASE, F. (1978) Ho]a Antofagasta. Carta Geo/ogia de Chile, 1:250,000. Inst.
Inv. Geol., Santiago. 142~~.

FLINT, S., TURNER, P. and JOLLEY, E.J. 1989. High frequency cyclicity in Quaternary fan-delta
deposits of the Andean forearc: Relative sea level changes as a response to aseismic ridge
subduction.In.: Macxlonald, D., (ed.), Sea level changes at active plate margins. IAS Spec. Publ.
(in press).

HARTLEY, A.J., TURNER, P., WILLIAMS, G.D. & Flint, S. (1988) Palaeomagnetism of the
Cordillera de la Costa, northern Chile: Evidence for local forearc rotation. 89,375-388. Earth
Planet. SC/, Lett. 89, 375-388.

KATZ, H.R. (1971) Continental margin in Chile - is tectonic style compressional or extensional?
Bull. Am assoc. Petrol. Geol. 55, 1753-1758.

MORTIMER, C. (1972) The evolution of the continental margin of northern Chile. Proc. 24th Int.
Geol. Congr, Montreal, Section 8,48-52.

MORTtMER, C, & SARIE, N. (1972) Landfonn evolution in the coastal region of Tarapaca
Province, Chile. Rev. Geol. Geog. Dynam. 21, 182-l 70.

NUR, A & BEN-AVRAHAM, Z. (1981) Volcanic gaps and the consumption of aseismic ridges in
South America. Geol. Sot. Am. Mem. 154, 729-740.

PILGER, R.H. Jr. (1984) Cenozoic plate kinematics, subduction and magmatism: South AfTWiCan
Andes. JI. Geol. Sot. Lond. 141, 793-802.
89

Subduction of aseismic ridaes at the Andean marain: a


maior factor in the sedimentoloalcal and structural
evolution of forearc basins

S. Flint,l, E.J. Jolley2 s P. Turnerz, G.D. Williams3 6 T. Budding

1 Department of Earth Sciences, University of Liverpool, P.O. Box 147,


Liverpool L69 3BX, U.K.
3 School of Earth Sciences, University of Birmingham, P.O. Box 363,
Birmingham, U.K.
3 Department of Geology, University of Keele, Keele, Staffordshire, U.K.

The coast of northern Chi!e comprises Mesozoic magmatic rocks and


Cenozoic - Recent shallow marine and alluvial fan/fan-delta sediments.
The structure, landform development and sedimentary response of the
coast between Antofagasta in the south and Arica (600 km to the north)
have been investigated to evaluate the influence of Nazca plate
subduction on sea level changes over Quaternary to Recent times. At
Arica the coastal range is in net extension, characterised by extensional
normal faulting and subsidence, in common with much of Chile.
South of Arica uplift is recorded by marine terrace development and
incision of alluvial fan surfaces; uplift reaches. a maximum south of
Iquique. The boundary between regions in net subsidence and net uplift
is marked by north-facing neotectonic normal fault scarps. Variations in
apparent uplift and subsidence are consistent with recently published
oceanographic records on relative sea level changes over a 30 year
period. Our data suggest that these regionally variable patterns of
coastal uplift along the north Chilean coast are controlled by the
subduction of an aseismic ridge, which overprints the effect of eustatic
sea level fluctuations.
Subduction of oceanic plate heterogeneities may provide a mechanism
for producing cyclicity in sedimentary sequences at a frequency equal to
or higher than glacio-eustacy in fore-arc and possibly back-arc
sedimentary basins. These sequences will be neither of global extent nor
of global synchroneity.
91

RECENT VERTICAL MOTIONS AND THE SUBDUCTION OF THE NAZCA RIDGE,


CENTRAL COAST OF PERU

Jose MACHARE * & Luc ORTLIEB **

* Instituto Geofisico de1 Peru, Aptdo. 3747, Lima 100, Pdrou

*a Mission ORSTOM, Aptdo. 18-1209, Lima 18, PQrou

Resumen

La deformaci6n causada por la subduccibn de la Dorsal de


Nazca se manifiesta por el levantamiento cuaternario de la
zona costera entre Pisco y Lomas, que llega a 1000 m. Los
modelos fisioos de deformaci6n podrian ser majorados mediante
la inclusi6n de nuevos datos cinematicos y geocronol6gicos.

Key words: Neotectonics, Aseismic ridges, Marine terraces,


Coastal Peru.

Background

A decrease of seismic activity, inhibition of arc vol-


canism and coastal geomorphic modifications, are reported
among the main effects of the subduction of aseismic ridges
(Vogt et al., 1976).
The Nazca Ridge is a major, aseiemic and volcanic,
topographic high that is being subducted beneath the Andes of
south-central Peru, between 14O and 16* S latitude. Several
tectonic and magmatic effects of this subduction have already
been recognized; these include the cessation of the Cenozoic
arc volcanism to the north and a lowering of the seismicity
level (Barazangi & Isacks, 1976; MIgard & Philip, 1977; Cross
& Pilger, 19821, and a strong coastal uplift as revealed by
long sequences of Plio-Quaternary shorelines (Teves, 1975;
Machard & Huaman, 1982; Hsu, 1988). Some preliminary model-
ling of the neotectonic impact of the subduction of the ridge
on the coastal region has been attempted (Moretti, 1982; Hsu,
1988) but still require improvements before a clear under-
92

standing of the regional geodynamic evolution is obtained.

The coastal uplift: a consequence of the Nasca Ridge subduc-


tion

Between S'S and 16'S, the Peruvian forearc region has


evolved under marine conditions and stayed below sea level in
Cenozoic times. Meanwhile, south of 16'S, the inland part of
the Peruvian margin (particularly the Coastal Cordillera and
Moquegua basin) remained emerged during most of the Tertiary
and the Quaternary (Machare et al., 1986). Therefore, the
recent uplift observed in the coastal segment between 14*s
(Pisco) and 16' (Lomas) may be considered as abnormal, and
should be interpreted as closely linked to the subduction of
the Nazca Ridge. Actually, it can be established that the
uplift is spatially coincident with and genetically related
to the Nazca Ridge.

Modelling of the deformation

To date, two models have been proposed to explain the


coastal uplift as an effect of the subduction of the Nazca
Ridge. The first one is a simple geometric model based on the
"similarity" between the shape of the Nasca Ridge above the
ocean floor and the topographic dome formed by the coastal
region (Hsu, 1988). This model did not take into account any
physical parameter constraining the basic rheologic behavior
of the overriding plate.
Earlier, Moretti (1982) had tested a static model in
which the Nasca Ridge is considered as a density heterogenei-
ty which produces a vertical upwards stress field of 0.35 kb
on a 200 x 400 km rectangular-shaped surface below an elastic
homogeneous continental plate. The best adjustment of
Moretti's model predicts an oval dome-shaped uplift extending
over 500 km along the coast which amounts to a maximum ampli-
tude of 1,100 m. The results of this model strongly depend on
two parameters: the ridge width and the flexural rigidity of
the overriding plate. It would be greatly improved by intro-
ducing a visco-elastic component (instead of a perfectly
elastic response), and by taking into consideration the
velooity of the southeastwards migration of the ridge along
the margin.

The relative motion of the Nasca Ridge and the margin

Because of the N080' azimuth of the Andean convergence


(Minster et al. 1974) the Nazca Ridge (mean strike= N040')
is obliquely subducting under the Peruvian margin (mean
strike= N14S0). Thus, it may be established that with a
convergence rate of 80-100 km/My (Pilger 1983) the ridge has
been scanning the margin southeastwards with a mean velocity
of 60 f7 km/My. From these kinematic reconstructions, it is
inferred that the Nazca Ridge began to directly affect the
14'-16'S coastal segment by 4 My ago.
Among other dynamic considerations which have not been
included in previous modelling, it is important to note that
the proJection of the present-day ridge axis onto the conti-
93

nent does not exactly coincide with the area which registered
the highest uplift rates (San Juan de Marcona), as assumed by
Hsu (1988), and that there are evidences that, in the north-
ern part of the coastal segment, some areas which have been
previously uplifted (in late Pliocene-middle Pleistocene) are
now experiencing subsident motions.

Amount and rates of uplift

In order to determine the main parameters of the defor-


mation (shape, rates, timing), it is necessary to reconstruct
the movements experienced along several coastal transects,
and if possible at different time scales. This goal is not
yet entirely achieved, but some new elements are being gath-
ered. The main difficulty is to assess accurate correlations
in time and space between localities referable to a given
horizontal datum (sea-level).
The geometric distribution of the highest marine sur-
f aces eroding the Coastal Cordillera, can be used to depict
the long-term deformation registered by the area in the last
few million years. The problem is that these abrasion sur-
f aces, which are interpreted as remnants of the last marine
invasion, are diachronic. Near the northern extremity of the
Pisco Basin (Paracas), the sea began to regress by middle
Pliocene, while more to the south (Sacaco), the last regres-
sion is considered to have occurred in the early Pleistocene
(Muison 8 DeVries, 1985; Machare, 1987). Though, by trying to
take into consideration these age differences, we propose an
elevation vs. latitude plot of these well-developed surfaces
(figure). This plot displays an asymetrical dome shaped

__,_-----e--w,
*Ye X,
,.d
\.
.!::.
marine surfaces. _e*'

: i
4 i .I_ A
Successive positions;
of the ridge axis and;
_a_f_f_Etedarea on-land
01 ,RO~rn f
.,-.._..-

95

QUATERNARY MARINE TERRACES ON THE PERUVIAN COAST


AND RECENT VERTICAL MOTIONS

Luc ORTLIEB= & Jose MACHARE'=

Mission ORSTOM au Perou, Apartado 18-1209, Lima 18, Perou


*z
Instituto Geofisico de1 Peru, Apartado 3747, Lima 100,
Pdrou

Resume

Un reexamen critique de l'ensemble des temoins laisses


par lee hauts niveaux marins interglaciaires la long de la
cdte peruvienne est necessaire pour reconstituer l'histoire
des mouvements verticaux ayant affect& cette marge active au
tours des deux derniers millions d'annees.

Key words: Neotectonics, Vertical movements, Quaternary


chronostratigraphy, Sea-level, Marine terraces, Coastal Peru.

Introduction

Paleo-sea level studies provide valuable data for the


interpretation of vertical motions experienced by coastal
areas, particularly during the last l-2 M.y. As marine ter-
races are relatively well-preserved in the arid coastal Peru,
they have been observed by many authors, but were seldom
thoroughly studied. Moreover, the discrimination between the
tectonic and "eustatio" factors controlling the present-day
elevation of Quaternary paleo-shorelines has remained a
puzzling problem for many authors (e.g. see discussions in
Tosdal et al., 1984; Laharie, 1985; DeVries, 1986). Thus, it
is time to re-examine the whole Peruvian marine terrace data,
in the light of the recent progress made in the interpreta-
tion of Pleistocene sea level fluctuations and in the chro-
nostratigraphic analysis of paleoshorelines (Ortlieb, 1987,
1990).
Through this review of the main sequences of Quaternary
shorelines, the Peruvian coast will be divided into four seg-
ment: the northern (from Tumbes to Chiclayo), north-central,
96

south-central (from Paracas to Yauca), and southern regions.

The northern coast

The Mancora, Talara, Lobitos and Salina "tablazos" are


wide Quaternary marine terraces extending along the northern
coast of Peru (Bosworth, 1922). Recently, DeVries (1984,
1986, 1988) modified the Plio-Quaternary chronostratigraphy
of the area by redefining the Mancora tablazo as a thin
littoral deposit overlying a new Taime Fm.. The Taime Fm. is
assigned a Pliocene age and the Mancora tablazo deposits are
interpreted to be earliest Pleistocene (DeVries, 1986, 1988).
Besides, this author recognized an additional ("Lower
Mancora") tablazo, intermediate between the Mancora and
Talara terraces.
Morphostratigraphic criteria strongly suggest that the
Lobitos tablazo was formed during the early last interglacial
high sea stand (Isotope substage Se, ca. 120 ka) and not
during the 40 ka or 80 ka high stands as once env&ioned by
DeVries (1986). The Lower Mdncora and Talara terraces were
formed during some of the Middle Pleistocene high sea stands.
A complex pattern of Quaternary vertical deformations
characterized the northwesternmost Peruvian coast, and in-
volved altogether some active faulting, an older local uplift
centered on the Cabo Blanc0 area (max. elevation of the
Mancora tablazo at +300 m), and a younger, more regional,
upli t motion with an estimated mean rate of 0.15-0.20
m/10 4 y. The last mentioned uplift, which was slightly strong-
er in the north (north of Mancora) than in the south (Paita-
Bayovar), is probably linked to some deep-rooted crustal
phenomena associated to the subduction of the Carnegie Ridge.
Five (?) Pleistocene high sea level stands were regis-
tered at about +lO, +50, +70, +90, and +170 m (Sebrier, 1978)
as small staircased abrasion platforms on the Illescas head-
land (Bayovar). These marine terraces are located on a fault-
ed, southwards tilted, block. The correlation between these
terraces and the above-mentioned tablazos is yet pending,
except for the lowest terrace (+lO m/Lobitos).
Emerged Holocene strandplains and beach ridge sequences,
preserved along low-relief coastal areas of the northern
coast ("Salina plains") were probably favored by a slow on-
going regional uplift. Though, such coastal features which
have sometimes been cited as evidence of repeated (coseismic)
uplift motions, are now interpreted as resulting from late
Holocene El NiRo oceanographic anomalies (Ortlieb & Machare,
1989; Ortlieb et al., 1989).

The central coast

From Chiclayo southwards for some 900 km, no emerged


Pleistocene shoreline has been described, although remnants
of Holocene shorelines are commonly found at up to a few
meters above MSL. This lack of Pleistocene marine terraces is
generally considered as an evidence for subsidence (e.g.
Machare et al., 1986). Nevertheless, the numerous emergent
Holocene coastal features along the central Peruvian coast
suggest that the area is not currently experiencing subsident
97

motions (Ortlieb & Machare, 1989). Additional studies along


this coastline should elucidate whether the area has really
been subsiding in the last 2 M.y. and what are the relation-
ships between the distinct' tectonic regimes prevailing in the
continental margin (with its actively subsiding basins) and
the neotectonic comportment of the onshore coastal area.

The south-central coastal segment

The most complete sequence of Plio-Quaternary marine


terraces from South America, which includes more than 20
abrasion platforms, is located near San Juan Marcona (Broggi,
1946; Legault, 1960). In spite of the good preservation
conditions of the landforms and of the fossil remnants, the
age determination of these marine platforms is still actively
debated. For instance, in the series of 13 staircased ter-
races at Cerro da1 Huevo, an isotope substage 5e age has
tentatively been assigned to the +I48 m shoreline (Hsu &
Bloom, 198S), the +90 m shoreline (Machare, 1987), and the
+65 m shoreline (Hsu, 1988; Hsu et al., 1989). The available
geochronological data from San Juan terraces include aminoa-
cid racemizfi ion analyses, as well as ESR (electron spin
resonance), C and Th/U measurements (Hsu, 1988; Hsu et al.,
1989). Hopefully, further geochronological analyses presently
under-process (at Geotop lab., Univ. du Quebec a Montreal, in
collab. with C.Hillaire-Marcel and P.Pichet) will contribute
to unravel several of the remaining uncertainties regarding
the chronology of this exceptional record of late Pliocene to
Holocene sea level oscillations.
According to the chronostratigraphic interpretations of
the lower terraces of the San Juan area, estimations of mean
uplift rates for she last few hundred thousand years vaf;y
between 0.47 m/10 y (Hsu, 1988) and 0.70 (i 0.05) m/10 y
(Machare, 1987). In any case, these are the highest uplift
rates measured on the Peruvian coast, and they are closely
related to the subduction of the aseismic Nazca Ridge
(Machare & Ortlieb, this volume). As the highest marine
terrace of the San Juan sequence (+780 m elevation) is as-
signed an upper Pliocene age (on paleontological evidence,
DeVries in: Machare, 1987), it may be assumed that the maxi-
mum amount of Quaternary uplift has been of the order of 700
m, and thus that the over-all mean uplift rate was of the
order of 0.40 m/103y in the last ca. 1.8 M.y.
At less than 50 km south of San Juan (Sacaco Basin), the
uplift motions have been much weaker, since earliest Pleisto-
cene fossiliferous marine beds are found at a +200 m maximum
elevation (de Muizon & DeVries, 1985). In Yauca-Palpa area, a
set of numerous Pleistocene beach ridges which apparently
document most of the Pleistocene high sea stands, is present-
ly under study.

The southern coast

The southern part of the Peruvian coast (Chala-Tacna)


was relatively favorable to the preservation of uplifted
marine terraces, that can be observed at varying elevations
between +lO and +200 m. Two areas, near Chala and Ilo-Ite,
98

which show series of staircased platforms with associated


fossiliferous coastal sediments, were selected for prelimi-
nary geochronological analyses (U-series and aminostratigra-
phy, in collab. with Geotop lab.).
At Chala, a sequence of eight marine terraces, preserved
between +20 and +300 m, had been interpreted as resulting
from the combination of four Pleistocene transgressive cycles
and five teotonic pulses (Laharie, 1970, 19851, but it is
most probable that each platform was eroded during a distinct
interglacial high sea stand. The first aminostratigraphic
results obtained in Chala area indicate that several coastal
deposits (but not all) that crop out at +25/+30 m correlate
with the last interglacial maximum. Though, the highest
remnants of the substage 5e paleoshoreline may be found at up
to +50 m.
At 110 and Ite, incomplete marine terrace data previous-
ly provided some hazardous interpretations which lead their
authors to underestimate the recent uplift motions: Tosdal et
al. (1984) overestimated the age of the low terraces (errone-
ously attributed to the Early Pleistocene), while Hsu (1988)
inferred from some aminoacid analyses that the deposits of
the last three interglacial maxima were in stratigraphic
superposition. New radiometric and aminostratigraphio results
assess a substage 5e age to the well-developed Pampa de1 Palo
terrace (+20 m) and indicate that several late Middle Pleis-
tocene high sea stands were registered at distinct elevations
up to at least +70 m (Ortlieb et al., in prep.).

References

Bosworth1.0. W22)- Seolow of the Tertiuy and Quaternaryperiods in northestm part of Peru.
RacRillan. London.346 p.
BroooiJ.A. (1946)- Bol. Sac. &l. Peru, 24, 3-24.
DeVriesT. M84)- & @cmRsot Rru. 73, l-13.
DeVriesT. (1986)- The geologyand pehontdogy of tabluos in W Peru. PhD, OhioSW.4lhiv., 1080P.
DeVriesT. MB)- J. 8ouh her. Earth Sci., 1, 121-136.
Hsu J. U$W- Ewgsd Rvkrrwy win tarracesin southernPeru1 sea level chaws and continmtal
mrgin tectcnics over the sbbcting Ram Riti. PhD,Cornell lhiv. 310 p.
HsuJ. k Bloa Aa(1965)- 6rol. Sot. Aw. M&r. Prow., 17 (71, 614.
Hsuet al. M89)- C&km. Sci. I&& 8, 255-262.
Laharie R. 11970)-Ier bngr.Latimmar.Reol. (Lima), 6, 145-157.
L&h&eR.W65)- Bull, Inst. Franc. Et. hdines, 14, 19-47.
tilt R.Z. (l%O)- Rewv study ofruin terraces in the krcona-San Juan area, 9cuthrn Peru.
lhiv. Richi9an W&ssI). 23 p.
Ra&ar4 J. (1987)- La aarpe continmtale C Phur rbgimestuhiques et sddimntaires conOzoiwes dr
l'want-arc &s &I&Scskrsles. mQrrDo&W., lhiv. Paris-B& Orsay. 391 p.
Hachar6J., 84brier R., !&man D. k Rercier J.L. (1986b Bol. !bc, 6eol. Peru, 76, 45-77.
kizon C. ds k DeVriesT. (1985)-Real. Rmdsch., 74, 547-563.
Ortlieb L. U987)- N6otect.h~ et variations du niveau earin au Quaternaire dans la &ion du Rolfe ds
Californie, Rexicn#.Tt&e Do&-k-Sci.,lhiv. Aix-Harseille II, Et.--- k ThbsesmT194#7 p.

Ortlieb L. (HO)- es sdeitted to 6lcbal Planet, Charms.


Ortlieb L. k Radar6 J. MW)- -IdINCUAMwtect Bull 12, 83-84.
CwtliebL., Radw6 J., Fownier II. 6 b&an R. (1989)- Bol. 9oc. Real. Peru, 80, (in press).
Sebrilr R. (1978)-La tecthica reciento da la zma de Bayovar.Cc&rib. Inst.6eofis.PerQ, 78-01. 29 P.
T&l R.H., Clark A.H. k Farrar E. (1984)- 6001. 8oc. Aw. Bull,, 95, 1318-1332.
99

NEOTECTONIC ACTIVITY IN THE NORTHERN SIERRAS PAMPEANAS,


ARGENTINA

Manfred, Arthur L. Bloom**, Daniel Malixia***

*Geologisches Institut, Universit& Karlaruhe, Kaiserstrasse 12, D-7500 Karlsruhe 1

*+ Department of Geological Sciences, Cornell University, Ithaca, N.Y. 14853. U.S.A.

*** UBA-CONICET. Dpto. Ciincias Geol6gicas, Buenos Aires, Argentina

Abstract

The Argentine Sierras Pampeanas (26 to 33s) are reverse fault-bounded mountain blocks of Precambrian to
Paleozoic basement rocks in the foreland of the central Andes. Uplift in the northernmost Sierras Pampeanas
fault blocks of Sierra de Quilmes, Sierra Cumbres Cakhaquies. and Sierra Aconquija started about 7 Ma and
became pronounced between 4 and 3.4 Ma. The movements culminated after 2.9 Ma, when the conformable
Mio/Pliocene Santa Marfa Group was over&rusted. faulted, and folded in the course of principal basement
uplift. These movements created climate and base level conditions that resulted in the formation of five
pediment levels in the deformed basin strata between 2.5 and 0.3 Ma. Tectonically induced base level changes
in the piedmont resulted in three tectonism-related pediment surfaces between 2.5 and 0.6 Ma, which attest to
the neotectonic activity in the Andean foreland. In conjunction with the principal uplift of the adjacent Funa
Plateau the n&themmost Sierras Pampeanas are bestexplained by E-W compressional stresses superposed on
a thiied lithosphere and inherited zones of structural weakness that facilitated uplift.

Introduction

The thick-skinned Sierras Pampeanas (26Oand 33OS)are characterixed by Proterozoic to lower Paleozoic
crystalline mountain blocks that were uplifted along high-angle reverse faults during late Cenozoic time (Fig.
1). Separated from each other by tectonic depressions that are filed with Tertiary and Quatemary sediments,
the sometimes more than 5000-m-high ranges protrude out of tbe low-lying Andean foreland (4, 8). The
ranges define a 450~km-wide belt of Lar&nide-type uplifts and are presently under E-W horizontal
compression (4). Between 28O and 33s the location of the ranges correlates well with a flat subduction
segment (cl@)of the oceanic Naxca plate. In contrast, the north&n Sierras Pampeanas (north of 28%) occur
over a seismic transition zone to a steeper subduction segment, which is located at about 25s. This
transition coincides with a geologic transition into the intra-Andean Puna plateau, an inverted Cmmceous rift
(Santa Barbara System, Sta. B.), the Cordillera Oriental (Co. 0.). and the thin-skinned Sierras Subandinas
(Sa. Sub., Fig. 1). The transition to different teetonic provinces suggests a relationship between the angle of
subduction and deformational styles in the overriding plate, One of tbe prime sites to demonstrate the tectonic
and geologic history of the Sierras Pampeanas with special emphasii on the young deformation exists in the
Santa Marfa Valley in the northern part of the Sierras Pampeanas. Within a distance of 50 km, elevations
change from 400 m in Tucumdn to 4000-5400 m in the basement blocks chat the delimit the approximately
2OC6-m-high Santa Maria Valley (Fig. 2).
100

Geologic setting and stratigraphy of the Santa Maria Valley

The basement blocks that border the Santa Maria Valley are the Sierras Aconquija. C. Calchaqules and
QuIlmes that mainly consist of schist, gneiss. and migmatites (Fig. 2). In the transition to the other
geologic provinces to the north the basement grades into low-grade metamorphic rocks. In the Santa Maria
Valley there ate seven Tertiary units overlying crystalline basement and allow to evaluate the Late Cenozoic
faulting history. The lowest unit is the 11 m.y. old Saladillo Formation (G. Bossi, oral communication), a
fme to medimn-grained sandstone 306 m thick that overlies the Aconquija basement. It possibly correlates
with the 10.6 m.y. old El Morterito Fm. that overlies the crystalline basement of Sierra Quilmes on the
west. In the Santa Marla Valley the sedimentary units are a Mm-Pliocene coarsening upward sequence and
include the San Jod. Las Arcas, Chiquimil, Andalhuala. Corral Quemado. and Yasyamayo formations (Fig.
2). Except for the Yasyamayo Fm, these formations are treated here as the Santa Maria Group (SMG; 1.8).
At the base of the SMG are sand-.and siltstones
of the marine San Jose Fm., which contains
fossil fmh, stromatolites, and foraminifera. It is
conformably overlain by brown to dark red
sandstones of the undated Las Arcas Fm. These
basal units are overlain by medium- to coarse-
grained sandstones and sandy conglomerates of
the approximately 1800-m-thick Chiquimil and
AndalhuaIa formations, which define an age range
between about 7 and 3.4 Ma (8). The strata have
westward transport directions and indicate a
lowland-type low-energy braided river depo-
sitional environment with lithologies related to
Sierra Aconquija and Sierra C. Calchaqules.
These sediments are covered by the up to lOOO-
m-thick Corral Quemado Fm, a medium- to
coarse-grained and blocky conglomerate typical of
alluvial fans. All sections display westward
transport directions and a composition derived
from the basement of Sierra C. Calchaqules and
Sierra Aconquija. The uppermost parts of the
conglomerate contain a dacitic lense with a
fission track age of 2.96 f 0.57 Ma. In the
northern part of the valley this formation is
unconformably overlain by the 298-m-thick
Yasyamayo Fm that consists of alternating
horizons of pelite and gypsum. Other excellent
chronostratigraphic reference horizons for the
evaluation of ncotectonic activity in that region
are the erosional remnants of five formerly
continuous pediments and their conglomeratic
covers that defme an age range between 2.5 and
-0.3 Ma. The first 3 pediments are tectonically
Gcnerakd geologic map of the Sierras Pampeanas; deformed and are 2.5.1.2 and >0.6 m.y. old (8).
Sierras Pampeanas stippled (SaP.); afmr (4) and (8)

Neogene and Quaternary Deformation

The presence of marine strata at the base of the SMG indicates that the region of the northern Sierras
Pampeanas was stiIl a lowland more recently than 11 Ma. Initial uplift began between 11 and 7 Ma, when
the Chiquimil and Andalhuala Formations began to be deposited in front of the uplifting Sierras Aconquija
andC. Calchaquies. The drastic change in sedimentary environments with alluvial-fan deposition between 4
and 3.4 Ma is interpreted with increased tectonic uplift. With pronounced tectonism the influx of the
conglomerates could be interpreted either as a function of increased proximity to an already moving thrust
101

sheet or as the beginning of uplift in the thrust sheet, in either case responsible for uplift of the adjacent
ranges. In the fm al&native the Corral Quemado Formation is a mountain front facies, which appeared at
the locus where strata were preservedbetween 4 and 3.4 Ma. Ihis implies that the Corral Quemado Fm exists
in the sub-surfaceof the migrating basement blocks and that the relief of the ranges cannot be inferred to have
changed significantly through time. In such a scenario the uplift during the initial phase before 4-3.4 Ma was
of greater magnitude. In this case the tectonism would be explicable with asymmetric basement uplift along a
westward migrating thrust front. To date, the lack of exact structural data from the east side of Sierra
Aconquija may support such a view. In contrast, bounding faults at the west and east side, and a partially pre-
servedpeneplain in Siam C. Calchaquies.exclude a westwardmigrating basement block that creetedspatially
similarfacieormdmigMedvrossitsownbesinf~inthefinalstagesoftheSantaMariadepocenter.
In any case, the principal tectonic event that de-
formed the entire Tertiary sequence, and which
must have been asscciated with major uplift in
theranges.islaterandtookplaceafter2.97f0.6
Ma (8) .In the course of these movements Sierra
Aconquijaand Sierm C. Calchaqulesand their ad-
jacent Piedmont regions were primarily affected
by fruits with NNE and NW strikes (Fig. 2).
The range-bounding faults involve throws in ex-
cess of 7000 m and dip as much as 8YE. Folding
is associated with reverse faulting. and in several
locations drape folds developed where basement
faults lose throw. In general, deformation was
characteri& by ESE-WNW oriented shortening
and vertical extension. Uplift of Sierra de
Quilmes was accomplished along NW-SE tren-
ding reverse faults within the range (Fig. 2).
Faults parallel to the trend of bedrock schistosity
h&ate predispositionof the basement to rupture
and uplift along older structures. Evidence for
recurrent tectonic movementspostdating the main
deformationafter 2.9 Ma is abundant in the Santa
Marfa Valley. Faults and folds that were active
during the Plio-Pleistocenehave been reactivated,
with an amplilication of folds and an increase of
dip in strata along faults. The continuation of
tectonic movements is well documented by
faulted and folded pediments in the Piedmont
regions and along the moutain-bounding faults.
For example, pediment II was once a continuous
surface that originated at the C. Calchaqules
mountain front and predates uplift of the asym-
metric northermost portion of the Aconquija
block. Thrusting and uplift along the Aconquija
fault after 1.2 Ma brought the higher-grademeta-
maphic basement rocks into direct contact with
Geologic map of the Santa Marfa Valley, after (8) the pediment (8).

The absence of large faults that cut the extensive pediments IV and V (0.6 - 0.3 Ma) does not imply that
tectonism has ceased 2-ka-old fault scarps in floodplain and alluvial-fan sediments attest to active tectonism,
and earthquakes in 1906 and 1936 with estimated magnitudes of 6 demonstrate the continuation of tectonic
movements to the present day.

Discussion and Conclusions

Range uplifts in the Sierras Pampeanas above the shallow subduction segment show similar chronologic
patterns as in the northrzn part of the province above the transition xone to a steeper segment. Initial uplift in
the Sierras Famatina and Velasco is indicated at about 7 Ma, and uplift of the Sierra Moradain the western
Sierras Pampeanas occurred after 5.7 Ma (6,7.9). Thus, the tectonic history of the late Cenozoic northern
...- --..-__

102

Sierras Pampeanas raises an important question as to whether the ranges are indeed related to processes
associated with the subhorizontal subduction of oceanic lithosphere or if perhaps other factors have caused
their uplift. For the northern Sierras Pampeanas, shallow subduction is incompatible with the existence of
Cenozoic volcanism as far south as 28% In contrast, volcanism in the region between 28O and 33OS
experienced a progressive decline in activity and migrated east beginning in the Miocene because of the
shallowing of the subducting plate segment (5). However. no such clear relationships between shallow
subduction and Sierras Pampeenas-style basement uplifts exist in the Andean foreland above the shallow
amagmatic subduction segment ofBcuador/Peru between 2 and 15S, where fcreland deformation is limited
and appears to be intbrenced by paleogeographic controls. Basement uplifts occur not only over steep but
also over shallow subduction segments (R. Coward, oral communication. 1988; 2). Consequently. plate
geometry does not exclusively determine foreland deformational styles, and certain crustal predispositions
may be mceasmy to cause widespread deformation over a shallow subduction zone. In the light of crustal
properties and distinct mechanical responses to an overall compressive regime uplift of the central Sierras
Pampeanas related to shallow subduction is not at variance with an equal tectonic style within the seismic
transition zone, if the similar crustal inheritance of the region is considered. In the transition zone the
mountain Mocks are structurally and morphologically equivalent to the central Sierras Pampeenas. However,
this situation changes gmdually in the area of the weeker basement and the Cmtaceous rift basins in the Santa
BarbaraSystem and disappears farther north (4). An explanation to understanding the northern Sierras
Pampeanas uplifts in this context is through a model that takes the close spatial position of the Puna.
lithqpheric thinning, and the active volcanism of the region into account, In the seismic transition zone the
cross-sectional topographic area of the Andes, with mean elevations above 3000 m. increases drastically and
signals the southern boundary of the wide Puna/Altiplano plateau. The northern Sierras Pampeanas are
transition&to the plateau, and their uplift may have been linked to the pmcesses that caused the final uplift
of the plateau. Isacks (3) explains the plateau by a combination of distributed structual shortening and tber-
mal~on. resulting from a thinned South American lithosphere, which is caused by a broad astheno-
sphuzc.we&e between a moderately inclined. but not horizontal. subducting plate and the overriding plate. In
such m. due to thinning, the strength of the upper plate is low and susceptible to thrusting and long-
wavelength thermal uplift. Because of the same type of inherited structures. paleogeogmphy, and lithology,
the region of the northern Sierras Pampcanas may thus have reacted to compressive tectonic stresses in a
similar manner as equivalent reeks above the region of shallow subduction between 28 and 33s.

(1) BaPii:G.E., and R.M. Palma, Reconsideraci6n de la estratigrafIa del Valle de Santa Maria. Provincia de
Tucum& Argentina, Actas, Congr. L&want. Gcol., V, pp.l53-171, Asoc. Geol. Argent., Buenos Aims,
Argentina.1982.
(2) Coward, RJ., and A.M. Aleman, Controls on structural styles in the foreland of Ecuador and northern
Peru;Gaof. sot. Am. Abs1r. Pr0g1ams. 19,629, 1987.
(3) Isacks,B.L.. The Altiplano-Puns and the Bolivian Orocline. 1. Geophys. Res., 93,321 l-3231, 1988.
(4) Jordan, T.E., B.L. Isacks. R.W. Allmendinger. J.A. Brewer, V.A. Ramos, and C.J. Ando, Andean
tectonics related to the geometry of subducted Nazca plate, Geol. Sot. Am. Bull ., 94,341.361,1983.
(5) Kay. S.M., V. Maksaev, R. Mocoso, C. Mpodozis. and C. Nasi, Probing the evolving Andcan
Lithosphere: Mid-Late Tertiary Magmatism in Chile (29-30S) over the modem zone of subhorizontal
subduction, 1. Geophys.Res., 92,6173-6189. 1987.
(6) Ramos. V.A., J.H. Reynolds, T.E. Jordan, and K.D. Tabbutt, Time constraints for the uplift of the
Sierras de Toro Negro, Umango and Espinal, western Sierras Pampeanas. Argentina. Geol. SOC.Am.
Absrr. Programs. 20, A61. 1988.
(7) Reynolds, J.H., D. Malizia, K.D. Tabbutt, N.M. Johnson, and J.J. Nickelsen, Chronology of Andean
Neogene tectonic events affecting the Camp0 de Talampa. La Rioja Province, ArgentinaGeol. See. Am.
Abstr. Programs, 20, A380.1988. .
(8) Strecker, M.R.. P. Cerveny, A.L. Bloom, and D. Malizia, Late Cenozoic tectonism and landscape
development in the foreland of the Andes: Northern Sierras Pampeanas (26-28S), Argentina, Tectonics, 8,
s17-534,1989.
(9) Tabbutt, K.D., C.W. Naeser, T.E. Jordan, and PP. Cerveny, Edades nuevas por m&do de trazas de
tlsfon de tobas Mio-Pliocenas en las Sierras Pampeanas y la precordlllem de Argentina, Actss, Congr. Geol.
Argentina, X. 222-224, 1988.
103

GEODYNAMICS OF THE BOLIVIAN ANDES AND FORMATION


OF THE ALLUVIAL GOLD DEPOSITS

GOrard H&rail", Mlchel Fornarl#, Vitallano Miranda"",


Giovanni Vlscara"""

ORSTOM,UR lH,CP 9214, La Paz, BOllVle

""GEOBOL, CP 2729. La Paz, Bollvle


"""1.G.L Facultad de Clenclas Geologicas, UMSA, CPl2198.
La Paz, Bollvle

The alluvial placers result from on the one hand, the


erosion of primary, in rock mlnerallzations. or from the
erosion of older secondary gold concentrations or from a
mlxlng of the two process, and on the other hand from the
concentration of gold and heavy minerals in the sediments
deposited by the co-related erosion.
so their formatlon 1s linked with the buldlng of
reliefs, the time when the erosion reaches the primary
mlnerallzatlons, and the presence of traps which can
retain selectlvly the aurlferous material produced by the
erosion.
In many cases the placers are scattered and of reduced
economic interest. However, peculiar structural and
morphological conditions can favour the trapping of big
volumes of aurlferous sedlments as, for instance, in the
North of Bolivia.
In this area, placers deposlts occur in the both slopes
of the Cordlllera Oriental, c.a. In the Altlplano (SW)
facing slope and the Amazonian (NE) facing slope. However
the conditions of their formation differ.
The fluvlo-glacial placers (1) partlculary those of the
Basin of Ulla-Ulla, near the border with Peru occur in
Quaternary alluvial sediments3 these me'terialsmalnly
consist of tills and out wash conlomerates and contain
gold only if they result directly from the erosion by the
glaciers of primary mlnerallzatlons. In this case the
-. -..,..ml_-_-- ~ -_-. .-_.

104

geomorphologlcal features are more important than the


structural features to explain their genesis.
Near the town of Oruro, in the area of Caracollo, the
small placers which occur, result from the Quaternary
reworking of Cenozoic conglomerates (Formation
Kollpana).The deposition of these conglomerates is
related to the motion toward the SW of a thrust which
corresponds to the South branch of the Coniri Fault, one
of the major thrust plane of the Altiplano. In this case,

SW
NE

0
I-
y,,,,,v --
m

m
z
m

Figure 1 I Strutural evolution and gold-bearing placers


generation on the eastern side of the Andean Cordillera,
North of Bolivia, and its Piedmont

l- Pre middle Ordovician strata. 2- Units subsequent to


Mid-Ordovician. 3- Sediments deposited on the Piedmont
since Lower Miocene. 4- Main primary gold mineralizations
areas. 5- Middle-Miocene to Pliocene gold-bearing
conglomeratei. 6- Rivers with sterile bedload. 7- Rivers
with aurlferous bedload. B- Rivers with low gold content
bedload. 9- Rehandling of the Mio-Pliocene gold-bearing
conglomerates by Quaternary erosion. lo- Imaglnal point
showing the underthrusting of the Brazilian shield. Owing
to uplift of Eastern Cordillera, gold mineralizations
crop out during the Mid- to Upper Miocene. The correlated
sediments are deposited in subsiding basins, in
relationship with the subandean front thrusts, subsequent
to underthrust of the Brazilian shield by the Cordillera.
105

the Formatlon Kollpana is an Intermediary collector of


tectonic origin and the placers formation 1s related to
both structural and morphological features.

In the Amazonian slope of the Cordlllera in Its Piedmont


(flg.1) the gold placers are more widespread. Numerous
deposits correspond to sediments of the recent terraces
and the bed of the rivers3 they result from the erosion
of nearby primary mlnerallzatlons.
This kind of placers occurs in most of the valleys that
go down from the Cordlllera (Pellechuco, Consata,
Tlpuanl, Challana river) but the speclflclty of this area
1s the presence of big placers, such as those of the
Tlpuanl-Maplrl basin or of the syncllne of the Tuichl
river (2,3).

These latter placers result from a long and complex


process, important quantities of sediments ( Cangalll
Formation, Tutumo Formation) were trapped and accumulated
since the middle Miocene in wide (3000-5000 km2)
hollows.In these materials, some placers of economic
interest occur, particularly at the bottom of the
proximal paleo-valleys but generally the grade is
lnfra-economic. However these materials provide gold
particles to the rlver which erode them, and slrve of
intermediate collectors allowing, so, the formation In
the actual rivers of placers deposits of economic
interest. The grade becomes higher that in the source
sediments and poor formation change by succeslve
reworking to economic deposits.

It appears that during the buldlng of the Andes in the


North of Bolivia, the gold transported by the rivers was
not totally trapped in the Cordlllera. and that some gold
particles reached the Piedmont area. During the
development of the Andean Orogeny, rellefs and hollows
are created which are younger and younger toward the
foreland (toward the NE)1 they are related to the
formatlon of succeslve thrusting fronts and piggy back
baslns.
Thls tectonic process starts at the end of the Oligocene
and if, for the time being, detrltal gold 1s known only
on and after the middle Miocene, it 1s probably because
only at that time, the erosion reached the primary
mlnerallzatlon.

References

(1) C.Herall et al., Geomorphology, vol 2, 1989


(2) H.G.Freydank, lned.rep, Denago 1964
(3) G.Herall et al., Khysos,l.3, 1986
107

NEOTECTONICS OF THE ANDEAN FOREDEEP BASIN


(MARANON BASIN) IN NORTHEASTERNPERU

Jean FranPois DUMONT* and Fredy GARCIA **

i Mission ORSTOM Apartado 18-1209, Lima, Peru

** Mission ORSTOM and IGP, Ap. 3141, Lima, Peru

Abstract

Neotectonics of the MaraRon basin has been studied using


f luvial pattern data. Migration of subsiding areas has been
identified by successives avulsions of the Ucaya 1 i River.
These data with others from the basin borders allow to define
the neotectonic evolution of the subandean/craton margin.

La neotectonique du bassin du MaraRon a et0 etudiee h


partir des structures fluviales. La migration des zones de
subsidence a et& identifiee par les captures successives de
lllcayali. Ces donnees, avec dautres provenant des bordures
du bassin permettent de preciser levolution neotectonique de
la marge subandin/craton.

Key words : fluvial pattern, neotectonics, subsidence,


Amazonian basin, Subandean zone, Brasilian craton, Peru.

Introduction

The MaraRon foredeep basin is a transition zone between


the extensive cratonic Brazilian shield, moving toward the
west, and the andean foothills pushed eastward by the
subducting Nazca plate. Field and remote sensing studies
(SLAR, Lansat and SPOT imagery) of fluvial pattern at various
stages of activity and abandonment provide a key to
investigate surface deformation of the Marafion Basin.
108

Geological setting

The MaraRon Basin is a large morphological depression


where all the rivers running down from the Andes between 2Q
and 149 south merge to- form the Amazon River. During the
Cenozoic the subsiding center of the basin migrated eastward
as a result of the folding and thrusting of the subandean
margin (Pardo 1982). 5000m of post Jurassic sediments
accumulated in the central part, and Tertiary deposits lap
eastward onto the Iquitos geanticline (Sanz 1974).

The Ucamara depression

Villarejo ([19431,1988) called "Ucamara depression" the


central part of the MaraRon Basin, a 25000 square km swampy
area, devoid of any elevation between the Ucayali and the
Ucayali rivers (fig.1). These rivers carry "white" silty water
running down from the Andes. The river divide areas are
drained and periodically flooded by "blackI waters of local
precipitation origin. The main black water rivers, follow
fossilized meander belt of the same pattern than the present
Ucayali River. They are interpreted as former reach of the
Ucayali River which have been abandonned after an avulsion of
the main river. Three successive avulsions of the Ucayali
River, cumulating a 1OOkm lateral migration toward the
Southeast have been identified from these fossilised fluvial
patterns,
The Tapiche area in the southeastern part of the Marafion
Basin was a non flooded area covered by upland type forest
(bosque de altura; Stiglish 1907). According to historical
data from direct witnesses, the area began to subside during
the twenties. A 750 square km area called Punga swamp is
presently 2m below low water level, which represent about 4m
of subsidence in 60 years. The phenomenon is testified by
remains of dead trees of the former upland forest conserved in
life position, and presently below permanent flooding.
Simultaneously, southeastward avulsion of the Tapiche river
occured. Presently the river to cross through the Punga swamp,
the white waters of the Tapiche River being limited from the
black waters of the permanent swamp by narrrows, levees, less
than 5m wide. Lower stream of the Punga, the Tapiche river
follow an older river valley caracterised by extensive point
bar type river banks and oxbows.

Andean crust and Brasilian shield relationship

The Ucamara depression is bordered southward by Contaya


and Moa Sierras, which belongs to the subandean lowland
109

(Dumont et al 1990). The Sierra de Moa is uplifted and its


northeastern border overthrusted over the Acre basin deposits
along the Tapiche fault zone (fig.1). Neotectonic activity of
the Tapiche fault area includes strong late Tertiary fold and
fault episodes, and continuing uplift of the Sierra de Moa
during Quaternary. Teleseismic data (Assumppao and Suarez
1988) suggest that the compressional regime is yet efective
between the two blocks, and concern both the Andean foreland
and the Brasilian craton.

Combination of compressional linkage in central Peru,


between the Subandean lowland and the Bresilian craton, and
active subsidence in the Ucamara depression, suggests a slight
anticlockwise rotation of the southern part of the Iquitos
geanticlfne. This rotation is responsible for the breaking of
the Iquitos geanticline in two parts, in the area where the
Marafion and Ucayali rivers merge and cross the Iquitos block.
In this area, normal faulting compatible with this rotatin has
been identified (Dumont et a1,19881.

Conclusion

Fluvial pattern studies give valuable data on the


neotectonic evolution of the Marafion Basin, and are coherent
with other types of data. Tectonically induced subsidence,
probably controlled by basement faults, generate sudden
avulsion of the main rivers. This result is in contradiction
with the RHs&nen et al's (1987) model of steady lateral
migration of rivers as a result of subandean tectonic, as well
in the Subandean areas (Dumont 1989) than in the Marafion
Basin. The effect of west Amazonian neotectonics on fluvial
changes and ecological evolution make pluridisciplinary
studies of these regions of a very high interest (Sale et al
1986; Dumont et al 1990).

Acknowledgments: This work was supported by two agreements:


ORSTOM/IGP and ORSTOM/IIAP. This is a contribution to PICG
project 279 "Terranes in Latinoamerica".

References

ASSUMPqAO M. and SUAREZ G., 1988. Source mechanisms of


moderate size earthquakes and stress orientation in mid-
plate South America. Geophysical Journal, 92:253-267.
DUMONT J.F., 1989. Morphostructural units of the Peruvian
Amazonia as related to Subandean tectonics and fluvial
dynamics. 28th International Geological Congress,
Washington, ~01.1~423-424.
DUMONT J.F., LAMOTTE S. and FOURNIER M.,1988. Neotectonica de1
Arco de Iquitos (Jenaro Herrera, Peru). Bol. sot. Geol.
Peru, 77:7-17.
_---

110

DUMONT,J.F.,LAMOTTE,S., and KAHN,F.,1990. Wetland and upland


forest ecosysems in Peruvian Amazonia : plant species
diversity in the light of some geological and botanical
evidences. Forest Ecology and Management, (under press).
PARDO A, 1982. Caracteristicas estructurales de la faja
subandina de1 Norte de1 Peru. Simposio "Exploration
petrolera en las cuencas subandinas de Venezuela, Colombia,
Ecuador y Peru", Caracas.
RAS&NEN,M.E.,SALO,J.S.,and KALIOLA,R.J.,1987. Fluvial
perturbance in the western Amazon river basin: Regulation
by long term sub-Andean tectonics. Science, 238: 1398-1401.
SALO,J.,KALLIOLA,R., HAKKINEN,I., MAKKINEN,I., NIEMELA,P.,
PUHAKXA,M., and COLEY,D.,1986. River dynamics and the
diversity of Amazon lowlands forest. Nature, v.322:254-258.
SAN2 V.,P.,1974. Geologia preliminar de1 area Tigre-Corrientes
en el Griente Peruano. Bol. Sot. Geol. Peru, 44:106-127.
STIGLISH G., 1907. Ultimas exploraciones ordenadas por la
Junta de vias fluviales a 10s rios Ucayali, Madre de Dies,
Paucartambo y Urubamba. Oficina tipografica de "La Opinion
Nacional",46lpp.
VILLAREJO A-,1988, [19431. Asi es la Selva. Centro de Estudio
Teologico de la Amazonia, Xquitos, 329pp.

Fig. 1 : Structural scheme of northern Peru. Large


arrows show the stress regime deducted from neotectonic
data (faults) from the Tapiche fault zone (black arrows)
and from the Jenaro Herrera area (white arrows).
. ,_,..
____,_
.,.__ ..,,

111

LATE CENOZOIC NORMAL AND STRIKE-SUP FAULTlNG IN NORTH


PERUWAN WESTERN CORDILLERA : AN EXAMPLE OF ALTERNATE EXTENSIONAL
AND COMPRESSIONAL TECTONIC REGIMES IN HIGH ANDES

0fM.r BELLlER*,Mid.1 SEBBiER*

l URACNBS (73g) G6ophysiquolt Olodynamiqur htemo. BLt.BIg. Unlversh6PARS-SLID,014w ORSAY

RESUME

Au coun du MiocAno,de8 deml-grabenr sorganisentparall6lemmt aux awidmtr majoursdo la Gordill6rr


Owidontslr du Nord POrou.ll ry awumulmt de8 d6p6ts fluvio-laoustrorau wun dunr phase dextmsion ENE-
WSW. Au N6og6nr sup. et au Guatemaire inf., la r6gion rst alon soumir 1 drux r6gimrr d6crochantsdont Ior
direotlonsdo oompresslonsent WNWESE puir N-S. Depuit Ie Pl6istoo6nemoyen, Ie r6gime ett caract6ris6par une
extensionde dimwon N-S. Ges r6gimer teotoniquersent interpr6t6rwmme la wndquenw der forces de volume
li6er 6 la haute topographicet de8 foroer de bordureli6er 6 la wnvergenoa de8 plaquesNaxwAm6rique du Sud.

Key Words : Halfqraben, Extension,Compression,Tectonicregimes,WesternGordillera,NorthernPeru.

I - INTRODUGTION_

The PeruvianAndes is a wrdilleran margin where the subductionis lcUvesincethe Jurassic.The,periodfrom


Early Miocene to Present-dayIs characterbed by a high convergencerata between the oceanic Naxu Md the
wntinentai South Ameriun plates: convergencedirectionbeing nearly orthogonalto the Peruviantrench. S6brier
and Soler (1989) have reported4 compressionaltectonicpulses which affect the GentralAndes after the Eocene
and Oligocene main deformationphases 3.e; 42 Ma Md 23-23 Ma dated phases).These pulses are dated: lower
Miocene (1517 Ma), middle Miocene (10 Ma), upper Miocene (7 Ma) and early Quaternary(2.5 Ma). The lower and
middle Miocenepulsescorrespondto locallydeformationalcrisis.After the maln upper Eooeneshortening(Incaic
phase)that affectsthe High Andes, severalbasinshave been initiatedalong major faults in the Western Cordillera
of northernPeru as for examples the Cajabamba, San Marwr, and Namora baslnr (fig.1). These ones were half-
gralwnr infilled by Miwene fiuvidacustrine deposits Md bordered by normal faults that reactivated old
wmpresslonal structures.The GajabambabasinIs situatedIn the westernedge of the NNWSSE trendingMa&on
ThrustMd Paid Belt (see M6gard, 1984), the Namora basin Is wlthln the E-W Gajamarca deflectionand the San
Mamor baskt, situated betweenthe GajamarcadefieotionMd the Marafion Thrustand Fold Belt, is bordered by
NW-SE trending rtructurer. The moan altitude of this studied region Is 3100 m (calculatedon an area of 10 Ooo
km2 [Belller,19891).The studyof the 8edimematYMd tectonicevolutionof these basins,wnstraint by radiometric
and paleontologlcaget, permit to show lvidenw for longer perioda of moderated Md lwailzed extensional
deformationseparatedby shortperiodsof compressionaldeformation.

II - MIOCENE SEDIMENTARYEVOLUTION

The Gajabamba Md San Marws bashr were filled by a thick fluvio-facu8trlne sedimentaryGroup (fig.2). This
group is characterizedby a mean 350 m thick IacustrineformaUontermed the Csjabamba fm and wnatituted by
alternatedshales,marls, lignitedshales,gypsum,diatomits.siltsand sandstonesdepositswhich are overlyingliffle
wnglomeratio fluvial series. ln this lacustrineformation we observed volcanidoam facier Md aynudimentary
normal faults indicating that voicanism and tectonioswere active during the basin evolution. This Iacustrlne
fomatlon Is overlapedby a meen 500 m thkk furglomerate formationtermed the Gondebambafm. This one, Is
wnstituted by mud-flowand fiuvfalsequencerat the bottom and warser torrentialsequencestoward the top. The
age of this group was lsUmatedlsrfy to middle Miocenethanksto diatom assemblagesexposed In U-IOIacustrine
deposits(Bellieret al., le Fourtarrieret al, in prepuation). The tectonicframeworkpermit to lsUmatethe age of
this Group probably ranglng between 13-17 h6aMd 10 f& I.e., between the iower Md middle Miowno tOCtCfIi0
pulses. The Gondebambafm Indicate that a climatic change had owured between the Gajabamba fm and the
- ,_-- -. I- ,_, _..

112

E
8

0 I
-..._--

113

CondObunk tmfhQ8. h rdditbn, ths 08t8strOphlc 008s~ 0on@0msr8tlo f8d.s of this Qroupprobrbly m8rksd
ths obsinQ of ths b8sins 8ssooirtsd to the miHl0 htbcens tsotonb pubs. Ws can lsUm8ts 8 moM drposit rsts
8nd oonsquontly 8 m88n subsldona nts in ths b8sins during this p&d: t8kin~ into 8ccountths docomp8ction
p8ruMtrrsws osn lsUm8ts 8 m88n loo0 m thbk, dsoomp8otsdssdimsnWy cdumn whbh hu bson doposItsd
du~w~Th~~ta~~~~mr~wbJdrnor~&~ 0.17tO.03 mm/vr.
~NunonbutnbRllrdby81~~thidc2~~~~trnnrd~~8fmloully
ovorfylngths Condob8mb8fm wlth 8 silQht unoonfwtty. Thir fomutbn w8s ch8motsrizsd by lbmont8ry fluvkl
squrnorr whbh puud pro~rouhnfy tow8rd ths top to boustkw soqusrws m8ds of rurds, siits. sh8bs,
dbtomlts, IiQnibd shsbs, oubon8tsd oonwstbns 8nd Umsstonoa h the middb put of ths boustdnr ssqusnw,
soms d8dUo tuffs wsro lntorbsddsd within di8tom bsds. This ktdbsts 8 wb8nb soth4tycontsmpor8nsousto ths
b8sln filling. 0-1 ths othsr hsnd, synssdknsntsry nomul f8ults Indk~@ 8 Nomul furlUng 8othAty. Ths
psbontobQbsl~8lysls of dbtom ssmpb Qw 8 upper Mbosno s~s (Ssllbr ot 8l.. loBob; Fourt8nbr st 8l., In
prspu8tbn). Rsoantfy, wdlomotrb 8n8lysss of ths N8mon d8oltb tuffs (Klk 8nslysss by M.G. BonhOmms) givs
M rgs Of 7.2tO.6 Ms for ths middb-upporput cf ths Nsmorsfm. Thb b ln rqrssmont with rgss of ths N8mor8 fm
rsnQlnQm ths middb 8nd uppsr Mbo8nr tsotonb ph8sss; I..., bstmn 10 8nd 7a h Ws h8vs lsti~tsd 8
msxlmum 300 m thbk, dooomp8otod ssdinwnuly columnof ths N8mor8fm whbh hu bssn dspositsd during 2-3
Ma This psrmlt to o8bul8ts 8 subsidsnos rrts of 0.12t0.03 mm/r.

Ul- MlCCENE TO QUATERNARY TECTONIC REGIMES

Ths Mice8110to Cu8tsm8ry tsotonb svolution of northern Psru intr8uwdillsrM b8sins shows sxtsnsbnrl
psrbds sspsr8tsd by comprsssiorul dsformrtbn crbia This wok&n wss invssti~8tsd uslnQtsotonb M8lysb 8nd
fsult kkwn8tlcr 8rulysls. Ansfysb of thsss fsult kinsm8tbs In tsmu of strsss wu dsvsbpsd usinQths oomputsr-
sided msthod pmpossd by Cusy (1979).This 8nslysis show svidsncs for 5 succsssivs Qsodynsmbst8t.s (Ssllisr,
1Qw):
1 : ?ho San Mucos ad Cajabambah8lf+rsbsns wars dsvsb+sd during ths mkldk k&cans
contsmpor8nsouslywith ths Wsstsm @rdilbrs uplift snd. Thb Mbcws b8sin8l svolutbn wss oe8v8l with sn
sxtsnsbn8l t8otonb roglms (l-110.3) ch8r8otsrixsd by 8 domirmnt ENE-WSW trsndlng tsnrlon sxpossd by
synssdfmonwy nomul f8ult 8ffsctkw th8 cJ8bsmb8 fm. mi8 wwsw trsnding8xtmrbn pdsts 8 minor NE-
SWtrsndin~ sxtsnrlon (2-fio.3)whbh 8ffsots 8lI ths reeksprbr to ths Nsmorr fm. Ths Furglomor8ts Condsb8rnb8
fm s8sms to indluts 8 tsctonb oh8nQSwhbh m8y corrsspondto ths 10 Ms oomprsssion8lpub.. mi8 pubs m8y
h8VOpfOdUordthrOlOsingOfthObuiM.
2 : Than, the N8nmf8 b8sins wu f~msd, bstwsan 10 snd (I M8, sbn~ WNW-ESE trsnding Md S dipping
mJoIfurltr.~~8~mmr~by~~ntuynomulf~WIwhiohind~~8ENE_WSW~nJonJ
tsotonbs (l-fio.3). This synssdlmontuy f8ulting dsfomwd ths d8dtb tuff lswls lndbstino that vokmim,
8xtonsbnal tsotonbs, snd ths N8mor8fillip mm oosv8f.
Thsn, ths Nsmor8, SM Msroos 8nd Csjsb8mbr buins mro 8ffscted by WNW-ESE snd N-S shortwings (3
8nd 4-flQ.3).Struotuml obswationr show th8t ths N-S trsndinQ oomprossb+lpostd8tss ths WNW-ESE trsndinQ
on..
2 : lho WNW-ESE trsnding aw~prsssbn (3-fig.3)produw fokfs, rsvsrss 8nd striks-slip f8ults sffsctlng sll ths
Mbosns dsposita This dsformsUon ph8ss msy bs centisr& first, 8s ths lsts Mioosn8 pubs drtsd st 7 Ms In
othr Csntr8l Andss put: thsn, ss M other comprss8bn8l pubs, prob8btybts Noogsns too; Lo., y-oungsrth8n ths
Nun0r8 fm.
5 : ThaN-StrondlnQoomprsssion (4-fig.3)8pf~us m8jor with nspscf to thr WfVW-ESE ons Md msinly in thr
cJ8msrca dsfbcUon. 2 producedrsgbrul fbxuntbn, folds, futltsd ovsrfokfs, rsvsrss 8nd striks-slip fruits 8s wsll
8s bobtod 8nd sn s&&n* twktnsl~8shss. lhsss dsfomwUonswsrs ths sffsot of ths Esrly Curtsmuy pubs.
Ths kinsmatbs uulysb drmonstr8tss th8t ths Nin (WNWESE snd N-S) oompresaionrldsfomwUon8wsrs
produosdwith strlks-sllp roglmoa
S : lho EarlyPloi8tocmo oomprsssion8l pubs b follovnd by 8 ns8rly KS trsndinQ sxtsnsional tectonics (5
fig.3). Thls tsctonbs b m8rksd by minor nomA f8ults whbh 8ffects ths Mssozoio bsd-rock,thr Nsogsns b8sin
filling 8nd ths Cu8tsm8ry fluvbl tsrmoss 8s wall 8s by 8 nomwl8&s f8ultz ths Chsquilb8mb8nomul fsult. This
on8 wu rssothmtsd In 1937, 8nd shows 10 m high vsrUc8l dlspbwmsnt of ~18cJslmor8kw crssts of ths last
Gboirtbn (Ssllisr lt 8l., lQ3Qc)IndicrtinQthat ths N-S sxtsnsion is ths prsuntdsy ststs of strssa

ln oonduskw, this study shows 8 doss rsbtbnshlp bstwssn tsotonb, ssdimsntuy Md volunb 8otivlUss
durlnQths Mboww to Prsuntdry svolutionof ths Wsstsrn Cordillsrs ln Northsm Psru. This ngbn is dw8ctsrizsd
by 4 m8jor tsctonlorsglmos: 8 Miocunssxtsnsbnal rrgkns hsvlng s nuinly ENEWSW trsndlng tsnsbn oosvsl with
Intr8oordilbrM bulns svolution snd volunism; L8ts Nsogons snd E8rfy Pbistocsns comprssslonal tsctonio
rsgimss having WNWESE Md N-S trsndlng shortsning: 8nd, s Mid-Pbistocsns to Presentday sxtsnsionsl
tsctOnicrh8vinQ8 N-S trsndlng t8nriOn.
This 8n8lysls shows that ths L8ts Csnoroio stress dirsctions us homogeneousboth In ths NNW-SSE striking
Cerdilbr8n ssgmsnt md in ths E-W striking C8jsm8ro8deflection.Thb suggests thrt rincr the ssrly Mbcsns thsrs
_(r__,_ __,.,.-_ .,_... _-.

114
wu no eignlRoyrt dHtoronUal routIon botwoon the E-W CJvnuu dofkdon and the NNW-SSE striking &rn
PoruvlanAndrrnBolt.
k lt hr boon propoaod In wuthrm and oontral Pwuvlan k&a (S&rlor ot al., was, roes), the vuiauonr ti
~~~ot~ouIn~m~n,~pur(okr~~~lottH~~~~bounduyfoc~e
du.to~~o~~~~~body~d~~Ohbhtopogr~.~v~tionrot~~~otmou
urdwlomodiflcrtknrof~bwndrry~rkcru~~topoOnphyhunotdrutiully~g~du~gtho
ootuidorod time period. The ohangw of boundary forow may be the offool of modHio8tionr lithor in the
oonvugona mto (s(brior ot al., 1985: 1oeS) ti, In fho 8lab-pull for0 (Sorol ot al., 1Oee) or in thr &a-mtwt
(Skrlor and Sokr; 1989). The MwPloi8tooww and wtlva KS trwIding lxtoMbMl tootonicrofthogonalto
~~dinction,irupWmdbynl8thn~ktrmn~bounduy~rmdbodytor#aThoktr
Miooono and udy Ploietooono oompmaeional twtonloa aoomr to be nlatod to an inorour of the boundary foroor
duo llthor to l high oommrgonoa r& or to a8l8b ruptum (Lllkr, 1989; S&riar and Solr, 1gSg). The kte Miocwo
ENE-WSW Wondlng lxtanrknrl took&a, k awolatod with prodominwt gravitatIonal body foroor duo to thr high
topogmphy and dlmlnutlon ol tJw boundary fomor b&g duo lithor to diminution ol oonvorgonoo rata, or to 8trong
8lab-puuor- lnduerd by l long alab. Momover, iho mod roglon llluetr8tor that thr Inoroaso of thr
vwikal8tmu la udfloionl to produoo lxtonsional drfonnrtlofu won H the moan topography ir nol wry high
(nruiy 31Coml.

REFERENCES :

Bollhr 0. (1989). - Thbw Univ. ParCSud, Orsay, 2gSp.


Sollior 0.. S6brirr M.. Gasu F., Fowtanirr E. and Roblrr I. (19esp). submitted to Gkdynamiqur - ORSTOM.
bllkr O., S(brlrr M., Fowtanir E.. Ourr F. and Roblrr I. (1oBgb). Annalrr Tootonkao, Vol.lll, nY, in preu.
blllor O., Machub J. and S(brirr M. (1oesC). Blotin do la Sooiodad Goologiu drl Peru, in prru.
Cuoy, E. (1979). Rev, G6ol. Dyn. Ologr. Phyr. 21, !i7-6&
ward, F. (1984). Journal Goologioal SocJoty London. 141. p.893-ooo.
8)brhr, M.. J. L M&or, F. ward, 0. Laubachrr and E. Cuoy-Gailhudir (1985). Teotoniu, 4, p.739760.
s(brior M., Morok J.L, Maohub J., Sonnot D., Cabmm J. and Slant J.L (1oSe). Teotonios, 4. p.895928.
S&ior M. and Solr P. (1980). G.S.A Special Papor : Andoen magmatlem end its toot&o uttlng, aocsptrd.
Soml D.. Moroior J.L. Koraudron Rand Cuehing M. (1seS). C.R Aud. Scl. Park, 307, r(rio II, p.lgSl-1086.

Fig. 3 - Neogene to Oustemery gcodynrmic evolution in


the Weeccrn Gxdillcn of Nonhem Peru.
1 : lntnsordillcnn Miocene ENEWSW (1) and NE
SW (2) extcnsionel tectonics
2 : lntnsordillcran Lte Neogenc-Early Pleistocene
WNW-@SE (3) and N-S (4) compressionaltectonics
3 : Intn-cordilleran Quetemary NS extensionel
lecconicl (5)
Boxes en : (1)Miccene normrl faults; (2) regional faults;
(3) Men6on Thrust md Fold Belt: (4) SubondeenThrust end
Fold Belt; (5) Subduction trench; (6) studied basins (N)
Nsmon ,(SM) San Marcoc end (CA) Cajabrmbr bash% (7)
compreeeionel direction: (8) Miocene enensionel direction;
(9) Neza plate eonvrrgencc direction
115

PREMIERS RESULTATS DES ETUDES DE FAILLES ACTIVES


DANS LES ANDES DEQUATEUR

LAVENU c\. *, WINTER Th.** et AVOUAC J.Ph.++

(Convention IPGH-EPN-CLIRSEN-ORSTOM)

* ORSTOM, Ap. Post. 6596 CC1 Quito, Ecuador


213, rue La Fayette 75480 Paris cedex 10, France

+* Laboratoire de Tectonique, Mecanique de la lithosphere


IPG Paris, 4 place Jussieu 75252 Paris cedex 05, France

Resume
Les premiers resultats de l'analyse des deformations
rlcentes en Equateur permettent de caracteriser le champ de
contrainte et d'ebaucher une interpretation g+odynamique de
cette partie des C\ndes. L'interaction des effets d'une
haute topographie compensle, de la subduction de la jeune
lithosphere oceanique, form&e a la dorsale Cocos-Narca,
sous le materiel d'origine oceanique constituant la tote
equatorienne et colombienne, et d'une convergence oblique,
induit le champ de contrainte qui se devel oppe dans les
Andes septentrionnales et est responsable de la migration
vers le nord de la partie occidentale de celles ci.

Introduction

Dans les Andes centrales, la tectonique recente est


maintenant connue et expliquee. Dans ces regions, le champ
de contrainte actuel resulte de l'interaction d'une
compression horitontale like a la convergence des plaques
Nazca et Amerique du sud et d'un effet topographique dQ a
une compensation isostasique des masses lithospheriques.
La tectonique tertiaire et quaternaire en Equateur est
un des exemples de la complexite des dBf ormations
intracontinentales pres d'une zone de subduction.
En ce qui concerne la tectonique quaternaire (fig. l),
une d&formation en extension tres importante est presente
actuellement dans le Golfe de Guayaquil OQ 4500 m de
sediments plio-quaternaires y ont et& reconnus. Elle existe
Bgalement dans la CordillCre Occidentale du sud de
1'Equateur (faille de Gironl. Plus au nord et faisant suite
aux grandes failles du Golfe de Guayaquil jusquau sud
d'Ambato, nous avons dtudie le systeme decrochant dextre de
la faille active de Pallatanga. Enfin, dans le nord de
116

l'Equateur, un &tat de contrainte compressif de direction


proche de E-W est responsable de l'activite de failles
inverses et decrochantes dextres dans la depression
interandine.
Les rrlsultats present&s ici sont la compilation de
plusieurs articles edit&s ou sous presse concernant une
partie des travaux effect&s dans le cadre de la Convention
IPGH-EPN-CLIRSEN-ORSTOM.

Les deformations en extension du Sud de 1'Equateur

L'analyse morphologique et structurale du systeme de


failles de Giron permet de mettre en evidence son jeu
recent et actuel en faille nor-male, compatible avec une
contrainte principale 0 1 de direction proche de NW-SE:
N135E-N145E ( fig. 2). Cette tectonique en extension est
Ogalement presente au nord de Giron (failles de Canar) mais
elle y semble moins importante et moins active.

Les deformations en compression du Centre de 1'Equateur

Elles correspondent au systeme actif des failles de


Pallatanga. Ce systeme correspond a la reactivation de la
partie meridionale de la suture creee lors de l'accretion
de l%rc Macuchi oceanique a la cata au Cretace superieur.
Ces failles debutent dans le Golfe de Guayaquil
(segment NE-SW) et se terminent au NW de Riobamba (segment
N-S).
Alors que l'analyse microtectonique de quelques sites
de la partie nord du systeme indique un regime local
dkrochant extensif, l'etude morphologique de l'ensemble du
segment interessant la Cordillere Occidentale montre
parfaitement l'existence d'une composante inverse (bermes,
collines deplacees) et dextre ( "pull apart graben" de
Pallatangal du mouvement. L.&tat de contrainte en
profondeur est bien ddcrochant compressif.

Les deformations en compression du Nord de 1'Equateur

Trois zones principales sont a consider&es, toutes


trois affectees par une tectonique en compression de
direction de raccourcissement E-W.
La zone entre Ambato et Quito constitue le
prolongement vers le nord du systeme de Pallatanga. Darts le
basrin de Latacunga, des flexures actives de direction N-S
affectent le remplissage plio-quaternaire de la Vallee
interandine.
.- -_ __.__l__.__
_.. ~..I_...__ -
118

La ville de Quito est construite sur une faille


inverse de direction meridienne A pendage ouest dont le
rejet peut Otre estime A 400 m. Des depots volcanique
r&cents (Formation Cangahual dessinent une flexure au
dessus de cette faille inverse et sont affect&s par des
microfailles failles normales de direction elles aussi N-S
(glissements bane sur bane, extension d'extrados et/au
bifurcation de faillesl. La direction de compression E-W
est confirm&e par un mecanisme au foyer en faille inverse
pure (Bonaz, comm. pers.).
Plus au nord, prbs de,la fr.onti&re avec la Colombie,
un grand decrochement de direction NNE-SSW a et& mis en
evidence, par la morphologie. 11 est dextre et s'amortit au
sud, pres d'lbarra, sur une structure en queue de cheval.

Lea deformations de la zone subandine

Dans la zone subandine du nord et du centre de


l'Equateur, les mecanismes au foyer des s&i smes r&cents
sont en faille inverse pure (Cifuentes t Lyon-Caen, comm.
pers.1.

Etat de contrainte quaternaire en Equateur (fig. 31

Dan8 le centre et le nord de l'Equateur, la contrainte


principale 0 1 est proche de la direction E-W de la
convergence. Dans la zone subandine et dans la zone entre
Ambato et Quito la contrainte principale intermediaire 0 2
est horizontale et N-S. Au nord d'Ibarra et au sud d'Ambato
la contrainte principale 0 3 est horizontale et N-S. Dans
le sud 1,a contrainte principale 0 1 est verticale et 0 3
devient NW-SE.

Conclusion

L'etat de contrainte, decrochant compressif, qui se


developpe dans les Andes du centre et du nord de 1'Equateur
eat induit par l'interaction dun effet de haute
topographie compensee (impliquant ici l'etablissement d'un
regime de contrainte intermediaire compressif decrochant)
et le couplage important a la suite de la subduction de la
jeune lithosphere oceanique form&e a la dorsale Cocos-Nazca
sous le materiel d'origine oceanique qui constitue la Cote
&quatorienne. L'obliquite de la convergence serait a
I'origine de l'ouverture du Golfe de Guayrquil et de la
migration vers le nord d'un Bloc Andin (Cdte et une partie
de la Cordillere Occidentale) le long de dbcrochements
dextres (fig. 4).
_.-. __I.._.._ _.._^_

119

A WEDGE MODEL FOR THE QUATERNARY TECTONICS OF THE ANDES OF ECUADOR

B Md Luca Ferrarl

Dipartlmonto di golonzr dolla Torm,Untversltyof Milan, Via Mangiagafli g4,20133, Milan, kaly

Jurquau Pl&too&r inf6rlrur la Cordill6re Ftealoet la Zone Subandinr ont 6t6er Inter888688 par drr frill88 obliques drxtrrr
ot dsr d6crochrments dextrer lvoc dlreotion NNE-SW. Dan8 IO m6mr temp8 la Vall6r htemndlnr a bt6r unr zone de
d6crochrmentsMrstro. lglon va proposer un mod&e oln6matiqur do8 mouvemmt8 du bloour croutal de la Cordillrro Real8
quo a oto lleve lt pouaab
VWNord rolatfvemont6 IavMt-pays Amuonlqur lt A la Cordlll6ro Ca3dmWe.
Key Wordr: Ecuador, Pnde8, Teotonlcr, Cuatemary

The NNE-SSW Andran Chatn In Ecuador Is charaoterlzed by the premoe of thr pamllel ranges of the Cordillrra Occidental
(CC) and thr Cordlllera Real (CR) divided by the n8rrow drprwalon of the lnterandrM Valley (IV) Md connrcted to ths
Amazonlan Platform bythr Sub8ndran Zone foothlllr. Thr deformation8 ocourred In thlr region slnw hfiocenr, am thr rewlt
of a oomplex Interaction among Nazca, South Amrrkzan Md Caribbean plater. Thr diffuse Md high selsmlolty of the
Ecuadorlan Andrr (with events rrnglng up to Ms=7) ~UQQOMthat thk region I8 8Ullwbjrcted to 8trong teotonlo activity. ln
thlr aroa 8omr work8 lnfora western dextral tmnrcurront boundary of northrm South knorlca whloh passes trough Ecuador
In an uoa comprlrrd between thr Cordillrra Cwldontal and the AmazonIan Platfom (Pennlngton, 1081;Foiningor Md
t%guln, 1983). &cent struoturalflrld work8 (Ferrarl and Tfbatdl, 1989 and In press: Tlbaldl Md Coltelll, 1989; Paquu4 et
8l., In pros& Tlbaldl Md Rrrarl, In press) provided a more predr ploturoof thr teotoniolvolugon of the Eouadorlan Andre.
The Integration of the structural data oolleoted In the fteld and their wmparlson wlth other grologicsl and geophlslcal
information8 prompt u8 to attempt the synthrsls on thr regional tectonics of Ecuador which I8 exposed In this work.

Evolution of the Quaternary deformations

The mrin 8truotunr of the ohaln were built during a major phurr of deformation (Pasquar6 et al., in press) which occurred
before Plelstoosnr Md produced malnly pure thrurtr. A ohMge In thr deformatlve mechanisms can be envisaged bstwem
the end of Plkcmo Md Early Plelstooenr. Indeed deformation occurred durlng the Cuatrmary both In the CR and In the N.
h many case8 thls new teotonlo sltuattonled to ths reactivationof the pm- lxl8tlng structureswith M alongutrikr Wmpon8nt
of motlon. Mlcmteotonlo results atlow to dlvldr thr quaternary deformatlve hlatory In two teotonlo pha8er occurrod during
Early-Mlddlo Pfelstoceno and tata Plrl8tocme-Holooene. tn the northom CR most of the thrust plan08 Md rever faults
developed during the previous phass show the 8uperlmposhion of sn obliqur-stlp 8triation. Freducsd stress tensor solutions
for thrr fault population8 Indicate that these structures were reactivated by a stress fisld with a mean E-W dlrection of the
greatest principal stress (sigma 1) (Fmarl and Tibaldl, 1gSgMd In press). The numerous NNE trending right-lateral strike-slip
faults which out all the prevlout 8tructure8, represent the evidence of a sscond phase of deformation characterlred by a
mean WSW-ENEdlreotion of the slgma I. The development of ther Wucturet can be datsd at the We Pleistocene because
theycuttheoldestrocksof ReventadorVoloanowhichylelded aradiomrtdcagoofabout0.35M.a. (INECEL, 1988). Lacustrlnr
Md pyroda8tlo depodts which fill thr N display the superlmposltion of a tensional phase on a older wmpresslve one. Late
Pfioo8neto Early Plel8tocenr rsdimsnts and lava flow8 are affeoted by fold8 wlth a mean E-W axis and reverse and strike-slip
faults which am wnslstent with a NlTCF-l&I stoma 1. Voloanlc and fluvial drposlts datlng from loO.lIW y. B.P. to Holocene
show no evldenoe of wmpressive deformation but are affected by NNE to N-S trending left-lateral normal and pure normal
faults supporting M average N 8(r dlrsottonof the least principal stress (sigma 3). The general architecture of the N , which
120

rommblra ~&on, 8nd ths 0vorall prodomlnrnor of loft-Woral oomponont of movonunts In thr fauits of both thr
plum wggw thatthearmhaaboondomitutodbya tootonbrogimowhichcJmngd fromtranoprosaiwto transtonsivs
~baldl and Mtolll, loco; Tibafdi and Forrul, In pros@.

Flg.1. Main Ou~tomuy too


tonb struoturos of northwe
stem 8outh Arnoriu. Dottsd
aroe lxtrudod crusW wedge
accordlng to the mod.1 pre.
untod in thls wok, 1.
fronoh, 2. thrust, 3. strik~-xilp
fault. 4. right- lateral strikr-
siip fauit with a rsvorss com-
ponmt, 6. Inaotivo thrust, 6.
ooranie ridgr, 7. block rota-
tion inferred from @some
gnrtlo data (aftor McDonald,
SW), 8. ori.nWfon of sigrM
1 doduwd from miorotscto-
nlcr (in Coiombla aftsr Ta-
boadr and Philip, 1988), Q.
rolrtivo block motion (aftsr
K&g md Bonini, 1985), C.
Caribbean Plato, 8A South
Arnrrica Plato, ht. Maraoaiba
Block, N. Nazca Rata, NA.
North Andem Block, CR.
Camrgio Ridge, P. Panama
City, 8u. BUoaramanga, 8.
Bogota, a. ouito.

Discussion and conclusions

in our intorprotatbn thr CR roprrrnts 8 wodgl-ohrprd orustal block which Is being uplifted and extruded toward ths
North as a oonsequrncm of the nrufy E-W oonvsrgoncr of the Nuu and South knrrican pWss (Fig. 1). The rrlWvo
motion of this biock with rospoot to thr adjrcwnt s&on oan lxpWn the right-iatemi and loft- Werai shear wmponant
rospeotively obrrvod In thr lastom CR and In the N sines Eariy PM-no. Thr *astern limit of the CR block is
roprounted by tha East Andran thrust systmm.Thr wostom limit is lxprourd by thr Tertiary suture zonr between oosanio
and oontkwnW GNIt whkh undorlirs thr N (Folningorand !Ssguin, 1983). 80th are inhrritsd struotural features reactivated
in Quatomary Umrs with a strike- slip componrnt of motion. From the Gulf of Gu8yaquil to the NNE two fault systems @it
aput: thr oastom onr, with a right-Moral wmponrnt of motion, limits to the South thr IV (Puna-Paiiatanga and
JambeltNamnjal f8ult systsm of thr Gsologloai Ms@of Eouador), orosses thr CR and then roachrs the East Mran thrust
system; ths wostom onr, with a left-latsrrl wmpomnt of displaamsnt, goss through the CC (Milagro-Guarand~ fault
system of thr Gsoiogicrl M8p of Eouador) and rrti@s the N North of Fliobamba where It connrots to the suture zone.
According to roomntnootoobnb survsys hovmvor (Soulas, 1989), thr ntos of displrormrnt would bo greater In thr onstem
syst0m aiong whioh would bo oencantr8tod from 809( up to ths 90# of thr tow motion.
Th0 seismicity of thr region applus to k consWont with thr proposed modri. shallow wonts ooiiectod integrating the
NaAq CERESIS and Ctmarv8torio ktronomico do Cult0 (lg81) uWogurs for the years 19051985 AD. showa that
uiomicity k oonoontrated along Wo broad bslts with a N-8 to NNE-88W orirnWion. Aput from thr obvious uismicity
related to thr tronoh zons, the seoond rismio ton, is oentsrsd on the iV and CR up to thr Guayaquil Gulf whsrr the two
belts merges. No graat ruthquakes UI rsportod slang ths CR South of Wtudo r 8, which reprsssnts the southern limit
whrro thr East MOM Thrust system Is rraotivated with a right-iateral component of motion.
Thr proviousiy lxPosod tsctonio mcdri CM bs convonirntiy lxtsndad to thr whole tootonios of Northern Andes (Fig. 1)
Thr right-iatsral strikeslip fault system of oastern CR oontinurs in the coiombian tsrritoty whsrr wismoiogioal studiss
121

(Pennington, lg61) suggest a right-lateral displacement along the East Andean Thrust system. The left-lateral shear zone
of lnterandean valley could be inferred northward along the fault zone of thr Csuca Valley where anticlockwiss rotation of
t&o Miocene rocks have been proved by paleomagnrttc studies (MacDonald, lg60). In this frame the CR crustal block of
Ecuador oould represent the southern termination of the North Andean Block of Kellogg and Bonini (1962), a broad zone of
upllfflng Md compresston which rosults from thr obliqur convergence betwrrn Nazce and South American plates with the
Influence of the CaribbeM onr In Its northern part. Thr conoept of rigid block however has to be carefully considered In
the light of recent gravity Md geodetlo measurements which showed different rates of uplifting within the block itself
(Liisohen, 1986).

Acknowledgments
This work was supported by a contraot between Universtty of MilM Md ELGElectroconsult, Milan. A T. benefied of a Ph.
D. grMt from Mtnistero Wlano della Pubbllca lstruzlonr.

References
Felnlnger, T. Md Seguln, M.K., 1983. Slmplo Souguer gravity Momafy field Md thr inferred crustal structure of oontinental
Ecuador. Geology, 1I, 40-44.

Ferrarl L Md Tlbaldl A, tge0. Seismotectonlcs of North-Eastern EcuadoriM Andes. Annatee Geophisicae, special Issue
on EGS XJV General Assembly, 39, (abstract).

Ferrarl L Md Ttbaldl A Recent Md active tectonics of the North-Eastern EcuadorlM Andes. J. Geodynamics (submitted).

INECEL (fnstltuto EwadorlM de Uectrlflcacl6n), lg66. Estudlo vulc~ol~lw de El Fbvontador. Cuito, Ecuador, 116 pp.

Kellogg J. N. Md Sonlni, W. E., 1962. Subduction of the Caribbean Plato Md basement uplifts In the overriding South
Arnerlean plate. Tootonics, 1,261.276.

Kellogg J. N. Md Bonlnl W. E., 1966. Reply to C. Schubert comments on Subductton of the Caribbean Plate and
basement uplifts In the overriding South American plate. Tectonics, 4,7,766-790.

U&hen E,, 1966. Gravity Md height changes In thr ooean-contlnent transltlon zone In westsrn Colombia. Tectonophysics,
130,141-167.

McDonald W.D., lg60. Anomalous palromagnetlo directions In Late Tertiary MdeoitiC lntruslons of the Caucs depression,
Colombian Andes. Tectonophyslcs, 66,3Xl-S46.

observatorfo Aetronomtw de Quito, 1981. Catalogo de Stsmos del Ecuador, 1900-1880.

Pasquare, G., Tlbafdl, A Md Ferrarl, L, in press. Fblatlonshlps between plate wnvergence Md tectonic evolution of the
Ecuadorian active Thrust Selt. In: Critlcel Aspects of Plate Tectonic Theory. Throphrastus Publications, Athens.

Pennington, W.D., 1981. Subduction of thr Eastern Panama Basin Md selsmotectoniw of North Western South Aneriw.
J. Geophys. Fbs. 66 (61 l), 1076%10770.

Soulas, J. P., lgeg. Definition de Iactuello plaque Csrribe. EGS XJVGeneral wmbly, Sarwlona, 1517 March (Abstract)

Taboda A and Philip H., IgSg. Quaternary activity of the Santa Mar-ta-Sucaramanga fault. EGS XIV General r%sembly,
Sarwlona, 1317 March (Abstract).

Tibaldl A Md Coltalll M., IBsg. Struoturaf setting Md Pfio-Cuatemary tectonic evolution of fnterandean Valley,
Ecuador. Annates Geophislcae, specld issue on EGS XlV General Assembly, 3839, (abstract).

Tlbaldl A Md Ferrarl L From Pliocene pure shear to Quaternary Ieft-hMd trurspression and transtension in the
MerandeM Vafley, Ecuador. J. Geodynamtos, (submitted).
123

NEOTECTONIQUE DU SUD-OUEST DE LA COLOMBIE : RESULTATS MICROTECTONIQUES


ET APPLICATION A L'ETUDE DE L'ALEA SISMIQUE SUR LE SITE DE POPAYAN

J.L. Bl&s+, W. Marin*+, G. Paris**, B. Sauret*, H. Vergara+*

l BRGM 4S/ENV - Domaine de Luminy 13009 MARSEILLE (FRANCE)


l*INGEOMINAS Regional de1 Pacific0 A.A 9724 CAL1 (COLOMBIE)

Dans le cadre du programme de microzonage sismog6otechnique de la ville


de Popayan (Cauca), affectee par plusieurs tremblements de terre histo-
riques dont le sCisme destructeur du 31 mars 1983, cet article pr&.en-
te les rBsultats de 1'6tude microtectonique r6alisCe conjointement B
1'Btude n6otectonique et geologique du site, en vue de determiner les
champs de contrainte recent et actuel et de connaltre le fonctionne-
ment des grandes failles.
L'analyse structurale de terrain a Bt6 effectuke exclusivement dans la
formation Popayan d'lge plioquaternaire et constitu6e d'ignimbrites,
d'agglom6rats et de niveaux de cendres alt6r6es. Le traitement des don-
r&es a Bt6 effect& grace aux programmes informatiques GEOS et CAREYLU
du BRGM (projection stkdographique et calcul du tenseur de contrain-
tes).
Trois Episodes de deformation successifs ont CtC mis en gvidence, B sa-
voir du plus ancien vers le plus recent :
1 - une compression E-W
2- une distension N-S B NE-SW
3 - une nouvelle compression E-W.
Les 616ments de datation disponibles permettent de rapporter le ler
Episode au Pliocene supkrieur et le 25me au Quaternaire.
La dernike compression E-W affecte des niveaux de cendres tr& &cents
. ._I^_____
.. . ,, .-.--.

124

et est compatible avec leS m6canismes au foyer des seismes du Nord des
Andes et avec ceux du SCisme du 31 mars 1983 (choc principal et rgpli-
ques1.
A partir de la connaissance de ces champs de contrainte recent et ac-
tual, on d&ermine les modalit& de fonctionnement des grandes failles
du secteur de Popayan.
l. X1. _I -- _-- _I ,_.+ --

125

TECTONIQUE
127

LATE JURASSIC RIDGE-TRENCH COLLISIONS AND DEVELOPMENT OF GULF


OF CALIFORNLA-TYPE BASINS ALONG THE ANDEAN CORDILLERA

Bryan C. Storey and Tony Alabaster

British Antarctic Survey, Natural Environment Research Council, High


Cross, Madingley Road, Cambridge CB3 OET, UK

R&urn6

De nouveaux r&ultats des prospections g6ologiques et geochimiques


de La G6orgie du Sud, qui fait partie deplacde du bassin marginal des
Rotas Verdes des Andes, suggerent que le bassin sest form6 dans un
milieu ressemblant celui du Golfe de la Californie. En con&quence, il se
peut que des collisions entre une systime de dorsales et une fosse
oceanique aient dirigi5 lorogenese des Andes dune facon non reconnu
jusqu8 present.

Key Words: Andean orogenesis, marginal basin, strike-slip

The tectonic evolution of the Andean Cordillera is intimately related


to long-lived subduction of Pacific and proto-Pacific oceanic lithosphere
beneath the South American margin during both the Mesozoic and
Cenozoic. Although broadly a convergent margin, repeated continental
lithospheric extension has resulted in intra-arc and back-arc basin
formation. Basin evolution is generally related to variations in
subduction parameters; namely: convergence rate; roll back; age and
hence thermal characteristics of the subducting oceanic slab; and
subduction zone angle. Surprisingly, one parameter rarely considered in
Andean orogenesis is the role of ridge-trench collision, notwithstanding
the current collision between the Chile Rise and Peru-Chile trench and
consequent formation of the Golf0 de Penas-Taitao basin (Forsythe and
Nelson, 1985). It is likely similar collisions occurred during the Mesozoic
and may have been an important, previously unrecognized control on
Andean orogenesis and ensialic marginal basin formation.
128

Although the Late Jurassic to Early Cretaceous Rotas Verdes


ensialic marginal basin system in the southern Andes (Fig. 1) is generally
considered to be a subduction-related back-arc basin, its tectono-
magmatic evolution has also been compared with the Gulf of California.
New field mapping, structural data and an integrated trace-element and
Nd isotopic study on the Larsen Harbour Complex on South Georgia (Fig.
l), a displaced part of the Rotas Verdes, suggest basin formation resulted
from oblique spreading in a transform system due to intersection of a
spreading ridge with a trench .(Alabaster & Storey, in press). Magmas
produced during the early stages of continental lithospheric attenuation
were derived by varying degrees of partial melting and fractional
crystallization from a low-ENd lithospheric mantle source enriched in
large ion lithophile elements (LILE) during an earlier phase of
subduction-related talc alkaline magmatism. Petrogenesis of these magmas
does not require contemporaneous subduction-related hydrous melting in
the mantle source and consequently we suggest the basalts were not
generated in a supra-subduction zone setting. Interestingly, basalts
generated during the early stages of rifting both from the Rotas Verdes
and the Gulf of California are essentially indistinguishable geochemically.
Moreover, in both areas, magmas produced during later stages of rifting
were derived from a high-ENd asthenospheric mantle source similar to N-
type MORB, unaffected by earlier LILE-enrichment.

Additionally, we have discovered geochemically unusual pre-


ophiolitic continental high-magnesium andesite (CHMA) dykes on South
Georgia and CHMA lavas on the once contiguous Antarctic Peninsula.
Studies based on the occurrence of Early Miocene age CHMA lavas in Baja
California and of Middle Miocene age in the Setouchi volcanic belt, Japan
(Saunders et al., 1987, Tatsumi and Maruyama, 1989) suggest CHMA form in
response to high geothermal gradients associated with atypical
subduction. These conditions may, in part, be generated by
encroachment of a spreading center upon a trench and subduction of
newly created, hot, oceanic lithosphere.

Structural data from a one kilometre wide ductile shear zone that
dissects the island of South Georgia indicates a major component of
strike-slip displacement, Strike-slip and oblique faults have also exerted
an important control on Andean tectonics (Dala Solda and Franzese, 1987).

We propose a model for the formation of the Rotas Verdes basin


which involves cessation of subduction along an oblique-slip margin as a
result of ridge-trench collision, and the migration of a ridge-trench-fault
triple junction forming a growing transform boundary (Fig. 2). The
change from subduction to strike-slip tectonics allowed local transtension
and upwelling of hot asthenospheric mantle. Opening of the marginal
basin may have resulted from instability in the triple junction due to an
embayment along the coast. This instability could have resulted in the
transform boundary migrating inland from the continental margin and the
carving off of a continental slice by oblique spreading in a transform
system. The present- day orocline in southern South America may be, at
least in part, an original feature that caused the triple junction
instability and opening of the basin. The basin was subsequently infilled
by volcanic detritus derived from an inactive remnant arc, whereafter
subduction resumed resulting in basin inversion and uplift.
.I ___ ..__.._I .,

129

The extent to which ridge-trench collision influenced the formation


of elongate basins of Late Jurassic to Early Cretaceous age along the
entire length of the Andean Cordillera is, however, unknown.
Interestingly, Atherton and Webb (1989) have suggested that formation of
the Huarmey Basin in the western part of the Mesozoic Peruvian Trough
was not subduction-related, but formed in response to crustal extension
similar to that postulated for the Gulf of California. It is therefore
possible to conclude that the tectonic evolution of the Andean Cordillera
has been influenced by oblique tectonics, migrating triple junctions and
ridge-trench collision; a scenario which has important implications for
Andean-type orogenesis, continental lithospheric extension and ophiolite
formation along convergent margins.

Alabaster, T. and Storey, B.C. Geology (in press).


Atherton, M.P. and Webb, S. Journal of South American Earth Sciences, 2,
241-261,1989.
Dalla Salda, L.H. and Franzese, J. Revista Geol6gia de Chile, 31, 3-13, 1987.
Forsythe, R. and Nelson, E. Tectonics, 4, 477-495, 1985.
Saunders, A.D., Rogers, G., Marriner, G.F., Terrell, D.J. and Verma, S.P. in
Tectonic Controls on Magma Chemistry (eds Weaver, S.D. and
Johnson, R.W.) 223-245, 1987.
Tatsumi, Y. and Maruyama, S. in Boninites and Related Rocks (ed.
Crawford, A.J.) 50-71, 1989.

Figure 1. Location map showing position of south Georgia relative to


Rotas Verdes (RV), South America and Antarctic Peninsula.
Map of South Georgia showing the ophiolitic rocks of the
Larsen Harbour Complex and the strike-slip fault zone
(Cooper Bay dislocation zone, CBD).
.-l--ll.. ._ -- _.l_,_-

130

Figure 2. a, Present-day tectonic setting of Baja California (BC).


NA, North America plate; PAC, Pacific plate; GP, Guadalupe
plate; RP, Rivera plate; CCS, Cocos plate; JB, Jalisco block;
MAT, Middle American Trench; EPR, East Pacific Rise;
MFZ, Molakai Fracture Zone. b, Proposed reconstruction for
part of the proto-Pacific margin of Gondwanaland during Late
Jurassic times, illustrating a Gulf of California type model for
the
131

NOUVELLES DONNEESMONTRANT UNE REMISE a ZERO GENERALE DE


L'HORLOGE ISOTOPIQUE K/Ar DANS LA PARTIE NORD DE LA
BAIE DE L'AMIRAUTE, ILE DO ROI GEORGE, ANTARCTIQUE.

E. SOLIANT Jr. * et M. G. BONHOMME **

* Institute de Geocihcias da UFRGS,


Av. Bento Gongalves, 9.500
(CEP 91.500) PORT0 ALEGRE, RS - BRASIL.
** Institut Dolomieu, URA 69 CNRS,
F-38031, GRENOULE - FRANCE.

Dans la Baie de l'Amiraut&, Ile du Roi George, Penin


sule Antarctique, la Formation Znosko'Glacier affleure entre
Mackellar Inlet et Ezcurra Inlet. Elle est constitude d'un
ensemble volcano-sbdimentaire dont l'dge etait r&put& meso-
zoique.
Des datations Rb/Sr ont montrt5 que son age de mise
en place est en fait pal&oc&ne (environ 60Ma). Les datations
potassium-argon deja publikes dans ia litterature s'btalent
de 67 a 26 Ma, dans la region de la Baie de 1'Amirautd.
Dans le but de r&soudre A la fois cette difference
entre les Ages Rb/Sr et K/Ar et la dispersion des don&es
K/Ar, de nouvelles datations K/Ar ont 6th effectuees en
roche totale.
Le diagramme dK/'"AArvs '"Ar/36Ar rdv&le une droite
isotopique dont l'intercept est proche de celui de l'argon
atmosphCrique pour un bge de 42,Ma.
Cette date eat attribuee a une althratfon generali-
s&e observee dans toute la rhgion. Ceci se traduit par une
carbonatisation et une chloritisation tr&s importantes, d'au
tant plus fortes que l'on s'approche des systemes faillks.et
plissds le long de failles en dkzrochement. Cette altdration
remet a zero le chronomktre K/Ar sans apparemment altkrer le
chronometre Rb/Sr. La relation entre cette altkation et la
tectonique conduit A modifier l'histoire stratigraphique
et geodynamique de la region,
133

TECTONIC TRANSPRESSION ALONG THE SOUTHERN SEGMENT OF THE


ATACAMA FAULT-ZONE, CHILE

M. Pincheira, R. Thiele2, L. Fontbotb

MineraIogisch-Petrographisches Institut. INF 236. 6900 Heidelberg, West Germany; ZDepar-


tamento de-Geologia y Geofisica, Universidad de Chile, Casii 13518, Corm 21, Santiago,
Chile

Los an&is cine&tico y de edmzodeformaci6n de las megafallas Los Colorados y La


So&a-Huantemi, Ias cuales son representativas de1 segment0 sur de la Mona de FaUa Ataca-
ma, sugieren una din&&a de falIamiento transpresivo, con movimiento si&tral, asociada a
la evoluci6n de1 margen Pa&co de. Sudamtica. Ambas me&&s son cl resultado de un
c&llamiento regional, convergente, primario, se&n una direcci6n N20E
Key \Vor&Te&nie transpression, Atacama Fault, Lower Cretaeeous, Chile

Introduction. This paper analyses the generation 8nd evolution of an area, considered
to be representative of the southern segment (28-293) of the Atacama-Fault-Zone (AFZ).
The studied area is located at the eastern flank of the Coastal Range, where the AFZ con-
trol the distribution of the major iron-mines of the region.

The region comprises volcanic and sedimentary rocks of Lower Cretaceous age. These
sequences were deposited at the transitional zone between a volcanic-arc and a back-arc-
basin. Overlaying and interfingcring from volcanic and calcarcous rocks is usual. The Lower
Cretaceous paleogeography appears to have been controlled by crust thinning processes
associated to probable extensional fractures which subsequently led into the AFZ. The
Mesozoic rocks from the region are deformed and intruded by granitoids of Valanginian-
Aptian ages. These units host the largest iron mines of the country. At the studied sector a
spatial coincidence between the sheared zone and the actinolitic alteration band (mainly
affecting intrusive and volcanic rocks) associated to the iron mines is observed. The geolo-
gic evolution of the region allows to establish that tectonic, magmatic, and metallogenic
episodes discussed in this study took place close to each other during Lower Cretaceous
and extended probably up to early Upper Cretaceous.

The analyzed fault-zone represents a periodically reactivated structure. It forms a line-


al-megastructure defined by an anastomosing system of fractures and folds, which divides
the region into two separate north-south blocks. West of the AFZ the sequences have been
134

deposited in a transitional range between a volcanic-arc and a back-arc-basin. The calca-


reous sequences east of the AFZ represent a marine platform environment. The sheared
zone displays an intrincate system of tight folds and echelon faults which contrasts with the
relatively simple deformation style of the Mesozoic sequences characterized at regional sca-
le by a wide N-trending synclinorium.

Megastructures and associated sinistral transcurrent fault system. The megafault Los
Colorados extends at least from the Boqueron-ChaAar and Los Colqrados mines to south
of the Huasco river (Fig.1); La Sosita-Huantemt fault zone extends from the
Chtiar-Quemado mine to the south through the mines La Sosita, Huanteme, and El
Algarrobo. The fault zone is usually 20-30 m thick, occasionally, as at the
Boqueron-Chafiar mine, it can reach up to 600 m thickness. At several points the faults cut
and limit Mesozoic intrusive rocks.

Different types of cataclastic rocks, including protomylonites and mylonites, have been
found. They develop on sedimentary, volcanic, and plutonic rocks. The occurrence of flow
textures in deep plutonic rocks suggests generation at relativelly high temperature allowing
ductile flow and partial recrystaltization. Despite the subhorizontal shear pattern, all invol-
ved rocks exhibit vertical slickensides which correspond to later vertical motions. As a
result of the vertical movements the western
Fig. I: Fault pattern at the studied zone:

Ri Riedels faulu, P= Pshean. n main iron deposits:


1= Doqu&n Chariar. 2= Los Colorados. 3 = Chatiar Quema-
do; 4 = Sosiws. 5 = HuaniemC. 6 = El Algarrobo.
136

of the tectonic transpression, sinistral transcurrent faults sheared ore and host rocks.
During early Upper Cretaceous shear movement with features of ductile-rigid regime persi-
sted. Finally, during later Upper Cretaceous and Tertiary, vertical displacements under
rigid-rupture conditions took place, which are illustrated by vertical slickenside surface.

_ Strain-stress analysis. Transpression experimental studies show that along shear zones
folds, thrusting, and fault systems develop according to a determinate model (Harland,
1971; Naylor et al., 1986). Related to simple shear folds and faults develop with a typical
en echelon pattern, the orientations of which show a rotational effect with respect to the
sheared zone in function of the deformation grade. An anastomosing system of longitudinal
faults consisting of several generations of synthetic Riedel (R) and associated P shear faults
develops in a final stage. This model can be applied for the present study. The lineal mega-
structure shows effects of horizontal compression, and the distribution of the associated
sinistral structures indicates that the whole fault system represents a sinistral shear zone.
Reference points that could indicate the magnitude of the horizontal displacement were not
recognized.

Origin of the deformation. The analyzed structures belong to a transpressive tectonic


stage associated with the openning of the South Atlantic Ocean, about 130 Ma (Popoff,
1988). At that time the Pacific continental margin was affected by horiiontal stress of great
magnitude, which could originate shear zones as the AFZ. The principal stress direction
must have had a similar direction to that of the lithospheric convergence between the Paci-
fic Plate and the South American Plate. The AFZ probably started as a extensional fracture
zone at the west margin of the Neocomian back-arc basin. Subsequently; between 110 and
85 ,Ma, i.e., as the expansion rates of the Pacific lithosphere rose up to 18 cm/year (Frutos,
1981), the distensive regime at the back-arc domain was replaced by compresive regimes in
accordance with Uyedas (1982) high stress Chilean subduction model (Fig.2). Subsidence
in the back-arc domain ended and tectonic transpression was favored.

FKUTDS. 1. (19SI)Tcaonophysia. 72. p. T21-T32.

I IAKLAND. W. (1971) GeologicalMagatinc. Vol. 108, N-1. p. 27-41.

MOMECINOS. D.P. (I~83)These. Don In&. Univ. Paris Xl (Orsay). 191 p.

NARANJD. J.A.; IIERVE. F.; PRIET0.X.; MUNIZAGA. F. (1984) Rcvisu Comunicocionn.34. p. S7.66.

NAYLOR. MA.; MANDL. G.; SIJPESI-EIJN. C.H.K. (1986) Journal of Suuchwal Ceolo~. Vol. 8. p.
737.7S2.

PICHON. R. (198l)Thae. DOCL3cmc Cycle, Univ. Paris Xl (Orray). 326 p.

POPOFF. hf. (1988) Journal olAfricsn Earth Sciences.Vol. 7. No. 2. p. 409.431.

UYBDA.S. (1982)Tcctonophysia. N. 81.~. 133.159.

UYEDA, S. (1983) Episodes.Vol. 1983.2: 19.24.

ZENTILLI, hl. (1974) Ph. D. The&. QueensUniv., Gnada. 394 p.


137

RNTOFRGRSTR - SALT4 - CHCICO PLAINS TRANSECT, CENTRAL CINDES.

R.H. Omatinit & H.G. Go&tell

With Contribution5

Geoi ogy: R. Alonsol, G. ChonQ. D.#, T. Bogdanict, R.H. Dmarini),


K.J. Router**, H. Soto. P.W and J.G. Vitamontel.

Sei smi cs: tl. Clranedat#, A. Carlo., P. GirselS, W. HeinsohnSS,


C.E. Di Persia*, H. SchmitzLI, 3.6. Viramontet and P.
WiQQettt.

Gravi t j: ?I. AranedaYt, A.CI. Cerrato@*, H.J. Goctretl, A.


Introcaso+, B. Lahmeyerttt ,S. Schmidt8t and S. StrunkLI.

Magnetotel lurics: J. Febrer++, D. KriiQerZ8, V. Rathtt and G.


Schwurbt.

Draft: R.R. Battapiiat.

1 Univerhidad National de Salta -- CONICET (Argentina)


rCI Freie Universitit Berlin (Fed. Rep. of Germany)
Stt Technische IJniversitat Clausthal (Fed. Rep. of Germany)
# Universidad del Norte, Wtofaqasta (Chile)
## UniLersidad de Chile, Santiago (Chile)
+ Universidad National de Rosario - CONICCT. (Argentina).
+* Instituto Espacial San Miguel - Buenos Aires {Argentina)
0 Yacimientos Petroliferos Fiecales, Buenos Aires. (Argentina)
nJ Universidad National de Buenos Aires - CONICET. (Argentina).

RESUtlEN

En el prssente trabaJ0 se eintetiia 105 re5ultados de la5


investlyaciones geolbcjicas y CjeOf i eicae rcalizados en el Norte de
Chile y Argentina CU+Y Finalidad fue obtencr informacidn sabre la
estructura de la lit6sfera Andina. Sobre el model0 litosf4rico
real i 2:ado a lCj:i;11)O1.3. se evaluan 1 as uni dadee tectcnicas \r
morfocstructursles de acucrdo a 5u actual asantamiento tectbnico.

INTRODUCTION

Geol ogi cdl and Geophysical investigation5 carried out by


Global Transect Pro.icct (IGL - GGT) and Proyecto kgenti no de
Transectas (CAPLI - CONICET) at 24O Lat. S. between 1985 - 1989
are shown. The Transect is 100 Km wide and spreads from the
Pacific Ocean to the Chaco Plains across the findean and Subandean
bel t3.
REGIONAL INDEX MAP 6~

PO-

TRANSECT
139

This work sumarizes the present knowlodge on the subjet but


some of the hypothesis we have reached to are subject for further
discussion, they are not meant to be considered definitive
conclussions.

The Geological-geophysical transect

The main morphoestructural units developed in the western


part of South America are shown in Fig. I. The most important
gaological features in the region is the Anden Cordillera ant its
magmatic arc development as a result of the interaction of the
South America and Pacific plates.
Hany genetical hypothesi s have been postulated among wi ch
ensialic origin of its Precambrian-Paleozoic basement,
lithospheric collision, combined models and al lochthonous terrane
accretions. The latter has not been verified by this 5tudy;on the
contrary a lot of distensive stage with rift basin generation and
subdue t i on processes along the Pacific margin of South America
were detected. Upper Precambrian (Puncovi scana Fm), Ordovicic
(Santa Rosi ta Fm) , Devonian-Carboni f ercus and Cretaci c-Eogenic
(Salt.5 Group), are the main rift stage development. On the other
hand, the active margins, have been proved to exist on 1 i since
the PQrmotrianic times to the Present times.
The lithorpheric model shows two contrasted areas located to
the East and to the West of Salta City meridian. The eastern
set t or shows a tectonic quiotnoss during the Phaneroooic with no
i ,mpor t ant di strophic processes that disturbed stratigraphic
comf ormi t.f of the stable pi atf orm 5Qdimentary sequences. Only
the narrcw Subandean belt shows thrust, overthrust and strong
foldings.
The western side, shows great deformation, important
shortening and crust thickening.
The Andean Orocl i ne, has a thick crust (65-70 Km). Its model
is from geological, petroloqical, gravimetrical, sei smi cal and
aagnetotel luri cal data.
Petrological data from the lithospheric xenolites included in
the Cretocic al kal ine basalt, shows that, at least till the
tretacic, basic and acid granulites built up the I ower crust
while the upper crust was formed by sillimanitic gnei ses,
biotitic schist and sediments. The successives development of
magmatic arcs from the Jurasic to the present timQ5, has
certainly modified this structure in the Andean Belt. It should
still be present under the Pastern side of the belt (Chaco
Plain3). Some of the seismic, magnetotelluric and gravimetric
anomal i es, have been interpreted as a consecuence of this crustal
changes.
140

!:

i
I
i

%
l-

b
F-

. .
I: x
I I
,._,.-._._-_

141

TECTONIC EVOLUTION OF THE CHILEAN MAIN CORDILLERA


BETWEEN 33 AND 35 SOUTH LATITUDE

Roynaldo Charrier', Hicardo Thielc*, Nodrigo Arcos** a II


3
Felipe Malbran***

*Depto. de Ceologia y Geofisica, Univ. de Chile, Casilla


13518, Santiago.

**Empress National de1 Petrbleo, CompaRia 1065, Santiago.

l*fiGeoestudios, G&nova 2095, Santiago.

Resumen

Las unidades (Jurisico a Presente) se acumularon en una


cuenca de tras-arc0 en el Mesozoic0 y en sucesivos arcos
volc&nicos desplazados hacia el este en el Cenozoico. Existen
cinco pisos estructurales cuyo plegamiento se debib a feno-
menos compresivos asociados a episodios de tect6nica global.

Key words: Central Andes, Chile, Stratigraphy, Paleogeogra-


phy, Tectonics, Structural Geology.

Introduction

This area is located in the southern part of the Central


Andes, inmediately S of the flat-slab subduction segment
(Fig. 1). It is one of the most studied areas in Chilean
Andes and is representative for the Main Cordillera in cen-
tral Chile. It, thus, provides a good basis to understand the
evolution of this andean region.

We summarize the geology and stratigraphy of the arca and


discuss the paleogeography, tectonic setting and deformation
mechanisms for each of its evolutionary periods.
142

Stratigraphy

The stratigraphic record (Fig. 2) consists of rock units


ranging in age from Middle Jurassic to Present. The age of
the mainly sedimentary mesozoic part is well constrained on
the basis of the ammonite content of the marine intcrcala-
tions. The age of the Cenozoic continental mainly volcanic
units is based on isotopic determinations. A strong and wide-
spread low temperature alteration of the volcanigenic units
precludes a more precise chronology of the Cenozoic units.

Paleogeography

During the evolution of this area two basically different


paleogeographies Here recognized: 1. a back-arc domain with
the development of a strongly subsident mainly marine basin
during Mesozoic, and 2. arc domains with the development of
intra-arc basins or small continental back-arc basins, or arc
domains located in graben depressions during Cenozoic.

The back-arc deposits accummulated during two transgression-


regression cycles: 1. Late Liassic-Bajocian to Kimmeridgian,
and 2. Tithonian to Neocomian. The basin was located to the
east of an emmerged and active volcanic arc and seems to have
been connected through narrow openings in the arc with the
paleo-Pacific Ocean. The presence of the arc caused an
eastward decrease in grain coarseness and volcanigenic con-
tent in the back-arc deposits. The development of carbonatic
platforms on the west margin of the basin during Callovian
and Tithonian-Neocomian was considerably inhibited by the
abundant detritic material supplied by the arc. Neocomian
volcanism in this area has a talc-alkaline nature.

Exept for a possible marine intercalation from the Atlantic


side in the lower Abanico=Coya-Machali Fm., which is suspcc-
ted to be of Maastrichtian-Danian age, the Andean domain was
totally emmerged during the Cenozoic. The volcanic ccnozoic
units are located in regions that formerly were part of tne
back-arc domain. They correspond to at least three volcanic
arcs, respectively shifted to the east of the previously
existent arc. The eastward migration of volcanism and of the
batholitic bodies in this area is a characteristic feature of
the Chilean Andes. Cenozoic volcanism is bimodal and of calc-
alkaline nature.

Tectonic Evolution and Interpretation

The development of a strongly subsident back-arc basin in the


Ilesozoic was interpreted as the result of a weak coupling of
e. . . .,l *-, , --._-.

143

Fig. 1. Location map. a. Main


subdivision of the Andes. b.
I ;
Morphostructural units be- wm~ t I

tween 33-35 S. c. Studied


area (stippled). PO: Pacific Fig. 2. Stratigraphic column
Ocean, CC: Coastal Cordillera, of the studied area with in-
CD: Central Depresion, HC : dication of the marine trans-
Main Cordillera, FC: Frontal gression - regression cycles
Cordillera, PC: Precordillera, and main paleogeographies
AF: Andean Foreland. s.1.: developed. U: unconformity;
approximate limit of flat-slab v: vertebrate fossils.
subduction segment.

the colliding plates due to steep-slab subduction. This could


have caused an extension tectonic regime in the back-arc
domain. The absence of a well developed back-arc basin during
Cenozoic was interpreted with a similar reasoning as the
result of a more flat subduction (Chilean-type).

Deposition during the Mesozoic was continuous. An unconformi-


ty between mesozoic and Cenozoic deposits was only mapped at
340 30s. During Cenozoic at least three unconformities sepa-
rate the continental volcanic and detritic units (Fig. 2).

The transgression-regression cycles during Mesozoic do not


coincide with global sea-level fluctuations. They can rather
be interpreted as the result of global diastrophic episodes
like variations in sea-floor spreading rates and the dif fe-
rent stages of the openning of the S-Atlantic.

The kimmeridgian deposits form a thick wedge that interrupts


the marine sedimentation. The absence of unconformity between
both marine cycles and the existence of an important jurassic
magmatic activity to the W suggests a thermally induced
uplift of the continental margin, possibly with the beginning
of subduction during the rift stage of the S-Atlantic. This
145

CHRONOLOGY OF FORElAND DEFORMATION IN THE PRECORDILLERA - SIERRAS


PAMPEANAS REGION OF ARGENTINA (2533%)

J.H.Revnolds, T.E. Jordan2, K. Tabbutt3, G. Re4, J. Vilas4, F. Bercowski5, J.P. MilanaS, J.A.
Beer26, J.F. Damanti2@, and V.A. Ranrosa

1 Department of Earth Sciences, Norwich University, Northfield, VT 05663, USA

2 Dept. Geological Sciences & INSTOC, Cornell University, Ithaca, NY 14653 USA

3 Department of Earth Sciences, Dartmouth College, Hanover, NH 03755, USA

4 Depanamento de Ciencias GeoMgicas, Unlversidad de Buenos Aires, Buenos Aires


1426, Argentina

5 Facultad de Ciencias Naturales, Fisicas y Exactas, Universldad National de San Juan,


5400 San Juan, Argentina

6 Pecten Oil Company, P.O. Box 205, Houston, TX 77001, USA

7 Department of Geology, Union College, Schenechtady, NY 12306, USA

Resume

Se determinaron las edades de la deformacidn de la corteza en la zona de 10s Andes


centrales par debajo de la cual existe la subducci6n de bajo angulo. En estratos en las cuencas de
antepafs se han medido edades por magnetoestratigraffa y por trazas de fisi6n en tobas.
DeformacMn en la Precordillera comenz6 hate 16 millones de ahos, y esta deformacibn y la
deformaci6n de las Sierras Pampeanas eran activos juntas en los riltimos 6 millones de anos.

Key Words: foreland basins, fission-track ages, magnetostratigraphy, Precordillera, Sierras


Pampeanas, flat subduction
146

Methods

Recent investigations resuit in a chronology of deformation in the Central Andean foreland


above the zone of flat subduction of the Nazca Plate in northern Argentina. The chronology is
based on a relatively prectse chtonostratigraphy in Neogene detrltal strata in several pans of the
Precordiltera-Sierras Pampeanas region. The data allow minimum rates of uplift to be assigned to
some ranges In the area. The new data reveal a rapkf evolution of the Central Andes which was not
discernible from traditional regional tithostratigraphy.
The foreland basin strata are dated usfng magnetic polarity stratigraphy and fission-track
dating of zircons retrieved from interbedded volcanic airfall beds. Tectonic events are constrained
either directly by dating cross-cutting relations or minimum ages are determined from changes in
conglomerate and sandstone ctast tiihotogies. Some events are dated, with less precision, using
indirect methods such as dating fluctuations in sediment accumulation rates and/or changes in
deposltional environments. Additional chronologic information is available from the rare Neogene
volcanic centers located in the study area.

Results

The data reveal that after the initial uplift of the Cordillera Frontal prior to 118 Ma, the upper
crustal locus of overthrusting migrated in a systematic west to east progression from 118 Ma to the
present in the Precordillera fold and thrust belt. At some time prior to - 8.2 Ma, crystalline
basement blocks began to rise along moderately dipping reverse faults in the northwestern Sierras
Pampeanas. Uplift of these ranges was less systematic; there is as yet no suggestion of a preferred
direct&n for the migration of tectonic activity in the basement uplift region. Considerable young
uplift (< 4.3 Ma) appears to have been focused on the western side of this tectonic province.
Minimum uplift rates (based on topographic relief) of several Pampean ranges are calculated
to be between 050.8 mm/a, over the past 4-7 million years. These numbers indicate a rapid rise of
the Laramide-style block uplifts contemporaneous with low-angle thrusting in the eastern part of
the Precordillera. The data suggest that foreland deformation is closely synchronized with the
initiation (-18 Ma) and propagation of flat subduction in the region.
147

ARC TECTONICS AN0 THEIR APPEARENCE AT DIFFERENT STRUCTURAL


MAGMATIC
LEVELS: EXAMPLES FROM NORTHERN CHILE

Ekkehard Scheuber and Klaus-J. Reutter l

* Institut fiirGeologie, Freie Universitat Berlin,


AltensteinstraBe 34 A, D-1000 Berlin 33, F. R. G.

Resumen

Los movimientos entre placas convergentes controlan la tectdnica de1 arco


magmatdtico. Se producen estructuras de acortamiento sin vergencia o con
vergencias bilaterales junto con fallas transcurrentes paralelas. Se de-
scriben estas estructuras para diferentes pisos tectdnicos.

Key Words: Magmatic arc tectonics, Central Andes, northern Chile

Introduction

Plate convergence has played a major role in the structural and


magmatic evolution of the Central Andes of northern Chile since the be-
ginning of the Jurassic (Andean Cycle, Coira et al. 1982). In a system of
converging plates the greatest amount of deformation is probably accommo-
dated in the shear zone between the two plates. However, the crust of the
upper plate may also be deformed. Besides the backarc area and the sub-
duction complex of the forearc, the magmatic arc is one realm that, con-
temporaneously to arc magmatism, may suffer very strong deformation
including structures of orogen-normal shortening and/or orogen-parallel
strike-slip (Reutter et al. 1988, Reutter & Scheuber 1988). In northern
Chile and adjacent areas, this interrelation can be studied effectively
as the magmatic arc shifted towards the interior of the continent during
the Andean Cycle. This eastward shift resulted in four arc stages: (1) a
Jurassic - Early Cretaceous arc in the Coastal Cordillera, (2) a Mid Cre-
taceous arc, not well documented, in the Longitudinal Valley, (3) a Late
Cretaceous - Paleogene arc in the Chilean Precordillera (Sierra de Mo-
reno, Cordillera Domeyko), and (4) the Miocene - Holocene arc in the We-
stern Cordillera, with extensions of magmatic activity reaching into the
Eastern Cordillera. The termination of the fossil arc stages was connec-
ted with an uplift of the arc's crust in the order of several km. This
uplift allows the investigation of different structural levels of the
148

magmatic arc's crust. The Jurassic - Early Cretaceous arc was upplifted
up to 12 km, the Late Cretaceous - Paleogene arc up to 5 km, whereas the
Miocene - Holocene arc shows near-surface structures. According to the
tectonic level an infrastructure where ductile flow is the prevailing de-
formational mechanism can be distinguished from a suprastructure with
folds and brittle faults.

The infrastructure

In the infrastructure a fully ductile level can be distinguished


from a semiductile level. The criterion for this distinction is the oc-
currence of S-C fabrics which are absent under fully ductile conditions
(Shimamoto 1989).

a) the fu77y ductile level

The deepest structural level crops out in the North Chilean Coastal
Cordillera NW and S of Antofagasta. Its lithology is built up by basic to
intermediate igneous rocks, to a minor amount also Paleozoic to Lower Ju-
rassic sediments are involved. The structures are characterized by more
or less distinct vertical foliations parallel to the orogen. The foliated
area reaches a width of some 15 km. Deformation took place under amphi-
bolite facies conditions at low pressures (500-6OO'C, <300 MPa, Scheuber
& Andriessen 1989), there are, however, also granulitic portions (RBRling
1989). Vertical folds and bending of foliation in shear zones indicate
that left-lateral strike-slip movement caused the deformation in this
deep level.

b) the semiductile level

Here the deformed zone is narrower than in the fully ductile level.
Deformation occurred in a mylonitic zone up to 2 km wide which is charac-
terized by a strong strain partitioning on the mesoscopic and microspopic
scales. Mesoscopically large strains are accommodated by mylonitic to ul-
tramylonitic bands which are separated by moderately deformed protomylo-
nites. On the microspopic scale C- and S-bands accommodated a major
amount of strain. The semiductile shear zones are clearly related to oro-
gen-parallel fault systems such as the Atacama Fault Zone of the Coastal
Cordillera (Scheuber & Andriessen 1989) or the West Fissure in the Pre-
cordillera (Reutter & Scheuber 1988). The mylonitic rocks were formed at
a depth of 5-7 km. In contrast to the fully ductile rocks, they contain
several indicators for the sense of shear which can be related to the re-
constructed direction of the motion of the subducting oceanic plate rela-
tive to S America. In the mylonites of the Coastal Cordillera the gene-
rally observed left-lateral sense of shear corresponds with the great ob-
liqueness of the (oceanic) Phoenix plate's motion (Zonenshayn et al.
1984). The fabric patterns of the Precordilleran mylonites are generally
more symmetrical or show a dextral sense of shear, they reflect a smaller
convergence obliqueness during the Paleogene.
149

The suprasstructure

This level is structurally characterized by orogen-parallel anti-


clines and steep faults which developed contemporaneously to the igneous
activity of the particular arc. Anticlines, in the core of which the pre-
Andean basement is always involved are characteristic of the Late-Creta-
ceous - Paleogene arc of the Precordillera; usually the anticlines are
compressed to such a degree that the limbs are steep or overturned and
the core is upthrust with respect to the limbs. Vergencies are developed
to the W and to the E. The flanks and especially the cores are frequently
intruded by shallow plutons which can in part be considered as synkinema-
tic. Structures of orogen-normal shortening are also found in the modern
magmatic arc of the Western Cordillera and Puna. Here conjugate reverse
fault systems with vergencies to the W as well as to the E are developed
(Schwab 1970). Structures of Neogene to Quarternary folding are also fre-
quent. Orogen-parallel strike-slip is revealed by en echelon fault arrays
(e.g. Riedel shears), vertical folds or stratigraphic discontinuities in
the suprastructure of the Late Cretaceous-Paleogene and in the modern
arcs. The sense of shear is dextral in the Paleogene, sinistral during
the Miocene-Quaternary.

Conclusions

The most characteristic feature of magmatic arc tectonics is the


coincidence in time and space between deformation and arc magmatism. Each
of the the four arc stages had a duration of several tens of million
years so that a pervasive heating of the crust and a corresponding high
geothermal gradient of about 6O'/km could develop. Deformations in the
magmatic arc are a response by the crust of the upper plate to the forces
generated by plate convergence. Due to the heating of the crust the de-
formations are focussed to the magmatic arc. Liquid bodies contained in a
crust which is subject to deformation reduce the differential stress to
zero and thus perturbate the regional stress field. In the heated crust
the brittle-ductile transition, which in quartz-rich rocks corresponds to
the 300'Gisotherm (Sibson 1977), takes place at a higher level than it
does under normal crustal conditions. In the arc stages of northern Chile
this transition was located at a depth of about 5 km. Creep processes and
the lack of strain hardening which can be assumed for the fully ductile
level also gain significance at the rather shallow depth of about
8-10 km. A significant drop in differential stress and an increase in
strain rate of the magmatic arc can thus be assumed. Magmatism also plays
an important role in the relation between infrastructure and suprastruc-
ture. Chong & Reutter (1985) proposed a model in which the deformations
in the suprastructure were thought to be triggered by intrusions that de-
stabilized the crust to such a degree that continuous compressive stres-
ses led the upper rigid level of the crust to shear off from its infra-
structure.
Magmatic arc tectonics reflect the relative motion of the conver-
ging plates. In case of oblique motion the vector is resolved into two
components, one acting perpendicularily and one parallel to the plate
boundary. The first leads to structures of orogen-normal shortening, the
second to orogen-parallel strike-slip motion (trench-linked strike-slip
faults Woodcock 1986). If both types of structures occur together, the
tectonic regime may be described as transpressive. There is no evidence
150

that the structures of crustal shortening are a result of far reaching


intracrustal overthrusting. It can be interpreted as distributed pure
shear crustal thickening.
A long time span for magmatism is necessary for a large-scale hea-
ting of the crust. Deformation, however does not seem to be continous but
it is accelerated during phases of increased spreading rates (Pilger
1984). This suggests that at low convergence rates the major amount of
stress is absorbed in the subduction zone and that only at high plate ve-
locities stresses are produced that are sufficent to generate deformati-
ons in the upper plate. If such high stresses are produced, they are pre-
ferably absorbed in the magmatic arc.

References

Chong, G. & Reutter, K.-J. (1985): Fenomenos de tectonica compresiva en


las Sierras de Varas y de Argomedo, Precordillera Chilena, en el
ambito de1 paralelo 25' sur.- IV. Congr. Geol. Chileno Actas 2: 2-
219 - 2-238, Antofagasta.
Coira, 8.; Davidson, J.; Mpodozis, C. & Ramos, V. (1982): Tectonic and
magmatic evolution of the Andes of northern Argentina and Chile.-
Earth-SC. Rev., 18: 303-332.
Pilger, R.H. (1984): Cenozoic plate kinematics, subduction and magmatism:
South American Andes.- J. Gee, Sot. London, 141: 793-802.
Reutter, K.-J. & Scheuber, E. (1988): Relation between tectonics and mag-
matism in the Andes of northern Chile and adjacent areas between
21' and 25-S.- V. Congr. Geol. Chileno Actas 1: A345-A363, San-
tiago.
Reutter, K.-J.; Giese, P.; GCitze, H.-J.; Scheuber, E.; Schwab, K.;
Schwarz, G. & Wigger, P. (1988): Structures and Crustal Development
of the Central Andes between 21' and 25'S.- In: Bahlburg, H.;
Breitkreuz, C. & Giese, P. (eds.): The Southern Central Andes, Lec-
ture Notes in Earth Sciences, 17: 231-261.
RiiSling, R. (1989): Petrologie in einem tiefen Stockwerk des jurassischen
magmatischen Bogens in der nordchilenischen Kustenkordillere siid-
lich von Antofagasta.- Berliner Geowiss. Abh. A-112, pp.73
Scheuber, E & Andriessen, P.A.M. (1989). Kinematic and geodynamic signi-
ficance of the Atacama Fault Zone, northern Chile.- J. Struct.
Geol. in press.
Shimamoto, T. (1989): The origin of S-C mylonites and a new fault-zone
model.- J. Struct. Geol., 11: 51-64.
Schwab, K. (1970): Ein Beitrag zur jungen Bruchtektonik der argentini-
schen Puna und ihr Verhaltnis zu den angrenzenden Andenabschnit-
ten.- Geol. Rdsch., 59: 1064-1087, Stuttgart.
Sibson, R. H. (1977): Fault rocks and fault mechanisms.- J. Geol. Sot.
London, 133: 191-213.
Woodcock, N.H. (1986): The role of strike-slip fault systems at plate
boundaries. - Phil. Trans. R. Sot. Lond., A 317: 13-29.
Zonenshayn, L.P.; Savostin, L.A. & Sedov, A.P. (1984): Global paleogeody-
namic reconstructions for the last 160 million years.: Geotecto-
nits, 18: 181-195.
_^ ._

151

GEOLOGY AND METAMORPHIC EVOLUTION OF IGNEOUS ROCKS; COASTAL CORDILLERA


OF NCHILE BETWEEN 2325 AND 2420%.

F.LUCASSEN, G.FRANZ

Fachgeblet Petrologie. TU-Berlin. Ernst Reuter Platz 1, D-1000 Berlin-12

Resumen

En la Cordillera litoral. SAntofaQasta, NChilr ocurrn rotas plutonicas que se prroentan sfectadas por un mrtrmorfismo
de In faoirs Qranulitia de alta trmperatura y baja presion. Debido al la penrtracion posterior de una fasS fluida 10s QnSiMS
do piroxrnos ss presentan parciafmmtr anfibolltfzados y miQmatfzados. Posteriormmto ss empluaran ourrpos Qab-
rolcor y tonaliticos. Ester currpos todavia furron akctados por dloho wmto drl mrtamorfismo de la facirs anfibolitiu. q
trrmlno drl cfclo magmatlco as documrntado por rl emplazamlento de dikes y de rotas volcanicas.

Palabrar claws: Cordillera Moral NChile, Orto~nsis~s, la&s Qranulitfca, mrtamorfismo wntrolado por Ias fluidSs.

Introduction

The Coastal Cordillera north and south of Antofagasta, NChiie is mainly build by volcanic rocks, in-
trusiva and sediments of the Jurassic magmatic arc but also contains some rocks of paleozoic age.
South of Antofagasta parts of the coastal cordiliera are uplifted along the Atacama fault system with
Increasing rates to the western blocks (R&SUNG, 1999). Our investigations concentrate on the
deepest accessable part of the fossil arc. First results presented here should be considered as work
In progress.

Geological setting

From the southern tip of Peninsula Mejillones down to at least Caleta Cobre extends an area (120 x
lo-20km) between the coastline and the western main fault of the Atacama system showing consi-
derably amounts of pyroxene gneisses and amphibdites.
The felsic pyroxene gneisses with minor amphiboiitized parts strike N-S along the coastline. To the
east they become increasingly amphibolitized and amphibolites with minor pyroxene gnelss relics
predominate. Foliation in the amphibolites Is sometimes well-developed striking N-S, locally subor-
dinate E-W directions, but always steeply dipplng. N-S directions are probably related to the juras-
sic-recent Atacama fault system but there may be also older N-S directions at the continental mar-
gin. Arnphibolites show locally frequent migmatitic textures that also affect to a minor extent the py-
roxene gneisses. Deformed mafic dikes can be found in the amphiboiites.
152

These rocks are intruded by prevailing small gabbro stocks also following the N-S trend. Diffuse
migrnatitic contacts to the host rock are common. The larger gabbros show rnagmatic layering. Fa-
brics are undoubtly magmatic.
Along the western side of the western main fault of the Atacama system a quarts diorite pluton cuts
very sharply all migmatitic textures of the amphibolite. At its edges it is usually foliated but the core
preserves its magmatic fabric.
All rocks except the gabbros are cut by steeply dipping, SE striking andesitic dikes, Ductile textures
in the country rock are unconformable cut and do not continue into the dikes.
Quartz lenses and veins appear in all rocks at all stages of deformation except the formation of py-
roxene gneisses.
Ouartz-feldspar, less carbonate, veines are sometimes frequent. They are not ductile deformed.
In the south andesitic and basaltic volcanic rocks of a terrestrial deposition environment partly cover
this plutonic-metamorphic sequence in isolated patches. Contact relations between these series are
not sufficiently known, but there may be despite the obvious tectonical contacts also unconformable
erosion contacts.

Metamorphic evolution (T-t-path)

The metamorphic history is presented in a T-t-plot (Fig.1) because of the lack of pressure sensitive
assoctatons in all rocks.

200

. .. . . .. **
0,
time (no scale) 2OOMa 190Ma %ZC.~.. lOOMa
153

Maximum pressure for the granulite facies may be up to 10 kbar indicated by the stability of an-opx-
cpx. The real pressure might be much lower. For the subsequent amphibdite facies the pressure
was c5 kbar (amph-pfag composition). Also garnet is lacking. Changes in metamorphic condiiions
are mainly recorded as changes in temperature, deformation and fluid access.
1. lntrusfon of the protdith of the pyroxene gneisses. Numerous smaller intrusions of gabbrolc
magma of refatMy homogeneous chemical composition buifd up the bathdith. Layering is not
found, differentiation products are very rare. Pyroxene gnelsses and amphibofites show sometimes
relics of a magmatlc fabric and more frequent pfagioclases with zoned an-rich (70 can < 90) cores.
2. Formation of the pyroxene gneisses at gram&e facies conditions. All stages of fabric develope-
ment from a magmatic to a fully recrystallised granobfastfc one are observed. Fdfations are quite
rare. The prevailing mineral association is pfag (an 50-60) - opx (enstatite) - cpx (diopsfde). Minor
phases are biotfte, magnetite, ifmenite, apatite, different amphibofes and occasionally other retro-
grade minerals like chlorite, quartz and epidote.
Temperatures (two pyroxene thermometry) range between 700 -900C depending on the employed
thermometer. For gf@ thermometer no systematic thermal gradient can be derived for different sam-
ples. Some pyroxenes are slightly zoned with increasing alumina and calcium (only opx) to the core.
Rims are very small and unzoned pyroxenes are found In the same samples. So the occurence of
slightly lower temperatured rims may be ascribed to retrograde exchange.
3. Formation of the amphibdites by ffukf contrdled metamorphism of the pyroxene gneisses. The
typical paragenesis is plag (an 30-40) - amph (one ore more generations) + biotite, quartz,
magnetite, fimenite. Other minor minerals are predominantly retrograde formations like chlorite,

0.1 : I
I 1
I I I
s102 Al203 Ill30 cao I20
1102 F.203 @I@0 na20 P

F10.2 shows quafltatively the gain and IOSS of element8 normalirrd to pyroxma gnrisser by the formation of thr amphl-
bolltes (primary Inhomogeneity should be exduded by sampling).

epidote, sphene, sercite. Amphibolitization of pyroxene gneisses can be observed at different sta-
ges. The main reaction is plag + pyx t K-fluid = amph t Mg,Ca-fluid. Changes in the whole rock
chemistry (fig.2) by this transformation show evidence for high amounts of fluid.
154

Tern ratures from amph - plag compositions and magnetite - ilmenite pairs range between 50&
SW8C. Cores of magnetite - Umenlte and amphiboles show compositions equilibrated at lower tem-
peratures before peak conditions. Some maflc dikes must have already intruded before or during
amphibolfte facles metamorphism because they are strongly ductile deformed.
4. Intrusion of the gabbros and the quartz dlorlte. Gabbros are chemically different from all other
rocks tending to more basic compositions. Primary divine and pyroxene are often completly chan-
ged Into amphibole whereas the magmatic fabric and plagioclas composition (an >90) remains un-
changed. SubsofMus amphlbolit&ation occurs at different temperatures and covers a range from
early corona textures (ca. 8ooC from cpx and secondary opx) to late actinolitic amphiboles.
The quartz diorfte pluton intruded after the peak of amphibolite facles metamorphism and is not af-
fected by migmatlzation. At least at the edges of the pfuton the rocks show features of subsolidus
amphibolitlzatlon.
5. Intrusion of the andesttic dikes and deposition of the terrestrfal volcanic rocks. Andesitlc dikes cut
all structures in the host rocks and are hardly affected by ductile deformation. Volcanic rocks (Form.
La Negra?) are metamorphosed under maximal greenschlst facies conditions (mineral paragenesis
In amygdales and alteration products). It Is not yet clear for this area wether this is caused by reglo-
nal (burial) metamorphlsm or autometasomatism. The base of the volcanics Is scsrecley known (In
other locatlons: jurasslc sediments, BUCHELT & TELLEZ, WSf3).The uplift and erosion of the base-
ment before deposition of the volcanlcs and ensuing sub&fence may be reasonable by geological
constraints and different metamorphic grades but Is not proven.

Acknowledgements: Thls work ls sponsored by the Deutsche Forschungsgemeinschaft (DFG)

BUCHELT, hi. & TELLEZ, C. (1oeS):The Jureeelc Le Negn Formetion in thr eree of Antolegaate. Northern Chile. In BAHL-
BURQ, Ii., BREITKREUZ,Ch. 6 GIESE, P. (Ed&), Lecture Notre in Eetth Sciences Vol. 17, Springer Verlag Berlin Heldel-
berg, 1988,171.182.
DAMM, K.W., PICHOWIAK,5.81 TODT, W. (1986): Geochemio, Petrologleund Geochronologieder Plutonitound der me-
tamorphm QrundgebirgeoIn Nordchilr. Berlinergoowiu. Abh., MS, 79146.
R&SUNQ, R, (loa0): Petrologicin rinrm liefon tiatenstockwork der jureeeiechm megmrtiechen Bogenr in der nord-
chilrnlechrn Kiistenkordillrroriidlkh vonAntofagaeta. Berlinergeowiu. Abh., Al 12
SCHEUBER, E. 6ANDRlESSEN, PAM. (in prer): The kinematic end geodynamic significanceof the Atecams fault zone,
NChile. J. Struct. Geol.
155

SEDIMENTOLOGICAL, TECTONIC & PALEOMAGNETIC IMPLICATIONS


OF THE MIDDLE JURASSIC OF CALETA CAMARONES &
QUEBRADA CHIZA, NORTHERN CHILE.
AN OCEANIC BACK-ARC BASIN MARGIN.
156

may in part be contemQoraneousw&h the Yokanks which haYe been dated at 186 Ma.
(Rogers,1985).
In places the eroded volcanic headlands are uncorJormablyovertaln by massive
MafomMonalcongkmeratss.Amixed sequmcedrnatinecarbomtes,vokanktastks
and clastks then lii uncoMmably on the congkmemtes, 01 directly on top of the
headlands. Vokankiastk sandstones and pebble beds are intert>edded wth the
carbonates atthe base ofthe sequence. Typkxlly the carbonatss we very fi~lntxl
Wm planar flne taminae, and represent the backmund sedtnenlalbn. They are
moderately Siliceous suggesting deposmkn aboye but near the C.C.D. Mid-Bajocian
anxnonitesoccuatthebaseofthesequence. Slunpswlhinthe carbonatesequence
stggestasedhMsovcehomthe~thev&anksrc.
Fauted againstthe cd&on&s, and possbly corWqxraneous wth them, ls a
thkk sequence of marine vokanklastks. These consist Ocwlell bedded Yokanklastk
sands,brecc~lava~andg~hotasMkv/?pe~b,derived~~the
Yo$oErc. The sequence is cutby nsptuniandylm showing the continuingphase of

WMng the basin are proxh-ral~okmoclastlcs. wing unconlonrrably on top ofthe


carbon&s and vokanktastks. These are aW+reccii, s&marine IaYa flmm,
showing~very little tmsprt, and may represent apron-fan depost~ adjacent to the
vokankarc.
EasWrds in Ousbtada C&a, a targe thkkness of vWbed&d cak-arenilic
ltnestones occu: These may be OrBathoMan age orpossbly cor&mpofaneous wth the
sequence at Caleta Camarones, They represent futhef back-arc basin stxiinentation,
more diil to the volcanic arc.
overlying these cak-arenRes is a sequence of Callovian or BalBonian marine,
hmkh sam#mes, s&tones and shales represeting a strallowingiqwards
sequence. lhese, togeWr with the cak-arentes below, are cut by appmx.tily
n&h-south bending basa&k d&es (the Cuya Dykes) whiih appear to act as feeders to
overlying highly altered has. The Cuya dykes represent the final phase of basin
extmsbn.
Middle-Calkvian marine sandstones, dated by unites, lie uncoti~ly on
top Orhe tavas,and are overlain by Ms and a Lavafiow which represents the youngest
Juassk &ala eaposed in the area.
Duing~UpperJuasskorearly~~usthebasinwas~p~ted,lLauLedand
sironglyeroded before depos&bn ofmflylng Cretixeous conUnerW sedtnents.

Tectonics

Despite ongoing subduction of the Nazca plate beneath the Andean margin
since the late Triassic, the Cm&l Cordillera of northem Chile is dominated by
e~nsbmalte&nics.
In addljon to a sl&igaphk change, a change in tectonicstyle also occus abng
the Coastal Cordillera. In the area ofshady,quebradas follow the paths of approxinale~
rmWe&soutkW trending fauts which cut the Mea Group sedimentS and haYe
dOWtYhMttlemtOtMMttl.
These txxthmb*M hnding faults have cut north-sollth Irending
basemetifau~ Mkh once conboiled the oceanic and ensialii back-arc basin margins.
These no&h-so&h fauls haYe been reactWed by the Alam Fault zone, Wkh
bounds the Coastal Cordillera to the south. Howmr, the mthem bkx% t=Ying
displaced the basement faults, has controlied the bend of the Atacama and associated
faub which swing out to the sea in the vicinitydthe Salar Orande.
. _... ._ . .

157

Fig.2
-_
K
J
n

N
I

Idh-\i
Q

0.
158

Palawrnagll&nl

In the area of study, palawrnagnetk studies indicate a general antickdwke


rolatM (Fig. 3a), which continues into so&hem Peru. In contra$ the Co&al Cordillera
b the sot&h k related cbckwke (Fig. 3b), (Haftley et al, 1988). Also In Peru, ck~H%
roQtionsocclras~~llthas4OS~uieretal,l988).
The~Cordillerakawhole~orhul-bkdcs,cutby~Llthand
~~~th-uulest$uts.Thebbdc~nactual~occuslnsnallscalebkcks.One
SIJ& bbck alCaleIa CamaroneS, a h~tdfined headhd, acbralv Showsa ckdmlse
datfon In conlwt to the general arWx%wke rolatbn of adjjcent bbds ,
demonsWhgb scaleofrotatkm.

Coftclusion

Sbadgaphk, tecbnk and pabnagnetk studies in the CBstal Ccrdilkta of


tlorthemnost Chile Indicate UW.an oceanic back-arc basin exls&M In the Middle
Juassk. AcontlnwWmargln arc exlsbd bthe so&h, represented by the La Nega
Fotmat&n.ThebanRlonbebmenensialiiand oceankback~basinsprobabtyoccus
lnthsvkln~d~Satar(jrande.Theistandaretothenorthis~ntperhapsdueto
undeWt&gandsubducbkn.

References

Hartley, AJ., Tuner, P. , William, Q.D. and Flint S. Patmmagnetim ofthe


Cordilleta de LaCosta, northem Chile: evidence for Cal forearc rota&n. E.P.S.L.,
89,375386,1988.
h&wW, T. , Laj, C. , Megad, F. , l?ofwch, P. , MRoUard.P. and Farfan
Me&ano, A. An accreted continetIal bane in norM&em Peru. E.P.S.L., 88,
182-l 92,1988.
Rogers, Q. A geochemkal ba~erse awoss the notth Chilean Andes. Unpbl. Ph.D.
thesis, Cpn Univ., Ml&on Keynes. 333 pp., 1985.
Salas, R., Kas& I?. , Montecinos, F. and Satas, I. aeOkgk y recwsos mineral@
del dqarbmento de Arka ; lnst lrwest Geol., paca, Chile : Memoria Uniwsidad de
Chile. 111 pp., 1966.
159

MISE EN EVIDENCE DE STRUCTURES TANGENTIELLES ANDINES DANS LE SUD DE


LA BOLlVIE (REGION DE TICA TICA, DEPARTEMENTO DE POTOSI) PAR SPOT.

LAUBACHER, G. lnstitut Francais de Recherche pour le Devetoppement en Cooperation


(ORSTOM) , B.P. 5045, 34032 Montpellier cedex 1, France.

Abstrsct.

Spot imagery was used for mapping and analyring the stratigraphy and teclonism of an area of
60 km x 60km in the eastern Cordillera of southern Bolivia. Paleozoic, Cretaceous and Tertiary rocks
crop out and could be subdivided into units using spectral and morphological criterias. Tertiary
tectonics look very spectacular and complexe, and large scale tangential movements from west to
east are inferred .

Mots-cl&: Image Spot, stratigraphie, recfonique tangentielle, CordilltVe orientale,


sud Bolivie

Introduction

Une bande Spot multispectrale (3 canaux dans le visible, pixel de 20m), couvrant une
zone denvlron 60 x 60km et situee au SW de Potosl (Fig.l), centre de la scene: 2001 S et
66018 W), a et6 acqulse pour realiser une cartographle stratigraphique et tectonique
preliminaire. Cinterpretation a 618 entreprise sans connaissance de la region de laquelle on ne
disposait que dune cartographic a grande Bchelle (carte geologique au l/l OOOOOOede GEOBOL).

Lutilisation du logiciel STIMDI, couple a une


imprimante VERSATEC (Centre National
Universitalre Sud de Calcul), permet de realiser des
tirages de compositions colorees a haute resolution
(RVB, ACP) optimales a Ibchelle du 1175000e.
Llnterprelation a 616 faite visuellement en utilisant
les criteres morphologiques et spectraux (les
compositions colorees RVB fournissant une
excellente image morphologique, alors que les tirages
a partir de IACP permettent une bonne
discrimination lilhologique). Un travail sur &ran
(logiciel PLANETES, Centre ORSTOM de Montpellier)
avec dautres traitements (rehaussement, ratio,...)
permet souvent daffiner Iinterpretation. La taille
des objets et du pixel sont des facleurs limitants et
les structures petite% m&me encore visible% sont
difficlles a Interpreter. Quelques photos aeriennes
ont Ogalement 616 utilisees.
Fig.1

La qualit des donnees Spot et les conditions daffleurement particulierement favorables


(pratiquement pas de vegetalion, lithologie souvent contrastee, erosion profonde ayant
largement degage les structures tectoniques ) ont facilite Iinterpretation et permis une
cartographle geologique preliminaire relativement detaillee.
-. ..---

160

Les grands ensembles lithologtques (Fig.2)

Le nord et Iouest de la scene sont occupes par un masslf volcanlque, dage NBogbne
(Schneider 8 Halls, 1985) dont les sommets depassent 5000m (presence de glaciers residuels
et dune ItIOphOlOQie gladaire IIIarqU6e avec vallees glaciaires et arcs morainlques). Au nord, le
plateau lgnimbritique de Los Frailes, encore peu Incise, recouvre directement les structures du
Paleozoique et du Cretace, alors qua louest sintercale une formation volcanique (Tertiaire
moyen 7) affect&s par un model6 glaciaire (vallees en auges, arcs morainiques). Des placages
discontlnus constituent des affleurements dans la partie SW de la scene: II sagit dignimbrites et
de cendres dont derivent de grandes quantites de sables Boliens formant par endroit des systemes
dunalres. Enfin, dans le coin NW de la scene, les ignimbrites sont surmontees par deux
affleurements de volcanites dont la morphologie suggere des coulees basaltiques.

Le substratum pr&volcanique affleure prlncipalement dans la partle centrale et


meridionale de la scene. Les caracteristlques morphologlques et spectrales des images
permettent generalement de &parer aisement le Paleozoique du Cretace et m&me de delimiter, B
linterieur de chacun de ces grands ensembles, des sous-ensembles lithologiques souvent
correlables avec des formations reconnues au niveau regional.

LOrdoviclen, Bpais de plusieurs milliers de metres, semble 6tre represente par


plusieurs sequences lutitiques et greso-lutitiques, avec une stratification soulignee par des
banes de quartzites intercales dans les lutites. Ces sequences sont surmonlees par des lutites
greseuses (400-700m depaisseur) a stratification peu marquee (diamictite Cancaniri du
Sllurlen InfClour 7, carte geologtque de Bolivie au lllOOOOOOe), puis par une sequence
de g&s-quartzites (200~SOOm), en gros banes durs, soulignant Ies structures tectoniques, et
qui pourralent correspondre a du Sllurlen superleur etlou du Devonlen (fm Llallagua
?). Le Siluro-Devonien semble en continulte stratigraphique, mais il nest pas certain quil en
soit de m6me entre lordovicien et le Silurien.

Le Cretace presente la serie la plus complete de la scene dans le synclinal du rio Totora
au sud de Tica Tica. En tenant compte, bien stir, de leur ordre de superposition, les contrastes
morphologlques
161

Fii. 2 : carte gBologique et structurale laite B partir de Spot.


1, Quaternaire alluvial et tluvioglaciaire; 2, coul6es basaltiques;
3. ignimbrites et cendres volcaniques et 4. volcanites nboghes
(Los Frailes) discordants sur les structures andines: 5, Cr6tac6
terminal et PalBocBne; 6, C&ta06 infhieur et sup4rieur; 7,
Siluro-Wvonien : 8. Silurien (Fm Canaceniri): 9, Ordovicien.

Fig. 3
162

Au sud de Calasaya et a louest de Tica Tica, le Paleozoique (formation Canoaniri)


surmonte la serie complete du C&ace par un chevauchement a vergence E . Sa trace NS
se suit sur limage sur plus de 10 km avec des pendages estimes allant de 30 a 60W. Dans sa
prolongation vers le sud et le sud-sud-ouest, dautres chevauchements, dont un tres plat, sont
rep&es.
A lest et au sud-est de Tica Tica, le front du domaine B deborde largement vers Iest,
dans/ou sur le domaine A (Fig. 2). II sagit de grands synclinaux form& de quartzites et lutites
de Stluro-Devonlen . Le style et la direction de la deformation, qui les caracterisent, contraste
avec ceux de IOrdovlcien qui les borde au nord et au sud. COrdovicien, mais aussl des ecailles de
Cr4tace semblent passer sous les synclinaux, apparemment par Iintermedlaire dun contact
tectonique plat. Une telle disposition (klippe 7) impliquerait des deplacemenis tangentlels
Important% avec un domalne B au molns en partle allochtone et decolIe au niveau du Silurlen
lutitique (formation Cancaniri) et chevauchant le domaine A autochtone ou subautochtone.

Enfin, ce domaine B est affecte par de grands accidents raides NW-SE, ayant
apparemment eu un jeu senestre souligne par la torsion a grande echelle, tres spectaculaire,du
synclinal de Cr&ac4 qui affleure au NNW de Calasaya avec une composante chevauchante. Ces
accidents delimitent de grands panneaux a linterieur desquels le raccourcissement se traduit
par des chevauchements. Ce sont peut-Otre ces accidents, jouant comme des rampes laterales.
qui ont permis dabsorber en partie les variations de lampleur du chevauchement frontal du
domaine B . Cest sans doute a eux que lon doit la forme atypique des grands synclinaux , en
partlculier la forme en polre allongee sulvant une direction WNW-ESE, que presente celui qui
est situ& a lest de Tica Tlca. Ces decrochements ne semblent pas affecter le domaine A dans la
partie es1 de limage; II est done possible que leur fonctionnement soit lie a la mise en place de
funite allochtone B.

Conclusion

Les structures tectoniques observables sur la scene Spot de la region de Tica Tica
suggerent Iexistence de deplacements tangentiels de grande ampleur a vergence est . En effet, le
front oriental du domaine B surmonte, apparemment par un contact tectonique plat, le domaine
oriental A sur plus de 10 km. Dautres contacts plats, qui ont et6 rep&es sur les images a
louest et au sud-est de Tica-Tica, impliquent Ogalement des deplacements horizontaux
Importants. Cependant, on ne voit pas clairement, sur limage, de relations entre le contact plat
frontal et ces chevauchements situ& plus en arriere. Ces relations conditionnent Iampleur des
d&placements : faut-il parler de chevauchements de grande ampleur ou sommes nous en
presence de veritables nappes?

Des travaux de terrain &cents am&rent Baby et al. (ce Symposium) a proposer, dans cette
m&me region de Tica Tica, lexlstence dune vaste nappe (nappe de Calasaya) dont ils estiment la
fleche a 42 km. Nos resultats, obtenus a partir de limagerie Spot, ne permettent ni dinfirmer
ni de confirmer une telle ampleur des d&placements tangentiels. Cependant, cette etude monlre
que les donnees satellitaires modernes sont maintenant susceptibles dapporter une aide efficace
a la cartographic stratigraphique et structurale surtout dans des regions mal connues.

Baby P., Sempere T., Oiler J.. Blanco J., Zubieta D., H&ail G. - Evidence for major shortening
on the eastern edge of the bolivian Altiplano : the Calazaya Nappe. (this Symposium).
Schneider A.. Halls C., 1885 - Chronology of eruptive processes and mineralitation of the Frailes
Karikari volcanic field; eastern Cordillera, Bolivia. Communicaciones. 35, 217.224., Univ. de Chile,
Santiago.
Sempere T., 1966 - Contribuci6n a la estratigrafla del Mesoz6ico boliviano en el dominio andino.
Publication Mission ORSTOM Bolivie, I, 34p.. La Par.
Sempere T., Otler J., Barrios L., 1988 - Evoluci6n tectonosedimenlaria de Bolivia durante el
Cretacico. Vto Congreso Geol. Chileno, 1.111,H37-H85.
Vargas E. 1981 - Comparaci6n entre la tnterpretaci6n geol6gica de la lmagen Landsat I y
fotograflas aereas del area Potosl -Tazna . Rev. Geoc. UMSA, I, 57-91.
163

EVIDENCE FOR MAJOR SHORTENING ON THE EASTERN


EDGE OF THE.BOLIVIAN ALTIPLANO : THE CALAZAYA NAPPE

Patrice Baby (l), Thierry Sempere (l), Jaime Oiler (2),


Javier Blanc0 (2), David Zubieta (2), Gerard Hdrail (3)

1 : ORSTOM, CC 4875, Santa Cruz, Bolivia

2: YPFB - GXG, CC 1669, Santa Cruz, Bolivia

3 : ORSTOM, CP 9214, La Paz, Bolivia

Dans la zone de transition entre Altiolano et Cordillere


Orientale vers 200 s, la cartographic h*une nappe dont la
fliche est d'au moins 42 km fournit une nouvelle preuve de
l'existence de raccourcissements considdrables dans les Andes
boliviennes. A l'echelle regionale, la giomdtrie de
l'allochtone apparalt gouvernee par celle du bassin
ashgillien-silurien.

Key-words : Bolivia, eastern Altiplano, thrusts, name,


paleogeographic control.

Introduction

At about 20 S, the transition zone between the Altiplano


and the Cordillera Oriental shows an unusual structural com-
plexity, which happens to be illustrated on the advertising
poster of the symposium. Field mapping and satellite imagery
analysis demonstrate the existence in this area of an exten-
sive tectonic nappe, the Calazaya nappe, emplaced on a pre-
viously deformed domain (1).

Structure and stratigraphy

The Calazaya nappe (fig. 1) overlies the Ayoma-Atocha (re-


lative) autochthon by means of a subhorizontal contact, which
is the southern prolongation of the Main Altiplanic Thrust
(CALP) (2). The nappe "roots" westwards into the Uyuni
allochthon.
The base of the allochthon consists of non-stratified fine-
grained diamictites of Ashgillian age, in which the dedolle-
164

-Several klippes define an "envelope" for the initial exten-


sion of the nappe (irregular folding within the southernmost
klippes suggests that gravity played some role in their final
emplacement). The geological map (fig. 1) shows that the
horizontal amplitude of the tectonic covering is at least 42
km.
The eastern part of the autochthon shows elongated and
narrow synclines with Cretaceous strata, locally deformed by a
N-S-trending left-lateral transcurrent deformation, and
thrusts. Some of these synclines are partly hidden below
klippes of the Calazaya nappe. The western part of the
autochthon shows wider synclines, E-verging thrusts and, in
two places, duplexes. Because they are beheaded by the basal
overthrust of the nappe, both duplexes formed prior to the
nappe emplacement, as did the thrusts.

Fig.2. Structural ssttins of the


Calaraya aree (framed). The CALFSiK
system is the eastern boudary of the
Uyuni allochton and the western
boudary of the Ayoma-Atocha domain.
The bcudary line of the Calazaya
nappe is shown With its envelooe
(see fig.1). Dotted areas: Late
Oligocene-Early Miocene foreland
basins of the CALP-SW system, and
minor piggyback basins.

Regional significance and paleogeographic control

The basal overthrust of the Calazaya nappe connects


northwards with the CALP and southwestwards with the Khenayani
Fault System (SFK) (fig. 2). The CALP-SFK E-verging system is
a major tectonostratigraphic boundary which separates the
(western) Uyuni allochthon from the (eastern) Ayoma-Atocha
domain (2). The SFK is a complex thrust system, with a basal
decollement also located in the Ashgillian non-stratified
diamictites (3). Based on cartographic and sedimentological
grounds, the CALP-SFK seems directly controlled by the georne-
try of the Ashgillian-Silurian basin (1,3) : the Uyuni
allochthon contains thick and deep facies of that age, and its
-sm.-, . ...

165

Fig.l.Geological map of Calazaya area. 1: Middle Miocene


and younger rocks. 2: Late Oligocene-Early Hiocene. 3:
Mesozoic-Paleocene. 4:Silurian.S:Ashgillian. 6:pre+sh-
gillian Ordovician. 7:basal overthrust of Calazaya nappe
and envelope of the nappe. 8:other faults. C:Calazaya
village. UY:Uyuni city.

ment is located; these rocks are overlain by thick Silurian


turbidites and shales, on which rest unconformably mainly
continental deposits of Mesozoic-Paleogene age. The autochthon
shows extensive outcrops of pre-Ashgillian strata and thick
Ashgillian diamictites; the basal Silurian unit is comparati-
vely thinner (when preserved below the Mesozoic unconformit$,
which rarely occurs); Mesozoic - Paleogene deposits are simi-
lar to those in the allochthon.
166

basal decollement is located in the lowest part of this se-


ries; in the autochthon, Ashgillian facies are more proximal
and apparently thinner.
Thick continental sedimentation of Late Oligocene-Early
Miocene age is recorded in the autochthonous Bolivar-Mondragon
and Lipez basins (fig. 2), and is believed to represent the
correlative foreland deposits of the E-verging deformations
described above (1,3,4).

Conclusions

The Uyuni allochthon was overthrust onto the Ayoma-Atocha


domain during the Late Oligocene-Early Miocene time span. In
the' studied area, this overthrust appears as a spectacular
nappe, which post-dates some deformations in both allochthon
and autochthon. The Ashgillian-Silurian basin geometry seems
to have deeply controlled the Uyuni allochthon geometry.

References

(1) T. Sempere et al., manuscript submitted to the C.R.A.S.,


1990.

(2) T. Sempere et al., 5th Chil. Geol. Gong., v.1, p, A127-


142, 1988.

(3) P. Baby et al., manuscript submitted to the C.R..-\.S.,


1989.

(4) T. Sempere et al., 8th Boliv. Geol. Gong., p. 45-46, 1986.


167

THE ALTIPLANO : A PROVINCE OF INTERMONTANE


FORELAND BASINS RELATED TO CRUSTAL SHORTENING
IN THE BOLIVIAN OROCLINE AREA

Thierry Sempere (l), Gerard Herail (2), Jaime Oiler (3),


Patrice Baby (l), Luis Barrios (3), Rene Marocco (4)

1 : ORSTOM, CC 4875, Santa Cruz, Bolivia

2: ORSTOM, CC 9214, La Paz, Bolivia

3 : YPFB-GXG, CC 1659, Santa Cruz, Bolivia

4 : ORSTOM, CCi 6596, Quito, Ecuador

Resumen

El Altiplano comprende fajas plegadas y corridas, zonas de


transcurrencia y cuencas intramontanas de antepais, tocios
elementos relacionados entre ellos en el tiempo y en el espa-
cio. Esta complejidad se debe a la posicidn de1 Altiplano en
el corazdn de1 oroclino boliviano.

Key words : Altiplano; foreland basins; thrust belts; wrench-


faults; stratigraphy.

The Altiplano-Puna is an enigmatic high plateau, 200 km-


wide and 1500 km-long, with a mean altitude of 3650 m (11.
Though it-constitutes a major peculiarity of the central Andes
and, more particularly, of the Bolivian orocline area, its
origin is far from being fully understood, partly because of
scarcity and dispersion of information. Recent integration of
geological and geophysical data (many of them unpublished and
borrowed from the Bolivian oil company Y.P.F.B.) within the
central Andean geotectonic issue has led to the proposal of a
new model (2).
Based on tectonostratigraphic grounds, we distinguish a
complex "western belt" of terrains (a terranes) separated
from an "eastern belt" by the Intra-Andean Boundary Fault
(FLIA) (fig. 1). Though it runs in an area mostly covered by
recent rocks, the existence and position of the FLIA are
inferred from very strong stratigraphic differences between
these belts, some geological data in less covered areas of
Peru, zones of high gravity gradient, and preliminary magneto-
telluric data. In our model, part of the western belt was
underthrust beneath the eastern one (2). This underthrusting
Fig.1. Simplified structural map cf the Boliviar
orocline. Dotted area:"western belt";deformed area in
white:"eastern belt" (see text) .Thin dot 1ine:drainage
boundary of the Altiplano-Puna. CANP: Main Andean Thrust;
CFP:Main Frontal Thrust; FLIA:Intra-Andean Boundary
Fault; CB:Chapare buttress; CI:Cuzco indenter; MI:Mizque
(crushed) indenter; SB:Susques buttress; VH: Vilcabamba
hinge (2); eSg:edge of Subandean deformation.

caused the Bolivian orocline to develop in the eastern belt by


inception of the Main Andean Thrust (CANP) system (fig. 11, in
an asymmetrical self-indentation of the South American margin
(2, 3).
This Late Oligocene major tectonic crisis resulted in the
"jump" of the Andean deformation front from a location west of
the Altiplano to an eastern position related to the CANP
system (3). During the Eocene-Middle Oligocene time span, the
Altiplano had been part of the external foreland basin of the
paleo-Andes. Because of the jump of the deformation front, it
became locked in the Late Oligocene between the main magmatic
arc and the rising Cordillera Oriental, and has evolved in an
endoreic setting since then. At all Neogene stratigraphic
levels, proximal facies as alluvial fans and braided rivers
deposits grade into lacustrine facies which may include lo-
cally thick evaporites.
,e-. I__-.

169

The structure of the Altiplano is quite complex. It in-


cludes fold-thrust belts, wrench-fault zones, and intermontane
foreland basins passing laterally to pull-apart basins (fig.
2), all related to one another in space and time.
A major part of the "morphological Altiplano" is located
within the Ulloma-Coipasa-Uyuni tectonostratigraphic unit
(UCU). South of the Coipasa strip (fig. 21, the Uyuni domain
is a NNE-trending, E-verging, mostly covered fold-thrust belt,
separated from its related foreland basin (the Lipez basin) by
the Khenayani thrust system (SFK) (4). North of the Coipasa
strip, the Ulloma domain is a NNW-trending complex basin,
structured by mainly W-verging faults. Its northern half
functioned as the foreland basin of the SW-verging Huarina
fold-thrust belt (FPCH) (3). More to the south, the basin
seems to have been controlled by "alternating" transtensional
and transpressional conditions. The Coipasa strip is bounded
by 2 subparallel, WNW-trending, left-lateral wrench-fault
zones. The complex NNW-trending Sevaruyo-Chita fold-thrust
belt (FPCSC) originated in an area of the strip rich in Creta-
ceous gypsum and halite, in relation with the left-lateral
movements of both its boundary faults.

The complex history of the Altiplano includes 2 main pe-


riods of continuous deformation approximately during the 27-19
Ma (3) and 11-5 Ma intervals. Deformation developed in the
fold-thrust belts and wrench-fault zones, and, to a lesser
degree, in the basins (foreland thrusts; push-ups). Relative
tectonic and magmatic quiescence reigned between 19 and 16 Ma.

Fig.2. Structural sketch-map


of the Bolivian Altiplano and
adjacent regions. Dotted
areas: main areas of thick
Late Oligocene-Neogene accu-
mulations (approximate). Thin
dot line: drainage boundary
of the Altiplano. Dashed
lines: political borders-
.CALP: Main Altiplanic
Thrust; CCR: Cordillera Real
Thrust; FCA:Chita-Arica fault
zone; FCC:Coniri thrust front
FSI:Sevaruyo-Incapuquio fault
zone; SFK:Khenayani fault
system; FPCH:Huarina fold-
thrust belt (SW-verging)
FPCSCaSevaruyo-Chita fold!
thrust belt (E-verging).
Ull:Ulloma; Uy:Uyuni. Other
abbreviations:see fig.1.
170

Activation of felsic explosive volcanism at about 11 .Ma is


noteworthy.

Crustal thickening below the Altiplano, which is thought to


represent crustal shortening (1,5), probably originated in the
underthrusting at the FLIA of part of the western belt, '. .
in the Bolivian orocline development (1,2). Because the Aitf-
plan0 lies in the heart of the orocline, its tectonic evo-
lution may be regarded as the "inner" record of the oroclinal
bending. Thus the importance of chronology and structural
geology in studying the Altiplano is crucial to define the
steps taken by the deformation in the orocline area. As a
consequence, the simplistic vision of the Altiplano functio-
ning as a tensional trough should be abandoned.

References

(1) B.L. Isacks, J. Geoph. Res., v. 93, p, 3211-3231, 1988.

(2) T. Sempere et al., 28th Int. Geol. Cong., v. 3, p. 72-73,


1989.

(3) T. Sempere et al., manuscript submitted to "Geology",


1989.

(4) P. Baby et al., manuscript submitted to the C.R.X.S, 1989.

(51 D. Roeder, Tectonics, v. i, p. 23-29, 1988.


.( I*.. _ .

171

THE AGE OF THE ZONGO PLUTON AND THE TECTONOTHERNAL EVOLUTION


OF THE ZONGO SAN-GAEAN ZONE IN THE CORDILLERA REAL, BOLIVIA.

Edward Fa, Alan H. Clark and Silvia M. Heinrich

Dept. of Geological Sciences, Queen's University


Kingston, Canada, K7L 3N6

Les mesures "Ar/"Ar sur micas echantillonnes verticalement


dans le pluton de Zongo (U-Pb, 225 Ma), montrent que les
donnees discordantes K-Ar (rejeunissement NE) resultent de
chrontours de muscovite et biotite orthogonaux, impliquant,
il y a 39 Ma, une rotation de 90' de la zone Zongo-San Gaban.

Introduction

The Cordillera Oriental of NW Bolivia and SE Peru is


underlain by granitoid rocks of Permo-Triassic age (e.g.,
McBride et al., 1983;
Laubacher, 1978) that
intrude Paleozoic meta-
sediments. However,
Bard et al. (1974) as-
'1?ywI?8"lU""
wull signed an "Eohercynian"
age to the variably
foliated Zongo pluton
in the Cordillera Real,
Bolivia (Fig. l), on
the basis of similari-
ties with deformed
rocks in central Peru
and on its relationship
to an apparent meta-
morphic culmination lo-
cated in the metasedi-
ments NE of the pluton.
Furthermore, in a SW-NE
transect through the
., __ .__-.___ ._ ..,. _,_
,.

172

Zongo pluton, K-Ar and 'OAr/'Ar, muscovite and biotite dates


are widely discordant, each set younging monotonically toward
the NE (McBride et al., 1987). Similar K-Ar age gradients
have been documented further NW in the Cordillera Real
(McBride et al., 1987) and on the NE flank of the Cordillera
de Carabaya of SE Peru (Kontak et al., in prep.). Farrar et
al. (1988) have termed this region of reset dates, which is
ca. 20 km wide and over 450 km long, the Zongo-San Gab&n
Zone, and have ascribed the overprinting to a predominantly
compressional tectonothermal event that occurred 39 Ma ago.
In this communication, we report two studies that bear on
the nature of the tectonothermal event. First, to establish
unequivocally the age of the Zongo pluton and its metamorphic
aureole, we present the results of U-Pb dating of zircons
from two phases of the pluton (Heinrich et al., 1988).
Second, to establish the three-dimensional nature of the
tectonothermal overprint, we report "Ar/"Ar dates of samples
collected above the valley of the Rio Zongo.

U-Pb and '0Ar/3aArdating


The two-mica, monzogranitic-to-syenogranitic Zongo pluton
comprises a pervasively foliated Kuticucho facies and, NE, a
weakly foliated, Sainani facies. Distinct zircon populations
were separated from samples of each facies, abraded to remove
material that may have lost lead during the tectonothermal
event that reset K-Ar dates, and analysed for U and Pb. The
analytical results, shown graphically (Fig. 2) yield ages of
FIG. 2a ZONG-8 FIG. 2b ZONG-7
KUTICUCHO FACIES SAINANI FACIE!3
OS 31oy
1 lQ~+lOOMr

'.06

04

.25 30 36 .l .2 .3 .4
207pb/266u 207pb,236u

intrusion (lower concordia intercepts) of 222 and 226 Ma for


the Kuticucho and Sainani facies, respectively. Poorly
constrained upper intercepts are suggestive of an inherited
zircon component of Proterozoic age.
173

New "Ar/"Ar total fusion dates for biotite and muscovite


from samples, collected at altitude above the valley floor in
the Zongo transect, have been integrated with previously
published K-Ar and "Ar/'Ar dates (McBride et al., 1987) and
unpublished r0Ar/3gArdates (Heinrich, 1988). The combined
biotite and muscovite data sets have been contoured in Figs.
3 and 4, respectively. The biotite l'chrontours@@ are sub-

FIG. 3. FIG. 4.
ZONGO BIOTITE CHRONTOURS ZONGO MUSCOVITE CHRONTOURS
W.Er6) W.E..S)

:t:
. .
4000

FACIES ..-.
MET*-
SEDS.
b
0 2 4 0 I 10 12 14km

horizontal and young with increasing depth, whereas muscovite


ltchrontoursl'
are vertical and young toward the NE.

Discussion and Conclusion

The essentially identical U/Pb zircon dates (225 Ma) for


the Kuticucho and Sainani facies of the Zongo pluton clearly
indicate a Triassic age of intrusion, and, in contrast to the
inference of Bard et al. (1974), imply that the wide meta-
morphic aureole in the metasediments NE of the Zongo body was
also formed at that time. This result supports our geologi-
cal observations which suggest that, in this transect, a
vertical cross-section through the Zongo pluton, with prog-
ressively deeper parts to the NE, is exposed.
The most plausible explanation for the intersecting mus-
covite and biotite K-Ar chrontours is that both patterns are
a result of heating from below, but that a ca. 90' rotation
(counter-clockwise looking NW) of the Zongo intrusion oc-
curred between the resetting times of muscovite and biotite.
The total resetting of the biotites near the NE margin of the
.-....-e.

174

intrusion occurred as the temperature of the rocks fell below


the biotite blocking temperature (ca. 250-C) at 39 Ma
(McBride et al., 1987). No muscovites in the Zongo transect
are totally reset and therefore the blocking temperature of
muscovite (ca. 350-C) was not reached. Thus, rotation of the
panel occurred while the temperature was between approximate-
ly 300 and 25O'C.
We attribute the tectonic rotation and the thermal effects
in the Zongo transect and the tectonothermal overprint else-
where in the Zongo-San Gaban Zone to crustal-scale shortening
associated with the Late Eocene Incaic orogeny.

Referenoes

Bard, J.P., Botello, R., Martinez, C. & Subieta, T. 1974.


Relations entre tectonique, metamorphisme et mise en
place d'un granite dohercynien a deux micas dans la
Cordillere Real de Bolivie (Massif de Zongo-Yani).
Cahiers de l'office de la Recherche Scientifique et
Technique Outre-Mer, Sdrie Gbologie, 6, 3-18.
Farrar, E., Clark, A.H., Kontak, D-J., & Archibald, D.A.
1988. Zongo-San Gaban zone: Eocene foreland boundary of
the central Andean orogen, northwest Bolivia and south-
east Peru. Geology, v.16, 55-58.
Heinrich, S.M. 1988. Geology and geochronology of the
Zongo River valley, Cordillera Oriental, NW Bolivia.
Unpbl. M.Sc. thesis, Queen's Univ., Kingston, 168 p.
Heinrich, S.M., Farrar, E., Clark, A.H., Archibald, D.A. &
Parrish, R.R. 1988. Age, uplift and thermal evolution
of the Zongo pluton, Cordillera Oriental, Bolivia. EOS
(abstract), v.69, 487.
Kontak, D.J., Farrar, E., Clark, A.H. 61Archibald, D.A.
Eocene tectonothermal rejuvenation of a Late Paleozoic-
Early Mesozoic terrain in the Cordillera de Carabaya,
SE Peru. Submitted to J. South Amer. Earth Sci.
Laubacher, G. 1978. Estudio geologico de la region norte
de1 Lago Titicaca: (Peru) Instituto de Geologia y
Mineria Boletin 5, 120 p.
McBride, S.L., Robertson, R.C.R., Clark, A.H. & Farrar, E.
1983. Magmatic and metallogenetic episodes in the
northern tin belt, Cordillera Real, Bolivia.
Geologische Rundschau, v.72, 685-713.
McBride, S.L., Clark, A.H., Farrar, E., 8 Archibald, D.A.
1987. Delimitation of a cryptic Eocene tectono-thermal
domain in the Eastern Cordillera of the Bolivian Andes
through K-Ar dating and "Ar/"Ar step-heating. J. Geol.
Sot. London, v.144, 243-255.
175

TECTONIC AND GEOTHERMAL HISTORY OF THE SUBANDEAN RANGES AND THE EASTERN
CORDILLERA OF SOUTHERN BOLIVIA.

Institut f&rGeoloqieder PreienUniversitatBerlin, Altensteinstr. 34A,1000Berlin 33, Germany

Resumen

Investigacionessobre la tectbnica y la geoternonetriamuestranque las tenperaturas nds altas


alcanzadasen la Cordillera Oriental tuvieron lugar duranteel Paleoaoicosuperior. En las Sierras
Subandinaslas paleotenperaturasnuncasuperaron10s 7OCguese alcanzaronen el Terciario.

KeyWords:Bolivia, Andes,EasternCordillera, Subandean


Ranges,TectonicEistory, Geothermal
Eistory

Thestudied area extendsalong2130lat.SbetweenVillanontesand S.Vicentein Bolivia. In the east


the section conprisesDevonian to Tertiary sedimentsof the Subandean
Rangesanda TransitionZoneto
the EasternCordillera. TheEasternCordillera is built up rainly of Ordoviciansediaentsoverlain by
Cretaceousand Tertiary strata and is borded in the vest by the Mtiplano. Altbougbthe lain
geologicalfeatures of the region havebeenrecognited,sany aspectsof the tectonic evolutionrenain
poorlyunderstood,andlittle data is available aboutthe geothersalhistory of the area.
176

The tectonic evolution of SouthernBolivia fros early Paleozoic up to the AndeanPhase, is


characterited uainly by veak epirogeneticsovemts. Wring Paleoxoicand early Mesozoictiles the
area vas situated at the vesteru border of the Brazilian Shield and, since Jurassic tiles, to the
east of the volcanic arcs. Epirogeneticsovexentsled to the forsation of several disconforsities,
vhichcm be correlated vith orogeneticphases:
- The lack of sedimentsof Llandeilianto Llandoverianage say be causedby an UpperOrdovician
OcloyicPhase; hovever, a correspondingthenal event in the studied area cannot, as yet, be
proved.
- Generally, paleoroic orogenetic wesents did not producestrong folding but caused a slaty
cleavagein deep-seatedrocks of Ordovicianand lover Silurian aqe. In the EasternCordillera
greatest depth and maximum tesperaturesvere reachedduringthe UpperDevonian(e.g. about28012
for Tremadocian rocks) folloved by uplift. AnUpperDevonianEohercynianPhaseis indicated by
three lines of evidence:1. in the EasternCordillera (e.g. the synclinal structure of Camargo)
veakly diagenetic Cretaceousrocks overlie, vith a very lov angle unconfonity, very-lov-grade
metamorphic Ordovicianstrata vith slaty cleavageand qua&-veins; 2. vhile the Silurian and
Devonianstrata do not shov significant lateral facies variations, carboniferoussedimentsthin
out tovardsthe EasternCordilleraandindicate that the vestera borderof the basin, near Tarija,
has experienceduplift since UpperDevonian/EarlyCarboniferoustines; 3. in someplaces,
Caxboniferous strata unconfomblyoverlie Devonian rocks(theextent of the stratigraphic gap is
not known).
- In the Subandean Rangessubsidencepersisted vith SOPinterruptions (Triassic? - LoverCretaceous)
until the start of the AndeanOrogenyin the Tertiary. Crustal shorteningled to significant
uplift of the EasternCordillera vhile intense lateral cospressionsvith folding and thrusting
characteritedthe vholearea.
Haturation studies on organic latter (reflectance-neasuresentson vitrinite and qraptolites,
infrared-spectroscopy)and clay sinerals (sainly illite-crystallinity) vere used for the
interpretation of thegeotheraalhistory. The maturityof sedimentsvaries betveenuudermature and
very-lov-qradesetanorphic.Thestratigraphic gapsof the OcloyicPhaseand (locally) the Eohercynian
Phaseare not markedby suddenincreases in rank, vhicb vould indicate a heating event, so that an
UpperDavonian(Eohercynian) beatingevent in the EasternCordilleracan only be provedindirectly,
as discussedabove.In the Transition Zonebetveenthe EasternCordillera and the Subandean Ranges
the rankdecreasescontinuouslyfror older to youngerstrata. burial curves shovthat in this region
the raximusdepth and temperatureswe not reachedbefore the beginningof the ReogeneQuechuan
Phase.The paleotemperatures of Devonian rocks at the easternmostpart of the Subandean Rangesnear
Villamontesdid never exceedabout 7OC.Thus, in this region the paleogeothermal gradient reached
about15C/km at the time of maximum heating (bate Tertiary) andrangedsignificantly belov 3OC/km-
at least since Permiantimes. In the Eastern Cordillera, hwever, the paleoqeothermal gradient
attained valuesof 3OC/ks or evenmorebefore the UpperCretaceous.
SEDIMENT MATURITY IN THE EASTERN CDRDILLERA AND SUBANDEAN RANGES OF SOUTHERN BOLIVIA

IC 1). r 0 IC
- ---*--7.- . .-.-.-o-. I detrital mica
Ic_____4-1 . . d___, l . Et
5 -. 5
\/ - -.__
--_
I - ._
ID -I 1 10
I
I

III l.
III 5. ~~--__
1
.-. . 1
I;1 &---- -___
.-.__ ---_ ~. ~__ c__-__w_-_
4 ---_
0 I
I -0

EASIERN
i,
Depth
w CORDILLERA
I
TRANSITION ZONE
I
SUEANDEAN RANGES
E
DePLh
I
km1 km1

D D

10 10

20 70

lzl Precambrian 0 Cretsceous-


lert1ary
m Neogene-Ouateroary
Volcanics
..-__m--

178

Conclusions

The tectonic and geothernal histories of the Eastern Cordillera and SubandeanRanges have differed
considerably since the UpperDevonian, although both are largely doninated by epirogenetic novenents.
From Ordovician to Devonian tines subsidence took place over the whole area. In the Eastern
Cordillera maxim burial depths and temperatures vere reached in the Upper Devonian/Lover
Carboniferousvitb a paleogeotbeml gradient around 3OC/km.In this area the Caxboniferousto Upper
Creataceousis generally characterired by uplift. Further to the east, hovever, subsidence or stable
conditions lasted until the Late Tertiary. Tbe SubandeanRangesvere relatively cool during this
tine span vith geotherwl gradients significantly belov SOC/kein the Hesoroic and values of about
15C/ks in tbe Late Tertiary.
179

STRUCTURE AND KINEMATIC EVOLUTION OF


SUEANDEAN THRUST SYSTEM OF BOLIVIA

Gerard HOrail (11, Patrice Baby (21, Marcimio Lbpez


(31, Jaime Oller (31,Oscar Lopez (31, Rend Salinas (3),
Thierry SemphrO (21,Garry Beccar (31, Humberto Toledo (3)

1 : ORSTOM, CP 9214, La Paz, Bolivia

2 : ORSTOM, CC 4875, Santa Crux, Bolivia

3: YPFB-GXG, CC 1659, Santa Crux, Bolivia

La structuration de la tone subandine de Bolivie emt db-


trite a partir d'une sgrie de coupes 4quilibr4es. On montre le
.rble des lithclogies et gaombtries shdimentaires ant6orog6ni-
ques sur la d&formation et on dticrit les transformations
g8om6triques acquises au tours de la deformation.

Key-words : Sub-andean belt, Bolivia, foreland belt, thrust.

The Subandean belt of Bolivia (13' 30'S - 15 30'S) is a


curved fold-thrust belt predominantly verging toward the east
and northeast (1,2). The structuration of this belt began in
the Late Oligocene and is still developing. The fold thrust
belt is bounded on its Andean side by the Main Frontal Thrust
(CFP), whereas its structures die out eastwards in the Beni
and Chaco plains. The Chapare buttress, which has a poorly
known structure, separates the northern Subandean (SAN) from
the southern Subandean (SAS) (3). On each side of the Chapare
buttress the pre-erogenic sedimentary pile (41 shows distinct
geometrical characteristics (fig. 11.

The SAN (fig. 2A1 is 60 to GO. km-wide and consists of.2


strips of different structure, separated by the Marimonos-
Toregua thrust. In the Southwest, vigorous reliefs are due io
thick sheets detached in the Silurian and redeformed by deve-
-.._.-,_~_^_.,_. _

180

lopment at depth of Ordovician duplexes. In the northeast,


broad synclines are bounded toward the foreland by a disconti-
nuous and complex thrust front produced by Ordivician duplexes
evolving to intercutaneous wedges. Shortening estimated by
restoratinn is 136 km, i.e 51%.

The SAS (fig. 2B) is 100 to 140 km-wide and also shows 2
strips of different structure, separated by the Mandiyuti
thrust. In the west, the structure is essentially due to

sheets detached in the Silurian, within which lift-off, fault


propagation and fault-bend folds developed. In the east, deve-
lopment of numerous duplexes generated antiformal stacks.
Total shortening increases from the Argentina border (70 km,
i.e. 33%) toward the North (136 km, i.e. 47%, at 21' 20'5; 139
km, i.e 53%, at 20' 5).

NW SElN S

Fig. 1 : The preorogenic sedimentary pile.

In the SAS, the low-angle slope of the basal decollement,


the multiplicity and regularity of the overlying decollements,
and the low thickness variability of competent levels in E-W
directions facilitated the eastward propagation of deformation
and the relaying of upper thrusts by deeper and more external
thrusts when the first jammed. In the SAN, in the contrary,
the high-angle slope of the basal d&ollement and the NE-
tapering of the lithological units affected by the deformation
led to frequent jamming of thrust propagation (piling-up of
thick sheets, reactivation of structures by more internal
thrusts).
181

I
E
Y
0
182

5rr
10 km

Fig. 3 : Thrust sequence in the Sierra de Caquiahuaca.

Construction of balanced cross-sections gives a regional


description of the Subandean structure (fig.21 and permits to
reconstitute . . .
the evolutlPn
._.
OT specific structures (fig.3) and
to show their lateral variation. In some cases (i.e. in the
southermost tip of the belt), development of easternmost anti-
clines (emerging or not) is due to formation of duplexes in
the Silurian where the paleogeoQraphic increase of the slope
of the basal decollement jammed the progression of deforma-
tion. In other cases (i.e. in the Serrania de Charagua), the
easternmost structures follow an increase of the slope of the
basal decollement due to flexion by thrust loading.

Geometry and kinematics of deformation in the Subandean


fold-thrust belt and foreland basin are controlled by (a) the
geometry oflthe pre-erogenic sedimentary units, which depends
on paleogeography of the basins where they were deposited and
on their:pre-Andean evolution , and (b) modifications of the
geometry of the belt during deformation.

References

(11 J. Oller, thesis UMSA, La Paz, 120 p., 1986, unpub.

(2) D. Roeder, Tectonics, 7, No 1, pp 23-39, 1988.

(31 P. Paby et al., C.R. Acad. Sci. Paris, t. 309, sCrie 1'1,
p. 1717-1722, 1989.

(4) J.M. L6pez, Inf. YPFB, No 1906, 1974, unpub.


-v---_

183

STRUCTURE OF THE CENTRAL BOLIVIAN ANDES: IMPLICATIONS


FOR OROGENY IN ANDEAN-TYPE MARGINS

Department of Earth and Planetary Sciences


Harvard University
24 Oxford St.
Cambridge, MA 02138 USA

Resume

Un estudio combinado de camp0 y de fotogeologfa de 10sAndes Bolivianos mostr6 que el


estilo de deformaci6n es tfpico de cinturones de escunimientos. El acortamiento cortical,
mk que 10sprocesos magmaticos, fueel mecanismo importante en el desarollo de 10s
Andes centrales. Este acortamiento es debido exclusivamente a la subducci6n de la litosfera
oceanica, sin intervenci6n de colisiones.
Key words: orogeny, crustal shortening, Bolivian Andes, balanced cross sections

Introduction

Exceptionally thick crust (55-70 km) and the high plateau of the Altiplano character-ire
the central Andes. While this remarkably thick crust has commonly been attributed to
mantlederived magmatic addition (Gill, 1981; Weaver andTarney, 1982). neither the
magnitude of this contribution nor the detailed style of deformation coincident with
magma&m have been assessed. This study describes the style and timing of deformation
in one part of the central Andes, the central Bolivian Andes, in addition to determining a
lower bound on the amount of crustal shortening that has occurred, as revealed by mapping
and balanced cross sections.

Methodology and Assumptions

Detailed geologic maps of the Cordillera oriental and Subandean Zone between
approximately 17.5 and 19OS,from 6315 to 67OW,were constructed from photogeologic
and field mapping at l:SO,OOO.A network of fourteen balanced cross sections was
constructed from these maps (Sheffels, 1990). The cross sections are constrained by
184

surface data only. Balance was determined on the basis of the consistency of bed lengths
and restorability of cross sections (Dahlstrom, 1969; Woodward et al., 1985).
Three assumptions ensure that the estimate of shortening is a minimum. 1) Subsurface
and eroded fold and fault geometries are required to be consistent with observed
deformational style. 2) Fault displacement along faults whose hanging wall cutoffs are
now eroded is assumed to be minimal; that is, hanging wall cutoffs are inferred just above
present topography. 3) Shortening is only calculated along the surfaces of decollement
inferred from the surface geology, the shallowest levels of decollement, although there is
evidence of deeper deformation.

Style of Deformation

The defoxmational style of the Bolivian Cordillera Oriental and Subandean Zone is
charac&ized by the thin-skinned deformation typical of foreland fold and thrust belts
(Bally et al., 1966). Although the observed style of folding and the steep to moderate dips
of most exposed faults in the study area have often been interpreted as indicative of thick-
skinned block faulting (Martinez, 1980), the repeated juxtaposition of distinct structural and
stratigraphic levels across those faults requires the dip of the faults to become shallower at
depth. Narrow, complexly-faulted anticlines typically define the surface expression of the
fault ramps. These structures and the presence of numerous backthrusts suggest a weak
surface of decollement.
Decollement is observed within three Paleozoic shale levels: undifferentiated lower
Ordovician shale, the Silurian Kirusillas Formation, and the middle Devonian Icla
Formation. These shale horizons lie within a sequence of Paleozoic, predominantly
marine, alternating sandstones and shales, which is overlain by variable, primarily
continental, Carboniferous to Recent rocks (Rodrigo and Castaiios, 1978; Pareja et al.,
1978). The level of decollement ramps up towards the foreland, with variations introduced
by tear faults and lateral ramps along strike. Distinct syles of deformation are observed in
regions defined by the level of decollement and the fault geometry. Along a particularly
large tear fault system, the Cochabamba basins have opened as strike-slip pull-apart basins.
The thrust front south of approximately 17.SS, the latitude of the bend or elbow in the
mountain belt, is defined by a triangle zone, while north of this latitude, it is emergent. The
origin of the bend can be explained, at least in part, by the presence of a wider Siluro-
Devonian depositional basin to the south; the thrust front has advanced further east where
SilureDevonian strata were deposited further east.

The age of the main deformational events appears to be Oligocene or younger


throughout the study area (Sheffels, 1988). Although the stratigraphic record is
discontinuous from the Carboniferous to the present, no large angular unconformities are
present until the Oligocene. Furthermore, Tertiary rocks exposed in the study area are
mvolvcd in deformation of the same style and strike as the older rocks. Older deformation
is indicatedin parts of the Altiplano, as well as to the north, in Peru, and to the south, in
Argentina, While the possibility of an earlier onset of deformation in the central Bolivian
Andes can not be definitively ruled out, the evidence suggests that most shortening
occurred since the Eocene.
185

A Minimum Estimate of Crustal Shortening

A lower bound on the amount of crustal shortening across the width of the mountain
belt was determined from five cross sections that form an east-west transect at
approximately 18015S. Well-constrained because the transect lies close to traverses made
in the field, this minimum estimate of shortening is 210 km.
The magnitude of this lower bound indicates that crustal shortening has played an
important, if not the dominant role in the formation of the Bolivian Andes. Volumetrically,
it can account for from two-thirds to three-quarters of the present-day crustal cross-
sectional area, implying that mantle-derived magmatic addition can only have contributed
the corresponding one-third to one-quarter. If the restrictive assumptions used to construct
the balanced cross sections are relaxed, increasing the estimate of shortening, or if the
evidence for additional shortening at depth and in the Altiplano is considered, then crustal
shortening could have built virtually the whole mountain range.

Conclusions
Thin-skinned deformation in the central Bolivian Andes has resulted in a substantial,
minimum amount of crustal shortening, with the following implications. 1) Orogeny in
Andean-type margins is not necessarily primarily due to magmatic processes. 2) The
convergence evident in the upper crust implies convergence in and therefore subduction
of the lower lithosphere, 3) Shortening in the Bolivian Andes occurred contemporaneously
with subduction of oceanic lithosphere; therefore substantial crustal shortening is not
uniquely the result of continental collision.

References

Bally, A.W., P.L. Gordy, and G.A. Stewart, 1966, Structure, seismic data, and erogenic
evolution of the southern Canadian Rocky Mountains: Bulletin of Canadian Petroleum
Geology, v. 14, p. 337-381.
Dahlstrom, C.D.A., 1969, Balanced Cross Sections: Canadian Journal of Earth Sciences
v. 6, p. 743-757.
Gil13ib 1981, Orogenic Andesites and Plate Tectonics: New York, Springer-Verlag, p.

Martine;, C., 1980, Structure et evolution de la chaine hercynienne et de la chalne andine


dans le nord de la Cordilltre des Andes de Bolivie: Paris, France, O.R.S.T.O.M.,
Travaux et documents de LO.R.S.T.O.M., no. 119 ,352~.
Pareja L., J., Vargas F., C., Sutlrez S., R., Ball6n A., R., Carrasco C., R., and
186

Weaver, B.L. and Tamer, J., 1982, Andes& magma&m and continental growth, in
Thorpe, RS., Andentes: Oro enic Andesites and Related Rocks: New York, John
Wiley & Sons Ltd., p. 639-6 %2.
Woodward, N.B., S.E. Boyer. and J. Suppe, 1985, An Outline of Balanced Cross
Sections: GSA Short Course Notes, GSA Annual Mtg., Orlando, Florida, 172 p.
187

NAZCA SLAB RETREAT VERSUS COMPRESSIONAL


DEFORMATION IN THE CENTRAL ANDES SINCE LATE
OLIGOCENE TIMES

*+
Pierre SOLER * and Michel SSBRIE&

+ ORSTOM. UR 1H. 213 rue Lafayette, 75010 PARIS, FRANCE

+* CNRS URA 730, Bgr 509, Universitk Paris&d, 91405 ORSAY CEDEX, FRANCE

Resume

Darts les Andes Centrales, les deformations compressives sont essentiellement le resultat d&e-
nements brefs qui s&parent des p&h&s de relatif calme tectonique. Ces evenements correspondent a des
instabilit6s dans la dynamique de la subduction andine qui semblent contr4lees par des a&s de la migration
de la fosse o&nique vers IOuest.

Key words : Cenozoic, Central Andes, slab retreat, subduction, tectonics.

Introduction

In Central Andes, both seismological and neotectonic observations indicate that most of the present-
&y deformation is compressional and occurs on both sides of the Andes, i.e., in the fore-arc and the retroarc
sub-Andean zone. The Andean Cordillera itself is very weakly seismic. Field observations allow the
observation of Recent and active crustal deformation within the High Andes. Most of the active faulting is
roughly located along the fault system that limits the High Plateaus from the Cordillera Oriental; some occurs
within the Cordillera Occidental or, in southem Peru, along the limit between the Cordillera Orxklental and
the Andean fore-arc. Most of the crustal deformation is character&d by normal faulting, except within the
Cordillera Oriental of central Peru which is located above a flat-lying slab segment. Huts, the High Andes
and the Pacific Lowlands are characttxixed by roughly N-S trending extension and are bounded on both sides
by compressional belts with a roughly E-W trending compressional regime. This state of stress may be
explained as the interaction between the boundary forces due to the convergence between the Naxca and the
South American plates and the body forces due to topography.
While subduction has been a continuous process since late Oligocene times, compressional
deformations appear mainly to have occurred during short-lived events. This apparent discrepancy may be
explained by variations in the oceanic slab retreat.
188

Main tectonic features of the Central Andes since late Oligocene times

Since at least 26 Ma, Andean deformations have occurred within a morphostructural framework
roughly similar to the present-day one, and the sub-A&an Lowlands have acted nearly continuously as a
retroarc foreland. where most of the Andean shortening has been accomodated by underthrusting of the
BraGhan shield beneath the Cordillera Oriental. In contrast, the High Andes and the Pacific Lowlands are
generally much less deformed than the sub-Andean retmarc foreland. One of the major problem which is still
in debate is to determine whether or not Andean tectonics has been continuous there. The observation. at a
regional scale, of toughly synchronous angular unconformities fmt suggested that tectonics had been
charactcrixed by short-lived tectonic phases, separating periods of tectonic quiescence. More recently, the
continuity of the subduction process, and the observation of syndepositional folding led some authors to
consider that tectonic phases corresponded to long-lived periods of compressional deformations in the whole
Andean domain. and that the short-livad, general&d compressional tectonic events were only climax pulses
within a continuum of deformation. Direct and indirect evidences for discontinuous tectonic events,
responsible for faulting, folding, erosion and subsequent unconformities, do exist in the High Andes and the
for&arc basins. For the last 5 to 6 My, successive compressional and extensional deformations have been
observed both in the intra-cordilleran and fore-arc basins of Peru and Bolivia (discussion and references in
Me&r et al., 1990). Moreover, Cenozoic angular unconformities are accurately dated at some places, with
time-span of approximately 1 My. In many localities however,
189

(1) Between 30 and 26 Ma, a major reorganization of plate dynamics in the eastern Pacific Ocean
took place and the convergence rates hap been very low untill the breaking up of the Farallon plate into the
Naxca and Cocos plates at ca 26 Ma. Oligocene tectonic activity is still poorly known. but it may be inferred
that no major deformation, neither compressionalnor extensional, occurmd during this period.
(2)Forthelast26Mytheconvergencehasbeencharactcrixedbyhfghrates(meanramof11.0f0.8
cm/year). a direction nearly orthogonal to the Peru-Chile trench (whim variations less than ilOo from the
present-day direction). and a lithospheric age of the slab at the trench in the range 35 - 65 My. The only
singular event during this period is the arrival of the Naxcaridge at the Peru-Chile trench at ca, 4 Ma, causing
shallowing of subducIion in central Peru.
During Urisperiod, the Andes have experienced a general compressive regime that appears to have
beennearly cominuous in the sub-A&an retrosrc foreland. As the short-lived generalixed compressional
tcctonic events occurred between longer periods of relative tectonic quiescence, they should correspond to
ineatabiities of a steady-state tectonic regime. The state of stress within the Andean Cordikra should be
controlled by the balance between boundary forces, that are due to convergence, and body forces due to
topography. As the High Andean Cordillera was mainly uplifted during the Miocene no major variation of
body forces may be invoked to explain the Ptiocene to present-day evolution of the Andean state of stress.
Even during the
190

Generalizedsketchofsubductiondynamicsin the CentralAn&s since late Oligocene times


A- Steady-statesituation,where mostof the westwardmovementqf the South-Americanplate
(VgoA& is accomo&ted by the retreat of the Nasca slab (i.e., the westwardmigrationof the
trench, VT - VSo&. The retroarc shortening (VR) is weak. Extensional tectonic regime
prevailsin the High Andes and thePacijicLowlands.
B- Situationduring a short-livedgeneralised compressionalevent. Most of the westward
movementof the South-Americanplate (VSOAM)is accomodatedby the retroarc shortening
(VR - VSOAM); the retreat of the Nazca slab is weak or even blocked (i.e., the westward
migrationof the trench VT - 0). and the slab is broken.Shorteningalso fleets the High Andes
and the PacificLowlands.

References

Mercicr. J.L., Sebrier, M., Lavenu, A., Cabrera, J.. Bellier, 0.. Bonnot, D., Dumont, IF., Huaman, D., and
Machart. J.. 1990, Changes in the stress pattern above a subduction zone of Andcan type: the Andes of
Peru and Bolivia during the PliocenePleistocene. J. Gcophys. Res.. submitted.
Stbrier, M., Lavenu. A., Fomari, M., and Soulas, J.-P., 1988, Tectonic and uplift in central Andes (Peru,
Bolivia, northern Chile) from Eocene to Present. Gkxiynamique, 3, p. U-106.
Sebrier, M.. and Soler. P., 1990. Tectonics and magmatism in the Pcauvian Andes from late Oligocene time to
the Present. Geological Society of America Special Paper Andcan magmatism and its tectonic setting,
R.S. Harmon and C.W. Rap&. eds., in press.
,~. -.-. . ---

191

RELAtIONSHIPS BETWEEN THE COCHAS-GRAN BRETAAA REVERSE FAULT


AND THE AZULCOCHA Zn-As-(Au) ORE DEPOSIT, CENTRAL PERU

C4srr MU602 & Lluis FONTBOT&

Mineralogisch-Petrographisches Institut der Universitat, Im


Neuenheimer Feld 236, Postf ach 104040. D-6900 HEIDELBERG.

Key Words: Zn-As-(Au) deposit, hydrothermal karst, fracture


contra 1, central Peru.

Setting and tectonic evolution

The Azulcocha area lies


between the eastern part of
the Western Cordillera and
the high plateau of the Peru-
vian Andes. about 45 km west
of Huancayo (about 12S and
7503W. at 4500 m above sea
level, Fig 11. The Azulcocha
orebody is emplaced along the
Cochas-Gran Bretafia fault
(Fig. 2). Between 1968 and
beginning of 1986 1.5 million
tons with 18% En and 2% As Fig. 1. Location of the
were exploited. Azulcocha area
Mesozoic and Cenozoic
sediments as well as Cenozoic granite-monzonite intrusive-
rocks occur in the studied area (Figs. 2 and 3) . The Cochas-
Gran Bretana fault constitutes the most important tectonic
element. Its main movement. as well as that of similar faults
like those of Carmen Chico-Cerro de Pasco and of Casaracra-
Junin Lake, is associated with the Quechua II and III phases
(MEGARD.1979: SOULAS, 19751.
In the mining area the Cochas-Gran Bretafia fault trends
E-W and dips between 30 and 45S (Fig. 31. This is a
transverse orientation with respect to its general NW-SE
...----

192

FIO. 2. GEOLOEICSKCTCH Of THE J*lUrutUAtl-~ULC~HA-~wulrlIA AREA. INDICATINQ IHC


MAIN TECTONIC ELEMENTS (HODIF. AFTER Em. 1919 AND m. 197!i)

FIG. 3. N - S CROSS SECTION OF AZULCOCHA ORE DEPOSIT


*...... . ...-- .---.-

193
direction (Fig. 2) . The transverse segment is characterized
by overthrusting in contrast to the dominant dextral slip
movement of the longitudinal Parts of the fault, The south
block consisting of limestones of the Condorsinga Formation
is thrusted over sandstones of the Goyllarissuizga Group.
East and West of the mining area the fault affects also the
Cercaeusuio and Chaucha Formations. The Cochas-Gran Bretafia
fault represents actually a set of Parallel faults which
strongly deform the affected rocks. Silt intercalations of
the Goyllarissuizsa sandstones react Plastically Producing
apparent dikes. Carbonate rocks are intensely fractured
and brecciated. favorins subsequent karstification. N30-40-E
faults with left-slip displacement are interpreted to be
linked to N-S shortening of the Quechua II Phase which also
formed the E-W folds and deformed older folds. The com-
pressive Pulses of the Quechua III Phase Produced N-S folds
(Azulcocha mine syncline), and a fault system trending N120-
130-E (sinistrall and N30E (dextrall .

The pzulcocha ore deposit

The Azulcocha Zn-As-(Au) deposit consists of an ellipti-


cal orebody (30Ox50x160ml emplaced in Condorsinga limestones
and controlled and limited by the Cochas-Gran Bretana fault
(Fig. 3). The ore occurs dominantly as open space filling, in
part cementing carbonate fragments in collapse breccias. The
ore appears mainly as colloform textures consisting of verr
thin whitisch-yellow to dark-brown sphalerite layers (scha-

--_
lenblende). marcasite, pyrite, and melnikovite (MUROZ. 1986).
Four depositional stages have been distinguished (Fig.
41 * Massive aggregates of pyrite and marcasite, in Part as

r
subidiomorphic grains, and disseminated sphalerite in 1ime-
stone are assigned to stage
I. Stage II comprises mainly
schalenblende, subordinate
barite and marcasite, and
minor amounts of salena and
Pb-As-sulfosalts (jordanite.
baumhauerite and duf renoy-
site). This stage displays
brumhruwl
tr several repetitive sequences
which of ten are interrupted
by brecciation and cemented
by a next deposition puls.
Pyrrhot i te with birds eye
texture.is occasionally ob-
served. Stage III consists
of marcasite and melniko-
Fig. 4. 8chematic paragonetic vite. Microprobe analyses
sequence _ show
_ that they contain. UP to
500 PPm Au and UP to 7% As. Abundant realgar and auripigment
with small acicular inclusions of Pb-As-sulfosalts occur also
in this stage III. Stage IV represents supergene oxidation
Products. Microprobe resu 1ts indicate that the first
. __.._.I_,* ,.,

194

sphalerit e generations are richer in Fe (UP to 1.5%). Cd- and


Mn-va lues range between 0.3-0.5% and 0.05-0.195 respectively.
As a whole the Azulcocha mine shows a relatively 1arge
variety of ore minerals and a broad spectrum of elements
(Tab. 1). The relatively high contents on Au should be
pointed out.
Table 1. Trace element contents In aelected ore samples from the Azulcocha
mine. ICP and except for
ICP-MB 1) analysis. Au (Acid lea-
ching/AA). Results for elements mainly in gangue minerals
contained
(Ca. Ug, Al, K) are aleo given. Other elements not included in the
table: B < 2Oppm. Na < 100ppm. P < 100ppm. Ti < 100ppm. MO < 6ppm.
Rh < 0.2ppm. Ag < 2ppm. Ba < Sppm. La < 2ppm. Re < 0.2wm. OS <
tppm. Ir* < O.lppm. Bi < 0.2ppm. IX* < lppm and U < 0.4ppm.
20 - zinc ore. IO - Iron ore.
-_- -___--- ___-_--__- -____--_-______.
Suoll Dwr. Ia n Al I, lb 61 b u In fb 1, nq I1 co II c4 v cr k Cl Fq bl K
I I I I I 444 448 444 441 $44 444 444 441 40) 444 JOI VI 441 404 I I I I

U-444-41 ZO-IO N 1.47 1.b lI,S 4.42 4 !2 111 4.1 42 4.4 14 21 21 4 7 2 14 121 4.42 0.41 4.11 4.45
U-444-42 10 4.1 4.42 4.1 1.4 4.43 2 1 S 4.1 1 4.1 (1 a 123 $7 LL 44 22 24 4.4S 4.41 4.:4 4.47
II441 IO-IO H 1.3 )I Lb 4.44 4 2b lSb4 14.4 4S4 2S.l (4 44 IS 4 I I 2 I%4 4.41 4.41 4.42 4.41
U-4+2 10-M W I.24 M 7,b 7.11 4 22 1241 3.2 S7S 2.2 71 42 b I 4 S b 4b4 4.11 4.41 4.4s 4.43
U-4-43 20-M N 4.23 4.4 12.4 be!4 4 27 7S4 7.4 111 4.S b4 14b 15 1 b b 14 S54 4.24 4.11 4.4S 4.42
K-4-44 IO-10 )4 4.44 4.5 11.3 43 2 14 444 3.4 14 4.b I7 34 S 1 S b 4 74 4.S4 4.15 4.41 4.41
U-o-03 IO-10 )4 4.14 4.4 14.1 b.33 4 2S 2SW 4.S 143 1.4 44 111 7 1 I 4 b 44 4.24 4.01 4.41 4.42
U-4+7 IO-IO1.S 4.44 4.1 11.4 I.43 2 I 77 4.S I4 4.2 4 S lb IS b I4 24 77 S.14 1.24 4.11 4.44
U-44-41 N-IO H 1.24 H 7.1 Lib 14 14 4Sb 13.2 bS? S.2 114 74 I4 4 S 2 S 14444 4.11 4.41 4.41 4.01
U-44-42 IO-10 )4 4.11 2.1 4.b b.!S 2 S4 117 S.7 11 4.) I7 74 7 1 4 1 s sb4 4.17 4.44 4.43 4.41
u-44-01 IO-IO w )2 14 4.3 4.41 1 111 2444 34.2 174 I.b b4 24 I1 4 4 I b 1424 4.41 4.41 4.44 4.42
W4-41 IO-10 1.: 4.14 )4 11.4 S.IS 2 1 IS 1.4 25 1.b :O 122 13 I 1 1 2 1 4.4b 4.41 4.44 0.41
U-IO-42 IO-20 )4 )2 )4 lS.1 4.01 5 12 423 21,s 44s 4,s 74 SS I7 5 7 5 3 II40 4.41 4.01 4.40 4.43
42.IO-OS IO-20 )4 4.25 4.4 12.4 I.44 S SI 49b b.7 74 2.1 46 44 lb 7 7 4 7 114 4.41 4.01 4.07 4.4s
4Z-WI IO-10 )4 )2 14 7,1 b.44 2 I 18 4.5 14b 7.2 14 194 7 I J 2 S 774 4.11 4.41 4.02 4.41
bZ-IS-42 14-10 >b )2 )4 14.4 4.41 I I1 24b4 1.2 1174 1902 121 279 11 I S 1 b 1104 4.4s 4.41 4.42 4.01

Homogenization measurements of fluid inclusions on ba-


rite yield two populations with temperatures of 180-230 and
250-29OC. Alterations is not observed in the mine itself:
however. silicification and dolomitization linked to N30E
fnults is recognized south of the ore deposit.
The Azulcocha orebody has been precipitated in a 1arge
cavity produced by dissolution and collapse brecciation
through hot fluids along an overthrust surface of the Cochas-
Gran Bretafia fault. The occurrence of ore fragments sealed by
new sulfide generations indicates that the fault experienced
some movement during ore formation. However, ore deposition
appears to have started after the main fault movement during
the Quechua II phase, for ore brecciation is subordinate com-
pared to that in adjacent mylonitized host rocks.

NESMD,F, 1197911 Bol, Ink 6tol. tlin, Ilet,, Lima, v, 8, S@rie D, 277 D,
AESARD! F, (19B71 I InI IKINKR, J,Y,E, I FRANCHETEAU, J, IEds.1 Awr. 6rophyr, Unlon, Washington,
p, 1108-1117,
IVJXlZ, C, (1986)r Inter, report Cir. Hinera 6ran Brstaiia, 31 o.iunoubl,l,
fwlz, c. l1988)r 11, teonisg, Lateintrerik Koll., Hannovw,Taqungshsft, p, 101,
SOULAS, J,P, 11975)Bull, Inst. Fr. Et, And,, Lira, v, IV, p, 121-156.
.-.. .._.B..

195

DEFORMATION AND AND GRANITOID EMPLACEMENT ALONG A


rEy;R CRUSTAL LINEAMENT: THE CORDILLERA BLANCA,
.

MICHAEL. P. ATa AND NICK PETFORD

Department of Earth Sciences, Universtiy of Liverpool, PO. Box 147, Liverpool,


U.K.

Resumen

El Batolito de la Cordillera Blanca representa el ultimo evento plutonico en el ciclo


Andino. Este pluton fue emplazado a lo large de un mega-lineamiento durante un
period0 de extension, ocurrido hate 5.2 Ma.
El margen occidental ha sido deformado por transpression durante el event
tectonic0 regionale Quechua 3, probablemente relacionado con un alto grad0 de
convergencia de la Placa de Nama.

Introduction

The Cordillera Blanca Batholith of NW Peru (Pig. 1) together with contemporaneous


ignimbrites of the Yungay formation are the youngest magmatic events of the Andean
cycle in Peru. They overly the thick crustal keel of the Andes. The Batholith strikes
NNW-SSE, and is 200 Km long, reaching nearly 7000 m in height near Yungay.
The western margin abuts against a major fault system which extends southwards
for a distance of nearly 400 km (Pig. 2.). To the west of this fault the
contemporaneous ignimbrites lie within the 15 km wide Callejon de Huaylas
intercordilleran basin, a subsiding graben of at least Pliocene age. K/Ar dates of 7.6 -
6.2 Ma indicate the graben was already in existence by that time.
K/Ar ages of the Batholith vary from 13.5 to 2.7 Ma indicating magmatism has
occurred along this lineament over a long period of time. The Batholith and
ignimbrites are clearly related to this major crustal fault system, which is still active,
and now shows a normal movement sense, which towards the south has an en-
echelon (N 140 E) left hand pattern (Sebrier &al, 1988). However, slip vectors
indicate a sin&n-al strike slip component, producing not orthogonal extension but
nearly N-S extension as shown by the recent kinematics of the fault zone. This
extension postdates the Pliocene compressional period.The Cordillera Blanca and its
southerly extension (Pig. 2.) lie above a high conductivity zone (0.1 SM-1) which
parallels the general Andean trend and lies 50 km or less from the surface (Wilson,
1985) . It is best explained by the presence of a small melt fraction (0.05) in highly
conducting rock. This may correlate with two low velocity zones found by Qcola &
Meyer, (1972) at 10 and 35 Km, also explained in the same manner. Geothermal
springs along the line of the fault and a measuredhigh geothermal gradient of 40-60
oC /km (Uyeda & Watanabe, 1970) confirm high heat flow along this megastructure,
and therefore hot material at a high level in the crust
..-. .--WI

196

4QAr-3gAr Step heating ages

A &ta&d 4oAr-3gArstudy confirm a late Miocene (5.20 Ma) age for the central
region of the Batholith, and shows that the Batholith youngs in a westerly direction
away from the undeformed core towards the intensely deformed western margin.
This younging, as revealed by K/Ar dating may be explained in terms of argon loss
due to deformation and recrystallisation as deformation increases towards the fault.
The timing of deformation is consistent with Quechua 3, the last major
compressional event in northern Peru.

Intrusion style, deformation and tectonoic environment.


The late assymmetricaldeformation (Pig. 3.) seen within the Batholith is very
different to that seen in ballooning diapirs with symmetrical shear strains, and
indicates the strong fault control on emplacement. Kinematic indicators imply the
Cordillera Blanca fault system has suffered sinistral strike slip motion during the
upper Tertiary. The structural_data summarised in Fig. 3. emphasise the deformation
occurred after the emplacement. The ductile/plastic to brittle deformation style seen
in the batholith is consistent with rapid hanging wall uplift of 1.4 mm&r.

Emplacement of the Batholith

It is considered that the source region was tapped at depth by the deeply penetrating
crustal lineament whose surface expression is the Cordillera Blanca fault system.
Melting was induced by decompression followed by fracture controlled ascent of the
liquid to high structural levels, then emplacement within the Jurassic (Chicama)
basinal rocks. Post emplacement fault movements, superimposed upon earlier
magmatic fabrics deformed the wastem margin and are essentially compressional in
form. Quaternary and recent faulting at the western margin of the Batholith near the
graben structure indicates a N-S stretching reaching 8% @onnot, 1984)

Extension and compression in the high Andes of Northern Peru


Analysisof the sedimentary, volcanic and structural data indicate there were 6 discrete
compressional periods from the upper Eocene to the present, of which Quechua 1 (20
- 12.5 Ma), Quechua 2 (9.5 - 8.5). Quechua 3 (6 Ma) and a Quatemary phase at ca
2.0 Ma are relevant (Megard, 1984). They have been related to periods of high
convergence rate. Basins forming between these periods were extensional.
Magmatism can be both extensional and compressional, although we think the
compressional periods were late in the intrusion sequences, as noted above for the
Cordillera Blanca Batholith. The presence of undeformed granite to the east of the
fault with an increasing apparent age as shown by the K/Ar data, together with the
Tertiary and neotectonic and argon data, suggests sequential intrusion in a strike-slip
environment. This was probably transtensional, with later intrusion occurring
towards the fault zone as the movements became transpressional. This may suggest a
half-flower type structure for the Batholith, with successive intrusion during the
transtensional periods.
197

Conclusions

In general, the prevailing regime during the Miocene, Pliocene and Pliestocene
appears to have been extensional with short lived compressional events perhaps
induced by convergence. The Batholith was intruded during extensional periods.
Strike-slip movements during the intrusion interval of the Batholith (13 - 3 Ma)
correlates with the predicted strike-slip movements from Nazca plate reconstructions
indicated for the Andean margin by Snyder (1987).

References
BONNOT, D. (1984). Neotectonique et tectonique active de la Cordillera Blanca et
du Callejon de Huaylas (Andes nord-peruviennes), thesis , University Paris-Sud,
Orsay, 115~.

MEGARD, F. (1984). The Andean erogenic period and its major structures in central
and northern Peru. J. Geol. Sot. London. 141, 893-900.

OCOLA , L.C. & MEYER, R.P. (1972) . Crustallow velocity zones under the Peru
Bolivia altiplano. Geophys. J. Astron. Sot. 30, 199-209.

SEBRIER, M., MERCIER, J.L., MACHARE, J., BONNOT, D., CABAREZ,J. &
BLANC, J.L.. (1988). The state of stress in an overriding plate situated above a
fault slab: the Andes of central Peru. Tectonics. 7,895928.

SNYDER, D. (1987). New Nazca plate reconstructions and implications for


intermontaine basin evolution in the Andes. In: Pacific Rim Congress 87. Aust.
Inst. Mining. Metal. 923-926.

UYEDA, S & WATANABE, T. (1970) Preliminary report of terrestrial heat flow


study in the South American Continent; distribution og geothermal gradients.
Tectonophysics. 10, 235-242.

WILSON,C.D.V. (1985). The deeper structure of the Central Andes and some
geophysical constraints. In Magmatism at a Plate Edge. (eds) Pitcher, W.S.,
Atherton, M.P., Cobbing, E.J. & Beckensale, R.D. Blackie Halstead Press, 13-18.
...^. WX_.^ (..1I .__.. ____._

Fig. 1. Gcologicsl sketch map of the Andean Batholiths of Peru.

Fig. 2. Enlargement of boxed area in Fig. 1. showing


the relationship bctwcen Ihe Cordillera Blanca fault
system. Batholith and associated stocks.

mylonilcs, ductile shear z6nes

Jolnl Inceese. SW dip

muscovilo lncomlno

biolite ali~nmcnt

schleiren mcwcrysl alignment

flow fabric

I
1 .2 km 3 4 5

Llanganuco Section

Fig. 3. Synoptic diagram showing the variation in deformation style in the granilc towards Lhc Tauh
system.
199

Martin Litherlsnd and John A. As&n

British Ceoloaical Uisslon,


c/o British F&assy,
Casllls 314,
Quito, Ecuador

Studies of the metemor~hic and plutoaic rocks of the &stern Cordillera


of Ecuador Indicate the presence of ophiolitic rocks, island arc terranes,
S-tyke branltes and -jot napbe complexes. The final accretion vas
MidJurassic In age but this may have folloved an older collision.

Key vorda: ?&ado):, Eastern Cordillera, Jurassic, collision, naypes,


s-type.

Introduction

An Anglo-Rcuatoriau technical co-operation project involving l,ersonael


from the British Geolog;ical Survey, funded by the Overseas Bevelokment
Administration, London, sad personnelffuadisb; from the Ecuadorian Institu-
te of Mines (INRMIR), has been studying the geology and mineral potential
of the Eastern Cordillera (Cordillera Real) of Ecuador since 1966. Specl-
fically , the project is devoted to the study of the metaaorkhic and older
klutonic rocks, the bulk of which are considered OS prPCretaceoua In
abe, rather than as equivalents of easterly and westerly umetazorkhosed
Crotaceous formations as subbested by Feininber (1975) and Bristow (1973)
respectively .
Evidence for kre-Cretaceous collision cm be summarised within a descrip
tion of a beneral vest-to-east aeotraverse across the cordillera, uhlch
Is divided into the Western, Central aad Eastern tectonic belts and indi-
vidual lithotectonlc divisions (Aspden et al, 1967; 198&r).

The Western belt

This comprises the low-pnde metamorphic rocks of certain inliers within


the Inter-Andenn graben , and the western slopes of the Eastern Gordillcrn
eastvards to the Baiios Front. The ucst-to-t?ast lithotoctonic divisions
are 0.3 follows:
The Guamote division is composed of a continental quartzite/phyllite
nssociotion of shsllow water, nolasoic sttinity, with o shullou-di~~in~
tectonic rc&e indicating: westward thrustin~,.
The Peltctec division compises u stccply-dippin~ tectonic mcl~n~e
of obhiolitic aifinity containing basalt, ~obbro, picrite, scrpntinite,
chcrts and scdirrcnts; with slices of Ties LaLuws-type Irarlite. The
200

eastern Inter-Andean graben fault marks the Cenozoic rejuvinntlon of


the Peltetec fault.
Tha Uakuazo diviaioa is of ateeyly4lpylnp graded turbidltes, chcrta
and andesltic volcanica of foreorc affinity. Chert samples hsvc ylclded
Csllovian-Cxfordian palynoflora and reworked Ordovicinn ncritarcha.
The Alao-Pauta division la of ateeply-dippinb: mete-andcsltlc greenstones,
og&merates, tuffs and graphitlc and politic phyllites of island arc
affinity. The Tampanchl msfic-ultr=flc complex intrudes this sequence
but ia probably an Early Terttiry Alaskan pike.

lba central belt

This lies between the Baiios and aub-Andean tectonic fronts and forma
tha msln mass of the hi&h cordillera.
The Boiioa front everywhere marka a 1ltholog;ical &an&e from Aloo-Paute
greenstones to aemipelites or Tres lagunas granite. In the north this
is also marked by a metamorphic change to medium Wade schists and para-
kneiases . In the south the front is marked by a &JO wide mylonite zone
and a mjor bold belt related to Cenozoic volcanic activity.
Tha Trea lagunaa granite is found almost throughout the cordillera east
of the Baiios front. Xt is essentially a tvo-mica granite vith S-typa
affinites, associated in places with veak Sn-W indications. Blue quartz
and smoky K-feldspar are characteristic; tarnet and cordlerite msy be
present. In the north the granite is essenttilly geissic and may be
fo:nd in zones with incfpient m&matlantlon. In the south the branlta
intrudes low-gade rocka and Is marked by shear zones vlth auken ~nelsa
separotink belts vlth little or no deformation. Rb-Sr plots for Tres
lacunas Indicate an Early Jurasalc age.
Nappa complexes may be found east of the Tree lagunas granite. The best
example Is east of Quito in the north of the cordillera vhere a major
flat belt extends for 20 ion across the strike to the sub-Andenn front.
The &3kjdXi comprise individual thrust slices of thin serpentinite, se&p+
lites, graphite and pelitic schists and a cdcic IUQIIetite skrun packet
which forma a rov of isolated high level klippes. These thrusts ride
eastvards over a belt of subvertically lmbricated rocka comprising the
Azsfran Urlc-aUaUne plutona and envelope Sreenstones, turbidites and
calcareous flysch: possibly another Island arc terrane.

Tha Pastern belt

This occupies the Andean foothill zone between the sub-Andean front in
the vest and the limit of Andean (late Miocene) thrustink In the east.
Rocks of the Central bolt alon& with Cretaceous formations of tho Amazonlc
cratonlc cover are thrust steeply over the essentially undeformed
Abiwga-Zamora plutonk-volcanic continental arc which Rb-St and K-Ar
studies Indicate as Esrly/MidJurasslc in age.
This narrov Cenozoic thrust belt (prcsumnbly Late-Niocene) is believed
to represent a rejuvlnotion of the major pre-Cretsct*us sole thrusts
vhich essentially murkcd the eastern limit of the pre-Cretaceous oroLeny.

Interpretation

Previously the Western Cordillera and coastal plain of Pcundor have been
interpreted aa ot accretionory oribin &brat et al, lYb6) of Late Crcta-
coouslCenozoic abC, rather than related to the simple Andcon model.
The discovery ot ophiolitic rocks, %tykO &pinitCS and nupkc coml~lcxes
in the Enstern Cordillera indlcatce even older accretionory/collisionol
wonts.
201

The vest-to-east llthotectonic sequence of the Weetcrn belt has been


interpreted as the MidJurassic accretion of the Alno-Poutc (island arc)
division between the S American plate to the east and, the Cbauchn-Arcni-
118s plate to the west, across the Poltatcc okhiolitic suture (Asplcn
et al, lYt171 1988). On this basis the Abitoaua-&more ylutonic chain
could be lnterbreted es a continental arc rclotcd to this accretion.
A major problem, however, Is the interpretation of the stype Tres Lo~unas
Lranite, nakpe comylexes and serpentinites the Central belt. These could
either relate to the some Mid-Jurassic Alao-Paute collision, or to an
older (Late Triassic/Early Jurassic) collision which would then define
the Baiios front as a MIdJurassic continental margin. The eastward tecto-
nic tranqort of the napke complexes might Indicate the overriding of
the South America plate by the colliding terrane, in contrast to the
Cenozoic subduction pattern.
kork will shortly bebin on the El Ore metamorphic basement block of the
IIuancabamba Deflection and it is hoked that these studies will shed fur-
ther llbht on the kre-Cretaceous history.

Referencea

LA*, Litherland, H., Duque, P., Salaxar, E., Rermiulez, R. and


viteri. P.. 1967. Un nuevo cintur6n ofiolitico en la Cordillera
Real, huador, y su posible sfbnificacidn rebiOSS1. Po~ithcnicS,
Quito, F&nografia de Cool., 5, Vol. XII, No. 2, kk 81-93.

&A., Litherlsnd, H. and Selaxar, E., 1968. Una interkretacibn


kreliminar de la historia colisional de1 Centre y Sur de1 Ecuador
y posibles controles para la geologia cenozoica y de minernlixa-
cibn kolimetAlica. PolitAcnica., Quito, Monografia de Geol.,
6, Vol. XIII, No. 3, yk 49-75.

C. R., 1973. Guide to the beoloby of the Cuenca basin, southern


Ecuador. Sot. Ecuat., de Geol. y Geof., Quito, 54 kk*

Feininger, T., 1975. Origin of petroleum in the Orieate of kuador. Amer.


Ass. Pet. Geol. Bull., Vol. 59, kk 11664175.

Lebrat, H., Hegard, P. and Dupuy , C., 1986. Pre-orobenic volcanic assem-
blages and position of the suture between oceanic terrancs and
South American continent of Ecuador. Zbl. Geol. Palaont., Teil,
1, Vol. 9/10, kp 1207-1214.
203

ALLOCHTONOUS TERRANES IN NORTHWESTERN ECUADOR

VAN THOURNOUT, F. l, QUEVEDO, I_. * l

l Belgian Technical Assistance in Ecuador (AGCD)

l * lnstituto Ecuatoriano de Mineria (INEMIN)

Resume

A brief desciption is given of previous work, lithologic assemblages, tectonic events,


structures, suture zones, metallogenic environment and geodynamic evolution of the Ecuadorian
Northwestern Cordillera.

Key words : accreted terranes, island arc, back-arc, sutures.

Introduction

Previously, the Ecuadorian Western Cordillera was interpreted as an island arc, accreted
in the Late Cretaceous (Henderson 1979).
An Eocene island arc and back-arc sequence on top of an accreted fragment of Cretaceous
oceanic crust is the interpretation resulting from more recent investigations. Continental type
talc-alkaline arcs, from Upper Oligocene to recent, rest unconformably on top of above units
(Eguez 1986, Lebrat 1985, Lebrat et al. 1987, Raharihaona 1980). The position of the
suture, at the eastern edge of these terranes, was better defined by Aspden et al. (1987) and
Lebrat et al (1986).
Present work describes the situation of the Ecuadorian Western Cordillera between 0 and
the Colombian border.

Llthologic assemblages

Basement of terranes of the Cordillera and the Coastal Plain consist of a serie of lithologic
units, with ages from Lower Cretaceous to Paleocene. Tholeiitic basalts and flysch-type
sediments, ending in a conglomeratic unit, are their main components, An Eocene assemblage,
consisting to the east of back-arc-type sediments and lava% and to the west of island-arc-type
volcanic deposits, with sporadic calcareous intercalations, rests unconformably on top of this
basement. Locally, it is covered, also unconformably, by several continental Oligocene to recent
talc-alkaline volcanic series.
Two main belts of batholiths can be distinguished among the various intrusive phases. One,
of Miocene age, is located near the axis of the Western Cordillera. The second, of Eocene age, is
located toward the west, almost in the coastal plain. It could be the root zone of above Eocene
island arc.
-._.ml ,,- .

204
. __.__--._

205

Tectonic events and structure

The history of this Western Cordillera is dominated by two main tectonic events, the first
of Maestrichtian-Paleocene age, the second of Oligocene age.
Structurally, an intricate, east-verging asymmetric tectonism is evident mainly in the
eastern part of most profiles, without presenting clear evidence of abduction. Sporadic
ophiolitic rock-outcrops, delineating a suture zone near the eastern base of the Western
Cordillera, are the expression of a late Cretaceous-early Tertiary accretion, without exluding
the possibility of earlier accretions towards the East (Aspeden et al. 1987).
Toward the West, gabbroic and ultramafic rocks crop out along another major fault zone,
possibly related to the second tectonic event of Oligocene age.
All of the Cordillera is sliced-up into longitudinal tectonic blocks through transcurrent
faulting.

Metallogenic environment

The metallogenic environment is caracterized by hydrothermal (epigenetic)


mineralization related to later intrusives, superimposed on top of a syngenetic context,
associated with the accreted ophiolitic and island-arc-type units.

Geodynamic evolution

The Ecuadorian Western Cordillera, and the Coastal Plain can be interpreted as an oceanic
fragment, accreted during the late Cretaceous-early Tertiary.
Eocene oblique subduction from the southwest built an island arc and its related
environments on top of the Cretaceous basement.
This evolution was ended by an Oligocene not yet well defined compressive tectonic event,
corresponding possibly to a general re-organization of Pacific subduction regimes.
After a possible westward jump, a more perpendicular subduction produced magmatic
activity typical of an active continental margin.

References

Aspden, J., Litherland, M., Duque, P., Salazar. E., Bermudez, R. y Viteri, F., 1987. Un nuevo
cinturon ofiolitico en la Cordillera Real, Ecuador y su posible significado regional. Politecnica, XII, 2,
81-93, Quito.
Eguez, A., 1988. Evolution Cenozotque de la Cordillere Occidentale septentrionale dEquateur
(00 15 s - 01 10 S ), les mineralisations associees. Th. Dot. UPMC. Paris,llGp.
Hendersen, W.,G.. 1979. Crelaceous to Eocene volcanic arc activity in the Andes of northern
Ecuador. J. Geol. Sot. London. 136, 367-378.
Lebrat, M., 1985. Caracterisation geochimique du volcanisme ante-orogenique de IOccident
Bquatorien: implications geodynamiques. Documents et Travaux du Centre Geologique et Geophysique de
Montpellier, 6, 118~.
Lebrat, M;, Megard, F. et Dupuy, C., 1986 . Pre-erogenic assemblages and position of the suture
between oceanic terranes and the South AmericanContinent in Ecuador. Zbl. Geol. Palaont., 1, (g-19),
1207-1214.
Lebrat, M., MBgard,F.. Dupuy, C. and Dostal, J.. 1987. Geochemistry and tectonic setting of
pre-collision Cretaceous and Paleogene volcanic rocks of Ecuador. GeolSoc. Am. Bull., 99,569-578.
Raharihaona, L., 1980. Petrographic des roches volcaniques et ante-orogeniques des Andes
Bquatoriennes: contribution a letude de leurs parageneses m&amorphiques.Th. Dot. Univ. Louis
Pasteur, Strasbourg, 166p..
_ ., _,_. .-.

.
207

METAMORPHISM OF THE CEUCA FORMATION, WI ECUADOR :


GEOTECTONIC lMPUCATlONS.

Luk Aquirro

Lab~rat~ire de P&ologle Magmatique, U.R.A 1277 CNRS ,FaoulU drr ScirnWr do St.Jo&ne. Univorxit6dAlx Marxellle III.
13397 MARSEILLE.Cedrx 13, FRANCE.

l_e8rWh8S UIdkitiqU88 d8 I8 Formation Celia (SW EquPtWr) Ont 6tb 8ffWt688 par un m6tarnorphleme d8 fSOl&
26Olit88t prOhnlte-pump8llyit8.Un gradient thermiqur mod&r6 1 fort, une faiblr pr8rriOn lithOetetiqU8,unr brute fo2 8t un
rapport aau da m8r/rWh8 piUt8t feiblr, car8ct6riwnt ce m&ernorphismr. Cecilr8SWmbl8 au m(tamOrphi8me da planohrr
od~lqU8 du Group8 MPbuiqU8 d8 18 ColOmbi8 8t 8ppul8 an WnS6qu8nC4 Ihypoth~sa dun baasin marginal (SVOrt6)P
lorlginr du volcanism8 Cellce.

Kay words: Mu8dOr, ~8taceour, Cellca, lowgrada metamorphiem,marginslbasin, l&and arc.

lntroductton

kl f&dOr (FiQ.1)VOk%%niC
WqU8nW8 Of M8WZOiCend ~nOzOiC 8Q8 BT88xpOWd dOnQ th8 W8St8rnbordrr Of the
COUntry.&llOnQ th8m, thr88 VOlCaniOfOrm8tiOn8Of 8eWntidly Cr8taceOUCQ8 er8 di8tinQUirh8d:(8) th8 marin Pifi6n
Form&On WithOW.SnlO Oh8mlcrl lffinitie8;(b) th8 merino Mecuchl Formationwhichr8pr8WnU en OW8nlCMend arc Wit8 Of
prrdOmlnMt tholrlitioWmporitlon , end (c) thr Celia Formation, moatiy marina, of c&-alkaline natura. Thir Formation Is
WnSid8Wd either aa 8 volcanloarc devaloped on en active wntlnantal mergln or u en aborted marginal baeln with extreme
WntinOntelCrust8ttOnuPtiOn.TheW throw08tWaOUS unite h8v8 been 8ffWtrd by lOwgr8d8 mrtamorphlem.
Thr Pit% end MaouchlWit88 are wnsidrred es ellochtonours./. and ara 6epemt8d from tha lnsialic,8UtOChtOnOU8,
clllca volcanl~e by 8 suture IinO which larQ8ly WlnCid88 with th8 pOritiOnOf the ~lOrW-GU8y8qUil Megashrar (Dr&f).
(f&ferencar concerning thir introductionto be found in Faininger &&iatow 1980, Bald& 1985, L@br8t lW5 , Aguirre h
Athrrton 1987 among othrrr)
Field rrwnnakanoe and sampling in the region8 of Zaruma, Portobelo, Pi688 and Cellca and rubwquont optioal and
Ch8miOellrrbor8tO~work were WJri8d Out. Thl8 paprr lnt8ndSt0 WtabliSh8 metamorphic mod81for th8 Cellu FOrm8tiOn 88

8 clue to th8 undrretanding Of th8 n8tUr8 Of ltSQbOtWtOniC


MfflnQ.
._.._.._-._ . .

208

Potrogrephy

The Celka Formation oonsiata of


volunk rooka with minor intercalated aedi-
m*nta. -rite fkwa pr-teaaaock-
ted wlUlfkwbreoduandlithktuff.of

wbordinate end oorreapond to baa&, da-


dte end oryaW tuft. The andeaitea are por-

d*r W,, - Ab ,,bpdOf,.dl


- chw-
roxrnr (OaJ&,$o,,) and homblendr. In
th* beaelta, th* phenouyata (2 mm) oorro*
pond to Pl4Jkol-e (Anac.aa&&t$r,) end
olinopyroxena (mean of Ca,Mg,Fe,,)
whrrru in the daoftea they conaiatof JbiU-
red pf@daae, hornbknde and lmbeyed
quartz. All these rocka ara modrrrtrly, al-
though lxtenafvely, altered. Seoondary
(=meWnorphk) mlnerela ue abundant in
winletaand amygdaloa and u patchy repla-
cement In primary oryaW and ln the
groundmaaa.

Gwchomlatry

me ohrmlatry of thr tilia andad-

Fig. 1 : Simplifiedgeologic map of weatern Ecuador. 1 = Maouchi tea haa been atudird by Lebrat(l985) who
Formation ; 2 = Cdiu Formatron ; 3 = PiMn Formation ; 4 - wndudrd thatthrr rocks are calc-alkalinr
undifferentiated Cretaoeoua aadlrnonta. DGM = Cokrea - Guxyyr-
quiimegaahrar. S-S = auturobetwem the wnUnrnW end oceanic and can be daaaiCed u medium-K andeai-
terrain8 (from &ulrrO 6 Atherton 1087 ; modified lfler Lebrat 1285). toa. Thalr REE pattern la lnrkhed In LREE
fnaet8hOW8 bcdon Of the crotaOooua vokJnio belt In th* Northorn end fallwlthlnthe rengr of conUnenW IJmci
Andes.
arc andealtea. Thrir Th/Ta ratio, nlatlvaly
high, la oloae to that of recent Andren ande-
sites. According to Lebrat (1985) and Lebrat eta/. (1987) thou lava8 are intermediate between Wand UC calc-alkdlne and
typical continental mergln andaaitoa end roaemble the magmaa rmplaced on l rather thin conUnonW ~uat.

Metamorphk mlnorakgy

Two varieties of llblte are present: (1) Ab I (An,An,.J


which reaultafrom aimpk aibitizationof the primary, Ca-tich,
plagiooleaaend ,(2) Ab II (An,& a late phaaa present aa a moaaioIn winlrta or aa neatly defined patcher in fresh and albltizrd
plaglookae phmocrysta.
t.eumcnUte end lUlblte eppeu aa patchy replacrmrnt in phenccryata and fillingveinkta. Ccexlatence of both reolitea wxa
observed In only one case.
Prehnite la moderatelyabundant aa patchy replawment of pla~koleae and In arnygdelea. HaIron wntent lxpreaaed aaXFeJ+
(= 100Fes+ / (Fe+ + AI,& ) varlea from 9.310 18.7. Prehnitr ooexkta with pumpellyfteand. additIonally (in one case), with
gunet of the andraditegroaaulu aeriea.
Pumpellylte la fairly abundant aa replawmmt patcher In plaglcclaae phenorxyata. 28 FeoOa content la almoat Invulably
higher than 10% which la typical of pumpellyiteaIn the zeolitaand prehnlte- pumpellyitefades (Uou 1983). The Wea+ wluea
r-- . . ----,

209

for ooewisting pumpellyits and rpidote show that Fes+ partitioning was squitabls: 21.7 to 34.7 for the former mineral and 21.4
ta 31.2for the latter. ln pumpsllyitss from spldots-free assemblagssKFes+ ranges from 23.6to 37.4. I\r a whole, pumpellyltrs
of the Celka rocks us richer In iron than those of the Macuohl Formation.
Epldote Is scarco and appears as replacement of plagloolase and clinopyroxme phrnocrysts In assemblages contalnlng
pumpellyhe, chlorlte, Utanlts, calcite and prehnlte.
Garnet of frarnboldal habit of ths sndraditegrossular series Is found lnrlde chlorite patches In the groundmass of a basaltic
flow where h oorxists with Iron-rich prehnltr, Iron-rich pumprllyite. IaumonUts, calcite and chlorite k closely rrsemblss In
oomposltion garnets reported by Coombs era/. (lS77) from basio volcanic rocks of Southern New Zealand metamorphosed
under prohnlte-pumpellyltr faoies.
Chlortte of dlabanUUo oomposition Is abundant as replacement of plagioclsss and clinopyroxenr, In amygdalra and as
lntersUUsl material In the groundmass. An Interlayering of chlorite-smectite Is suggested by the composition of some brownish
phyUostlloates dcher In glOs than the dlabanthes.
Tltanlte Is fairly oommon as small gralns insldo plagioclase and cllnopyroxene crystals, in amygdalea and as spots In the
groundmass. its NsOs and FeO* oontents are hlgh (K= 393% and 3.02% respectively); the main substitution wrresponds to
(Al+ Fe*) <->Tl. These features characterbe titanite In the very low grades of metamorphism (Mrs 6 Coombs lS77)

Metamorphlo pattern

The mineral assemblages and the chemistry of lndlvidual mineral phases Indicate that the Cellca volcanics have been
affected by lowgrade, nondeformational, subgreenschist mstsmorphlsm. The assemplagaa: (1) laumontite+ stilbite
(prehnite and pumpellyite absents); (2) laumontite + grossular-andradits garnet+ prehnite + pumpellyite; (3) prehnite +
pumpellyite t Al-rich that-rite + (smectites); and (4) epldote t At-rich Utanite + pumpellyits (prehnite absent) are diagnostic
of the zeolits and prehnlte-pumpellylte faclea.
The chemical characteristics of the mineral phases previously described (e.g. KFes+ values of prehnite, pumpellyite and
lpldote; Al and Fez+ contents In titanits; composition of chlorite among others) suggest a metamorphism at a moderate to
steep thermal gradient, weak load pressure, high 10s and rather small ssswster/rwk ratio. This pattern approaches that
observedinocean-flwrmetamorphlsmand Iscloaertotheonedescrlbedforthe DiabssicGroupof Colombla(Aguirreinpress)
and the Peruvian Casma Group (Offler et a/. 1930) than to the model known for the Ecuadorian Macuchl Formation (Agulrre
& Athsrton lQ97).

Ths Celica Formatlon Is considered as a preorogenlc, caloalkaline, arc volcanism developed on the static western
margin of South America between the Alblan and the Maastrichtian (Kennerley 1939, Feininger 6 Matow 1999, Lebrat 1935
and references therein). An alternative setting has been suggested by Aguirre (L Atherton (lQ97) as part of a wider model for
the Cretaceous evolution of the western border of South America. An aborted marginal basin nature , with strong attenuation
of the wntinental crust, Is attributed to the Celica Formation according to such model, The general analogies in metamorphic
pattern with the Colombian Diabasic Group and the Peruvian Casma Group established here would favour this last hypothesis.
However, some important differences with the &ma pattern are also observed in the Cellca volcanica: (1) absence of
greenschist facies assemblages; (2) absence of the high-T zeolite wakakite; (3) lower geothermal gradient; (4) higher values
of fOs and Fe3*-activity. These differences are here interpreted in terms of the degree of evolution of a marginal basin. Thus,
the steepest thermal gradients, with their accompanying metamorphic pattern, should be found in aborted marginal basins
with strong attenuation of the continental crust. In these basins (e.g. the Carma Group) the metamorphism is transitional
between burial and ocean-floortype (Aguirre efel.1939). Onthe other hand, In those basins wheremaximum degree of crustal
thinning has been achieved (e .g. the Mica case in Aguirre & Atherton 19Q7)or In those where the continental crust has been
totally eliminated (e.g. the Colombian Diabaslc Group), a decrease of the thermal gradient occurs and the metamorphism is
of ocean-floor type.
. .,_.,_.
_. . .^ _.

210

AQuirre, L 1989. Metamorfiamo prrorog6nico oretioico y marco geotect6nlw. Cordillera Mdental de Colombie @erfil
Buga-Buenaventura). Rev/& Geofdg~a de We, In preu.

Aguirre, L 6 Atherton, MP. 1987. Lowgradr metamorphism and goolectonio setting of the Meouohi Formation, Western
Cordillera of Ecuador. huma/ ofA4efamorphic Geofogy, 5,473-494.

Aguirre, L, Levi, 8; 6. Nyatrtim, J.O. 1988. The link between metamorphism, volcanism and geoteotonio utting during the
ewMfon of the Andes. In: Zvolutfon of Metamorphlo Belti (edr. Wy.J.S., Cliff, RA a Yard1ey.B.W.D. ). Geological
Soobfy ofLondon, SpeciaI Pub&at&n NW. 223.232.

Baldook,J.W. 1985. The Northern Andre: l rovlewof the Ecuadorian Pacific margin. In: -The Owan Basins and Marginem (ode.
Nakn, AEM., Stehll,F.G. & Uyeda, S.). 74 181-217.

Boles,. JR 8 Coombr, D.S. 1077. Zeolite facier alteration of aandrtoner in the Southland Syncline, New Zealand. Amerfcan
Journal of Science, 277,982-1012.

Coombe,D.S., Kawachl, Y., Houghton,B.F.. Hyden,G., Pring1eJ.J. &Williams, J.G. .lQ77.


Andraditr end andraditegrouular solid aolutionr In very-low grede regionally metamorphosed rocks In Southern New
Zealand. Confribufionr fo Mineralogy and Petmkgy, 6~3~229246.

Fllnlnger, T. (L Bestow, CR 1980. Cretaceour and Paleogenr geologic hIstoryof coastal Ecuador. Geokzglsc~ Rundechau,
60.849874.

Kennrrlry, J.B. 1980. Outline of the geology of Ecuador. Overseas Geobgy and Minerel Resoumer. Institute of Geological
Sciences, Great Britain, 55, 17 p.

Lebrat. M. 1985. Caract6rlsation ghochimique du volcanismr antrorog6nlque de Ioozident (quatorlen: Implications


g6odynamiques. Documents et Tmaux du Centre G6ologique et GBophyslgue. CGG (CNRS), Montpellier, 118 p,

Lebrat, M., MBgard, F., Dupuy, C. 8 Dootai, J. lge7. Geochemistry and tectonlo setting of pro-collision Cretacoour and
Paleogrne volcanic rocks of Ecuador. Geological Sociefy ofAmerica Bulletin , ~,568578.

Lieu, J.G. 1983. Cccurrenoer, compositions and stabilities of some Ca-AI hydrous silicates in lowgrade metamorphic rocks.
Memoir Geolcgic~ Sccbsty ofChina,5 ,4766.

Offler, R, Irgulrre, L ,Levl, B. & Child, S. 1980. Burial metamorphism in rocks of the Western Andes, Peru. Lirhos, 13,31-42.
211

NEOGENE STRESS PATTERN IN SOUTHERN ECUADOR

LAVENU A.+, NOBLET Ch.++ and WINTER Th.+++

* ORSTOM, Ap. Post. 6596 CCI, Quito, Ecuador.


213 rue La Fayette, 75480 Paris cedex 10, France.

** Botquelen, 56610 Arradon, France.

*++ Laboratoire de Tectonique, MBcanique de la Lithosphere


IPG Paris, 4 place Jussieu, 75252 Paris cedex 05,
France.

CIPGH-EFN-CLIRSEN-ORSTOM Convention1

Resume

Results of microtectonic, structural and sedimentologic


analysis of the deposition of Mio-Pliocene continental
basins permit to describe the Neogene stress pattern in
Southern Ecuador. These three kinds of data show that this
region was submitted to a continuous compressive stress
pattern, from Upper Oligocene to Pliocene time.

Introduction

Intracontinental basin deposition is a characteristic


of Andean geodynamics. Basins formed after the Upper Creta-
ceous emersion of the Andes. These elongated basins are
located in or along the High Andes. In the Ecuadorian
Andes, a great continental sedimentation occured during
Late Cenozoic time.
Tectonic and volcanic events which occured during and
after sedimentation are recorded by the depoiits.
Sedimentologic and tectonic analysis is one way to retrieve
erogenic mechanisms. In this paper we compare results
obtained in a microtectonic analysis of Mio-Pliocene
continental basins of Southern Ecuador with sedimentologic
and structural data.
212

General sedimentary evolution of tertiary basins of


Southern Ecuador

The tertiary deposition of the intramontane basins of


Cuenca, Nabon, Loja, Malacatos and Zumba crop out between
the Eastern and Western Cordilleras, along N-S and NNE-SSW
trending major faults, at elevations ranging from 1600 to
3000 m.
The deposition of these basins (1500 to 4500 m thick)
consists of continental deposits with several intercala-
tions of volcanic deposits. It is always organized in two
megasequences. The first one (fining- and thinning-upward),
which shows an evolution from proximal to distal facies
(alluvial fan, braided river, lacustrine facies), characte-
rizes the opening of the basins. The second one
(coarsening- and thickening-upward), which presents an
inverse evolution, characterizes their closing. This last
megasequence begins with megaturbiditic, deltaic and/or
distal alluvial deposits and follows with a prograding
alluvial sedimentation in the whole of the basin area.
In Cuenca basin, a maximum basin life of about 20 My
may be estimed. Assuming a ratio of sediment decompaction
of about 2.0 for the deep lacustrine sediments and about
1.5 for the rest of the basin deposits, a thickr.ess of
about 6000 m may be estimated and a sedimentation rate of
300 m/My may be deduced.

Microtectonic analysis

Microtectonic studies permit us to characterize the


state of stress in this region using numerical methods to
compute it. Most of microtectonic station studies revealed
faults bearing several generations of slikensides (2 or 3)
and a relative chronoloqy was observed.
CI N-S trending compression:
-. Five sites, measured in
the Saraguro Formation CU. Oligocene) and at the base of
the Biblian Formation (L. Miocene) in the Cuenca and Nabon
basins reveal a roughly N-S to N140E trending compression.
A NE-SW trendinq compression: Six sites, measured in
the same formations in the Cuenca, Nabon and Malacatos
basins, permit to characterire a roughl,y NE-SW trending
compression. But the maximum principal stress varies in
trend from N31E to N59E.
213

A E-W trendinq compression: Ten sites, measured in the


same formations (Cuenca and Nabon basins) and in Middle
Miocene deposits (Loja, Malacatos and Catamayo basins)
reveal compression striking between N81E and N107E.
A NW-SE trending compression: In Cuenca, Nabon and
Catamayo basins microtectonic results clearly show a N120E
to N135E trending compression.

Comparison between microtectonic, synsedimentary folds and


sedimentologic data

TWO major fault systems trending N30E and N-S control-


led the Tertiary basins after the deposition of the
Saraguro Formation. The deposition and filling of basins in
South Ecuador is affected by synsedimentary tectonic
structures on kilometric scale. Systematic detailed studies
of these structures were made in all the basins deposits
but especially in the Cuenca basins deposits.
The fining- and thinning-upward of the first mega-
sequence characterizes an increase in the tectonic
subsidence (Lower Miocene opening basin). This opening is
marked by a sedimentary wedge, which affected all the first
megasequence. This synsedimentary progressive tilting is
suggestive of a normal component of motion along the NE-SW
trending boundary fault, induced by Lower Miocene trending
compression.
The coarsening- and thickening-upward of the second
megasequence characterizes the progressive closing of the
basin (Middle Miocene and Pliocene time). The E-W Middle
Miocene compressive event initiated the closing of the
basin inducing right-lateral motions along N30E trending
fault system and probably reverse motions along the N-S
trending fault system.
The E-W and the NW-SE trending compressions caused the
formation of progressive erosive unconformities, suggestive
of synsedimentary folding with reverse beds.
The analysis of the progressive unconformities
suggests, more, that variations in tectonic intensity
occured during the sedimentation.
Thus, microtectonic and synsedimentary fold data
attest to a strike-slip component of the movement along the
boundary faults from Lower Miocene to Middle Miocene, and
mainly reverse kinematics from Middle Miocene to Pliocene.
214

The shortening axis deduced from microtectonic analysis


must furthermore have experienced an indirect rotation from
NNE during the Lower Miocene, to SE during the Pliocene,
and are consistent with synsedimentary folds and sedimento-
logic data.

Discussion and conclusions

Microtectonic studies indicate several successive


tectonic events. These events may be ascribed either to one
or several "instantaneous" deviatoric stress tensors, which
could correspond to deformational climaxes (1 to 2 My) or
to a mean deviatoric stress tensor derived from slikensides
produced "continuously" during the lifetime of the basins.
The convergence of results derived from microtectonic data
is more agreement with the first solution. Thus, we infer
that the states of stress derived from orientations of
slikensides reflect compressional deformation climaxes
similar to those which are at the origin of both the syn-
sedimentary folding and the progradation of proximal facies
onto more distal facies.
Sequential analysis of basins deposits clearly shows
tectonic deepening, then shallowing of the basins
Synsedimentary deformation which affected all deposits
in the basins attests for continuous compressive tectonics
during the length of deposition.
So, these three kinds of data (sedimentologic,
structural and microtectonic) indicate one conclusion:
southern ecuadorian continental basins were under
continuous compressive stress from Upper Oligocene. to
Pliocene time. Intensity of the deformation varies in time
and space, and attests to deformational climaxes within a
deformational compressive continuum.
In the Ecuadorian margin, this Tertiary tectonic
continuum appears ta be in agreement with the steady-state
dynamic behaviour of subduction, which controlled Tertiary
Andean orogeny.
___. I _-_.w,

215

CENOZOIC THRUSTING AND WRENCHING IN THE CORDILLERA


ORIENTAL, COLOMBIA: FIELD DATA AND EXPERIMENTAL
INSIGHTS.

Peter Cobbokr, Pascal Rii. Carfos Ulloa.

&Mre Armoriesin dEtude Sfructurale de8 Sodes, Universit4 de Rennes ,25042 RENNES C&q France

INGEOMINAS, Diagonal 53 No. 24-53. Bogota, Cdombia.

The high plateau (about 2OW m) of the Cordillen Oriental fs mainly due to post-Miocene crustal thickening, but
~ts~a~llem k far from betng a srmple knear fold-and-thrust belt. Instead it has a V-shape on a map, the apex pointing

In the area of the apex (Boya&) the cordillera is bounded by active faults in the foothills to the east (Llanor
basin) and west (Magdafena valley). The en&e Corditlera is here a large pop-up. Within the Cordillera are fofds and thrus$
of Cenozoic age. Prominent is the flat-lying Soapaga thrust, which has carried basement rocks (Precambrian and
Paleozoic metamorphic rocks of the Floresta massif) eastwards over Tertiary sediments, the transport direction being
about 10fP. Another prominent reverse fault. possibly of higher angle, is the Boyaca fault, which cuts out the inverted
eastern llank of the Arcabuco anticlfne. Large antidines such as this one have cores of Jurassic rocks: these are inverted
Jurassic rifts.
To the north (Bantandsr and Notts de Santander), the Cordillera Oriental strikes NNW. It is sharply bounded to the
west bv the Bucaramanaa fault zone. This fs an obliiue-slio zone. with components of westwards overtfuustfna tover 4 km
vet& throw) and left-l&af wrenching (perhaps 2oi) km): lndiikiual faults in tftii zone are sometimes nearly itire thrusts,
sometimes nearly pure strike-slip faults, more often oblique-slip faults, aft with nearly parallel strikes. The geometty Is that
of half a flower rikucturTj0 There
Tr 22 are also antithetic faults with right-lateral offsets: these often bound blocks that have
rotated clockwise by a domino mechanism. Finally, there are flat-lying thrust faults with northerly strikes. Prominent
amongst these is the Falla ds las Mercedes, which dips ISo or so westward and has put Precambrian basement on top of
Tertiary sediments.
The southern Cordillera Oriental ftas a northnortheasterly strike. It is bounded by active thrusts and strike-slip
faults, right-lateral in this area. Strike-slip motions appear to dominate the active tectonics, but thrusting was more
fox-t in the Tertfary.
Thus the entire Cordillem Ortiental appears to have been generated at the restraining intersection between
conjugate wrench rystems, left-lateraf in the north, right-lateral in the south. This may account for the high plateau, uplifted
some 3500 m since the Mine. It may also account for a component of shortening along strike (north-south) in the apex
(Boy&), where the Mesozok snd Tertiary cover has been folded into tfght domes and basins.
Near the Caribbean coast,. tie Sirm Nevada de Santa Marta and the Serranh ds Perfjc span another, more
acute, restraining intersection. thi bme between the Santa Mana-Bucaramanga fault system in the southwest and the
tight-lateral Oca system mat pamlfefs the continental margin to the north. In thii area, wrenching has predominated over
crustal thickening. The general direction of contraction ls northwest. Thus on a continental scale, the Cordillera Oriental
marks a transition between crustal thickening with east-west contraction fn the south and right-lateral wrenching in the
north.
We have done two reties of experiments with analogue models scaled for gravity. The first set investigated
coeval wrenching and thrusting upon parallel faults, either at the scale of sedimentary basins (flower structures), or at the
scale of mountain belts. The experiments show that partitioning of fault motions, between almost pure thrusting and almost
oure strike-slio faultino. is oossible under certain conditions. esoedaliv in the oresenm of detachment horizons at depth.
These experiments w&elf&e explain the structures &e&d ki the BLaram~ga fault zone.
The second set of experiments InvestigaIed lateral transitions between crustal thickening and strike-slip faulting
at continental scale. The transitions were obtained by applying suitable lateral boundary conditions upon the model
continents. The usa of a rigid indenter provided transitions between pure crustal thickening in front 01 the indenter and pure
wrenching at the side. On this basis we attribute me structural style of the northern Andes to a transition between
convergent plata mot&s In the south (Naxca and South America) and transwrrent motions in the north (Caribbean and
South Amerfcan plates).
217

UPLIFT AGE OF THE GARZON MASSIF (EASTERN CORDILLERA,


!&COLOMBIA) IN RELATION TO THE INFILL OF THE ADJACENT S.NEIVA
BASIN

Wageningen Agricultural University, Dept. of Soil Science and Geology, Duivendaal 10, 6700 AA
Wageningen, The Netherlands

The present investigation was carried out in order to determine the age of the uplift of the Garx6n
Massif, the most southern extension of the Eastern Cordillera of the Colombian Andes, and its effects
on the sedimentation within the adjacent S.Neiva basin. To that purpose fission track age
determinations were done on apatite from samples taken at different elevations within the massif and the
~tratigraphic and sedimentary relations of the deposits filling the adjacent SHeiva Basin were studied,
while K-k detaminationson thesedeposits provided the necessary time control.

The S.Neiva Basin constitutes a broad tectonic depression situated between the Central and Eastern
Cordilleras in the south of Colombia. It is filled with some 3500 m of fluvial material deposited during
uplift of the Central Cordillera and f900 m of volcaniclastic and volcanic sediments derived from the
Central Cordillera volcanic arc. The fluvial sediments belong to the Gualanday and Honda formations;
the vokaniclastic and volcanic deposits are grouped into the CliganteFormation. This latter formation is
subdivided into three members: a lower and upper conglomeratic member and a middle volcaniclastic
member.

K-Ar determinations were carried out on biotite and hornblende separates from samples taken at
stratigraphically controlled positions. The middle volcaniclastic member of the Gigante Formation was
dated at 8.3-7.0 Ma and an age of 6.66.2 Ma was obtained for the upper conglomeratic member. The
lower part of the Honda Formation was dated at 16.1-14.6 Ma. The age of the lower conglomeratic
member of the Gigante Formation was estimated at 11-8.3 Ma.

From the effective track retention temperature of apatite and the present elevation of the highest samples
it is calculated that the Garz6n Massif was uplifted approximately 6.5 km. Apparent fission-track ages
of apatites date the uplift at some 12 Ma ago. This implies that the Gigante Formation was deposited
after the uplift S.S.
Paleocurrent directions from the Honda Formation and lower two members of the Gigante Formation
are to the east.,indicating that the uplifted massif at first had little influence on the drainage pattern of
the basin. Only 5 million years after the uplift, during deposition of the upper conglomeratic member,
the drainage system in the basin changed direction and &watering was to the North.
In the northeastern part of the studied area, the Gigante Formation shows interfingering with
conglomerates with a provenance east of the present basin while to the NW only the upper
conglomeratic member interfurgers with these conglomerates. Apparently, erosion products from the
uplifted massif spread further west in time, reflecting the increasing activity of erosion and denudation

Evzi compared to the vast amounts of fluvial sediments produced during uplift of the Central
Cordillera, the thickness of the deposits resulting from uplift of the Garx6n Massif is minimal. This
fact and the fact that the uplift took place some 12 Ma ago, but that the rivercourses were influenced
only 7 million years ago, lead to the conclusion that the uplift of the massif hardly influenced
sedimentation processes in the adjacent basin.

There are two possible mechanisms which may have prevented deposition of the major part of the
erosion products into the SHeiva Basin:
1. An alignment of intramontane basins can be found within the Garx6n Massif parallel to its western
border along one of the major faults. Gec&ctrical and sedimentological investigations indicate that one
of these basins, the Pitalito Basin, is at least 1200 m deep and is probably filled with Pliocene to
218

Pleistocene fluvial deposits. However, the electcicplconductivity studies do not exclude a much deeper
basin that than could be filed, at least the deeper part of it, with erosional products derived from
Jurassic intrusives borderhg the basin. Therefore,it is conceivable that the inrramontanebasins were
formedduring the uplift and acted as a sediment hap, preventingthe erosion productsfrom entering the
S.Neivo Basin.
2. In the PutumayoBasin dhctly to the ePft of the Gti Massif, some 700-800 m of UpperTertiary
fluvial sediments are found. Possibly the msssif was tilted to the east during uplift and the sediments
were depositedpreferentiallyin the PutumayoBasin.
219

CRONOLOCIA DE LAS ACREClONES DE TERRENOS ALOCTONOS


EN LOS ANDES COLOMBIANOS

Jean Francois Toussaint* y Jorge Julian Restrepo*

* Universidad Nacicnal de Colombia, Facultad ds Ciencias, Apartado AOreo 3840, lbdellfn, Co1

Resumen

La esquina noroccidental de Suramerica estg formada por un mossico de varios


terrenos acrecionados al Escudo Guyan& en diversos tiempos. Los terrenos
orientales poseen un basamento continental y heron suturados durante el
Paleozoico tardlo y el Cretgcico. Los Terrenos occidentales tienen basamento
ocehico y fueron suturados durante el Crethico y el Miocene medio.

Palabras claves: Aloctonla, Terrenos, Andes, Colombia.

Resume

La region nordoccidentale de IAmerique du Sud est formde dune mos$que de


plusieurs Terrains qui ont fait collision avec le Boucller Guyanals durant
diverses epoques. Les Terrains orientaux possedent un socle continental et ont
ete soud& pendant le Paleozdique superieur et le CretacC. Les Terrains
occidentaux ont un socle oceanique et ont fait collision pendant le Cretac6
et le Miocene moyen.

Mots clefs: Alloctonie, Terrains, Andes, Colombie.

Introduccidn

Nuevas investigaciones permiten suponer que el basamento de 10s Andes Colom-


bianos no es autdctono sino que esta formado & bloques aldctonos suturados
entre sl. En esta nota se presentara una breve descripcih de 10s principales
terrenos aldctonos, as1 coma un esbozo de la cronologfa de las acreciones.

El Bloque Autdctono y los Terrenos Al&tonos

El territorio colombiano est&i formado de oriente a occidente por:

- Un Bloque Aut6ctono (BA), unido al Escudo de Guyana al menos desde finales


del Precambrico y que corresponde a las regiones de 10s Llanos Orientales,
del Caquet6 y Amazonas. Su corteza es continental, con un espesor
estimado de unos 35 km. Las edades radiometricas del basamento
220

precambrico sugieren un importante evento tectometamdrfico transamazdnico


seguido por un rejuvenecimiento nickeriense. El Paleozoico inferior esta
representado por sedimentos cambro-ordovicianos, mientras que el Paleozoico
superior esta totalmente ausente.

- El Terreno sospechoso Chibcha (Ch), con basamento continental precambrico,


comprende la Cordillera Oriental, el Macizo de Santander, el flanco este
de la Cordlllera Central y la parte sureste de la Sierra Nevada de Santa
Marta. Su Gltimo evento tectometamdrfico es caledoniano. Rotas sedimen-
tarias marinas del Paleozoico superior recubren en discordancia las unidades
metam6rficas paleozoicas inferiores. Desputs de unirse al bloque autdctono,
un importante cinturdn magmatico afectd su borde occidental durante el
Jurasico 3 Tw
221

Durante el Cretgcico el Terreno Calima se amalgama al Terreno Tahaml


pero este conjunto no estaba suturado todavla al Ch. Importantes eventos
tectometamdrficos con metamorfismo de alta y media presidn y con
tectonismo marcado por apilamiento de neppes se producen durante la
amalgamacidn

Los Terrenos Calima y Tahaml amalgamados durante el Cretkico se


acrecionaron al conjunto formado por el BA y el Ch al finalizar el Cretkico
0 prlncipiar el Cenozolco. La diferencia entre ambos conjuntos antes de
la unidn es principalmente Clara en cuanto al tectonismo, el cual es
distensional al Este y compresional al Oeste de la frontera representada
por la falla, probablemente de rumbo, de OtB-Pericos.

Durante el Miocene el terreno Cuna se acreclond al Bloque Andino formado


por el mosaic0 de 10s Terrenos anteriormente unidos. La sutura de Dabeiba-
Pueblo Rico corresponde a una serie de escamas y cabalgamiento hacia
el oriente del Cu sobre el Ca.

Durante esta Ciltima colisibn, 10s Andes Colombianos sufren importantes


acortamientos marcados en particular por la formacidn del megahorst de la
Cordillera Oriental con su borde E cabalgando hacia el oriente sobre el
BA y su borde W cabalgando hacia el occidente sobre el valle del
Magdalena. As1 la morfotectdnica actual de 10s Andes Colombianos,
esquematlzada en el torte profundo localizado a 6N (Fig. I), parece ser
en gran parte una consecuencia de la colisidn del Terreno Cuna con el
Bloque Andino.

Conclusiones

A diferencia de 10s trabajos que representan la evolucidn del sector


septentrional de 10s Andes segGn una simple acreci6n de un dominio oceanic0
al dominio continental suramericano, este trabajo sugiere que la construccidn
del territorio colombiano se produjo por varias acreciones sucesivas de terrenos
al6ctonos con basamento tanto continental coma ocefinico. La Gltima acrecidn
de un bloque continental serla finicretacica y la Gltima de un bloque oceGnico
serla miocena. Estas conclusiones corroboraron que la parte septentrional
de 10s Andes es et resultado de procesos geodin6micos muy diferentes a 10s
que actuaron en 10s Andes Centrales.

Este trabajo es wa contribucih al Proyecto 279 Terremr en Amkica Latina de1 PICC.

Referencias

Etayo-Serna, F. et al. 1986. Mapa de Terrenos geoldgicos de Colombia.


Publ. Esp. Ingeominas, Bogota 14: l-235.
Restrepo, J.J. y J.F. Toussaint. 1988. Terranes and Continental Accretions
in the Colombian Andes. Episodes ll(3): 189-193.
Toussaint, J.F. y J.J. Restrepo. 1989. Acreciones -sucesivas
-._ en Colombia:
un nuevo modelo de evolucidn geolbgica. V Congr. COI. Geol., Bucaramanga.
Memorias 1: 127-146.
Toussaint, J.F., J.J. Restrepo, H. GonzBlez y A. BermGdez. 1989. Transecta
6N (Andes Colombianos). Reunidn sobre Transectas de America del Sur.
Mar del Plata, Arg.: 1-4.
-*-. I m-s.-

223

BASSINS
225

ASPECTS OF MESOZOIC ANDEAN BASIN DEVELOPMENT IN SOUTHERN


SOUTH AMERICA.

ROSENFELD, Ulrich l

l Geobgisch-Palaontologtsches lnstitut der Universitat,


Correnstrasse 24, D-4400 Mgnster , FRO.

The Paleozoic orogenies at the Pacific margin of southern South America have come to
an end by the Gondwanlde orogeny (P-Tr). In Triassic times the region is caracterized by
various intramontane basins. Transgressive marine Upper Triassic is only found in some
regions.

The Lower to early Upper Jurassic sedimentation is transgressive. The South Peruvian
Basin, the Andean Basin and the Patagonian Liassic Basin (replaced by the Austral Basin in the
Oxfordian) contain characteristic sedimentary sequences which give hints to the basins
devebpement.

All the basins were separated from the Pacific Ocean by island arcs in changing extent.

The Jurassic history is partly terminated by the Araucanlan diastrophism (Middle


Kimmeridgian). It produces only few tectonic and sedimentary instabilities in the South
Peruvian Basin, but a regional unconformity followed by a transgression in the Neuquen Basin,
and the shifting of sedimentation in eastern Extraandean regions in the Austral Basin.

In the Cretaceous, the mentioned basins show different developments. Several tectonic
events, in part regionally restricted, cause further transgressions and regressions and the
arise or disappearance of depocenters. In the Middle Cretaceous, the subsidence of the Atlantic
basins of southern South America increased, whilst the Pacific ones increasingly filled up in
the course of the further Cretaceous.

Obviously both the break-up of Gondwana and the beginning of the Andean tectonic
megacycle affected the Pacific margin of the southern South America from north to south in
different ways which may depend on the respective specific plate tectonic position.
.._.,..%
__---
227

THE ORDOVlClAN PUNA BASIN OF NW ARGENTINA AND N CHILE: FBOM BACK-ABC TO


FORELAND BASIN

Heinrich Bahlburg

lnstitut fur Geologie und Pallontologie, Technische Universitgt, Ernst-Reuter-Platz 1, 1000


Berlin 10, Fed. Rep. Germany

Resumen

Perfiles sedimentologicos, con fauna graptolitica, levantadas en las series turbiditicas


ordovicicas de la Puna septentrional (noroeste de Argentina y norte de Chile) permiten una
division estratigrafica m&s detallada de la serie sedimentaria. La analisis de cuenca llevo a
una nueva interpretation geodinamica de la cuenca desde un tipo tras-arco hacia un tipo
ante-pais.

keywords: Puna, Argentina, Ordovician, sedimentology, stratigraphy, foreland basin.

Introduction

In the southern Central Andes of NW Argentina and N Chile Ordovician elastic rocks occur
in shelf facies in the Cordillera Oriental and as basinal turbidites in the westwardly adjacent
Puna. Measurement of detailed sections in the northern Puna (north of approx. 2430s)
together with new finds of graptolites led to more precise dating of sedimentary successions
and to the proposal of a new stratigraphic subdivision as well as a new geodynamic
interpretation of basin evolution (BAHLBURG under review, BAHLBURG et al. under review).

Depositional history of the Puna Basin

In the Tremadoc, sedimentation in the eastern part of the northern Puna (Sierra del Cobre,
Fig. 1) began with a transgressive succession of intertidal quartzarenites and intercalated
pebbly mudstones giving way to turbidites indicating increased subsidence during the Lower
Ordovician.
In the western part of the northern Puna the oldest fossiliferous rocks are represented by a
volcanosedimentary unit (Volcanosedimentary Succession (VS), Fig. 1) of middle to late
Arenigian age, which include the middle Arenigian Aguada de la Perdiz Formation (Fig. 1) in
the Chilean Puna and upper Arenigian strata present in the Argentine Puna. The VS is about
3500 m thick and consists in its lower part of basaltic to andesitic lavas, hydroclastitic
rocks, tuff breccias and debris flows. In the upper part, alternating ash tuffs and
228

volcanoclastic turbidites document the fining upward trend of the succession. The VS
represents a submarine apron on the back-arc flank of a mainly subaerial silicic volcanic
arc of an eastdipping subduction zone at the Pacific margin of Gondwana (BREITKREUZ et.
al. 1989). There is no evidence of active volcanism in the Ordovician after the Arenig.
To the east of the depositional site of the VS, the Puna Turbidite Complex (PTC)
encompassing the Lower and Upper Turbidite Systems (LTS, UTS) formed from the late
Arenig to the Llandeilo (-?lower Caradoc)(Fig. 1). The PTC includes parts of the Coquena
Formationand the Calalaste Group (Fig. l)(BAHLBURG et al. under review). The LTS is at
least 2700 m thick (BAHLBURG et al. 1988), the UTS at least 900 m. The volcanoclastic
turbidites of both systems are usually laterally continuous deposits of depositional lobes.
However, channel complexes reach thicknesses of up to 200 m. Sediment transport was
axial and almost uniformly directed towards the NNW (Fig. 2). The turbidite successions of
the approx. N-S striking elongate Ordovician Puna Basin are comparable to elongate,
efficient turbidite systems of active thrust basins (MUTTI 1985).
The absence of regional proximal-distal trends indicates, among other evidence, that the
depositional system was not a point sourced submarine fan. Detritus originated almost
exclusively in the line source of the magmatic arc (Fig. 2) which seems to have been inactive
after the Arenig. Immaturity and bad sorting of turbidite arenites in fining upward as well as
coarsening upward cycles of different order attest to short transport distances and minor
weathering and reworking in small repositories and shelf areas throughout all cycles and
sections. Although pnd order cycles in both turbidite systems appear to be paralled by
respective global sea level changes, 3rd and gth order cycles have different trends from
section to section. In combination the data show that sedimentary trends and apparent sea
level changes are primarily related to tectonic activity in source regions and the basin itself.
Global sea level changes during the Ordovician (FORTEY 1984) only complemented the
influence of tectonic activity on depositional patterns.

Geodynamic evolution

The Ordovician basin in the Argentine-Chilean Puna developed in the Arenig as a back-arc
basin (Fig. 3). Incremental sediment decompaction leading to calculated geohistory plots
shows that from the late Arenig onwards the basin in the western part of the northern Puna
subsided rapidly at rates of up to 1.1 mm/a. Subsidence rates of this order of magnitude are
typical of foreland basins (ALLEN et al. 1986). Basin subsidence most probably originated
from the overthrusting of the Arenigian arc complex onto its former back-arc basin which
transformed the back-arc basin to a foreland (successor) basin. This was accompanied by
tectonically induced deposition of thick turbidite systems (LTS, UTS). The events were
probably caused by the onset of collision of the allochthonous Arequipa Massif with the
Gondwana margin in NW Argentina and N Chile (e.g. RAMOS 1988). As a consequence of
eastward thrusting of the arc complex and the development of a foreland basin, a flexural
bulge formed at the eastern margin of the basin. Formation of the bulge led to uplift and
eventual emergence of the shelf in the Cordillera Oriental in the early Llanvirn and therefore
to the end of the Ordovician stratigraphic record (Fig. 1) in this area during the Guandacol
diastrophic phase. The Puna Foreland Basin was closed during the Oclbyic Orogeny at the
Ordovician-Silurian transition. Folding of the basin fill was caused by the terminal collision
of the Arequipa Massif.
An important role in the geodynamic interpretation of the Ordovician Puna basin and this
segment of the Gondwana continental margin plays the extended magmatic belt of the Faja
Eruptiva de la Puna Oriental. This belt was interpreted as a magmatic arc consisting of
submarine ignimbrites intercalated with and contemporaneous to the sedimentary rocks
desribed above (e.g. COIRA et al., 1982). The Puna basin was therefore alleged to have
been located in a fore-arc position. However, the outcrops of the Faja Eruptiva Oriental
north of 2430s consist exclusively of silicic, porphyric and equigranular partly sheared
high level granitoid intrusions crosscutting folded Late Ordovician sedimentary rocks. The
intrusives therefore have a maximum age of very late Ordovician (see also MENDEZ et al.
- ., _.. ^ _
.._-. _..-__

229

Northern Puno Sierra Cordillero


nw#irW0rlw I del Cobre Oriental

Lianvirn ;zo;; $ TurbiisysWn


-__, ____, __,r- ----------_
M%ZrLWU ;wLyV --r?----7--
Arcnig
---7 --__ -7 __&~__-__?_-. Chiquwo
0
8 t Formotion
I
Tremadoc I

;
Nwr
Cambrian

Fig. 1: Stratigraphic table for the region under consideration; modified after ACENOLAZA 8,
BALDIS (1987) and MOYA (19881, and own results.
.:
,., ,. , ;;

m..F
., ... , .

.
.
..
:,;.. : .

. ,

,,,. .
.:, :.

.:::
Fig. 2: Sedimentological model of the
,.
,
0, Puna Foreland Basin (Middle
.,
.
Ordovician).
..

,;) .: ,.

Fig. 3: Schematic sketches of the


geodynamic evolution of the
Ordovician Puna Basin.
230

1973). Geochemical data and the abundance of (meta)sedimentary xenoliths indicate that
the granitoid magmas suffered considerable crustal contamination. The granitoids are cut
by N-S striking subvertical shear zones along which leftlateral strike-slip movements of yet
unknown extend occurred. Shearing took place in a low pressure regime at temperatures of
between 300 and 500%: as opposed to quartz grains, feldspars show no recrystalization
phenomena.
Conclusively, the Faja Eruptiva de la Puna Oriental in the northern Puna is not the mag-
matic arc contemporaneous to the studied Ordovician sedimentary rocks but a belt of late-
to post-tectonic intrusions of very Late Ordovician to Silurian age. The axis of the actual
Early Ordovician magmatic arc was probably located to the west of the VS. This arc may
have been related to the magmatic belt of the Faja Eruptiva de la Puna Occidental (PALMA
et al. 1986) situated west of the Puna in northern Chile.

References

ACENOLAZA, F.G. & BALDIS, B. (1987): The Ordovician system of South America.
Correlation chart and explanatory notes.- IUGS Publication No. 22, 68 S.
ALLEN, P.A., HOMEWOOD, P. & WILLIAMS, G.D. (1986): Foreland basins: an
introduction.- Int. Ass. Sediment. Spec. Publ. 8, 3-l 2.
BAHLBURG, H. (under review): The Ordovician basin in the Puna of NW Argentina and N
Chile: geodynamic evelution from back-arc to foreland basin.- Geotekt. Forsch.
BAHLBURG, H., BREITKREUZ, C. 8 MALETZ, J. (under review): The Ordovician
sedimentary rocks in the Puna of Argentina and Chile: New stratigraphical data based on
graptolites.- Newsl. Stratigr.
BAHLBURG, H., BREITKREUZ, C. 8 ZEIL, W. (1988): Geology of the Coquena Formation
(Arenigian-Llanvirnian) in the NW Argentine Puna: constraints on geodynamic
interpretation.- Lect. Notes Earth Sci. 17, 71-86.
BREITKREUZ, C., BAHLBURG, H., DELAKOWITZ, B. & PICHOWIAK, S. (1989): Volcanic
events in the Paleozoic central Andes.- Jour. South Amer. Earth Sci. 2, 171-l 89.
COIRA, B., DAVIDSON, J., MPODOZIS, C. 8 RAMOS, V. (1982): Tectonic and magmatic
evolution of the Andes of northern Argentina and Chile.- Earth Sci. Rev. 18,303-332.
FORTEY, R.A. (1984): Global Earlier Ordovician transgressions and regressions and their
biological implications.- In: Aspects of the Ordovician System (Ed. by D.L. BRUTON),
Paleont. Contrib. Univ. Oslo 295, 37-50.
MENDEZ, V., NAVARINI, A., PLAZA, D. & VIERA, V. (1973): Faja Eruptiva de la Puna
Oriental.- 5. Congr. Geol. Argent Actas 4, 89-l 00.
MOYA, M.C. (1988): Lower Ordovician in the southern part of the Argentine Eastern
Cordillera.- Lect. Notes Earth Sci. 17, 55-70.
MUTTI, E. (1985): Turbidite systems and their relation to depositional sequences.- In:
Provenance of arenites (Ed. by ZUFFA, G.G.), Nato ASI series C 148, 65-93, Reidel,
Dordrecht.
PALMA, M.A., PARICA, P.D. & RAMOS, V.A. (1986): El granito de Archibarca: Su edad y
significado tectonico, provincia de Catamarca.- Rev. Asoc. Geol. Argent. 41,414-419.
RAMOS, V.A. (1988): Late Proterozoic-Early Paleozoic of South America - a collisional
history.- Episodes 1 1, 168-174.
231

pndean basin dvnamics in northern Chile

G.D. Williams1 , P. Turner% S. Flint,3 I. Stipmsonl 3


E.J. Jolleyz and A.J. Hartley4

1 Department of Geology, University of Keele, Keele, Staffordshire, U.K.


2 School of Earth Sciences, University of Birmingham, P.O. Box 363,
Birmingham, U.K.
3 Department of Earth Sciences, University of Liverpool, P.O. Box 147,
Liverpool L69 3BX, U.K.
4 Department of Geology, University of Wales, Cardiff, U.K.

The Andean forearc of northern Chile comprises Triassic-Recent arc-related


magmatic rocks and sedimentary basins. Origin, position and internal facies
architecture of the sedimentary basins can be related to arc dynamics and stress
regime.
The Andean margin has been in net extension/subsidence over much of the Mesozoic-
Recent, related to poor coupling between the Nazca and South American plates. Regional
contractional events, linked to plate rearrangements and changes in convergence
angle/rate, include mid Cretaceous back-arc basin collapse (major source rock) and
retro-arc thrust belt formation. This crustal reorganisation was succeeded by episodic
eastward volcanic arc migration and associated foreland thrust front propagation
through Tertiary times.
An intermontane basin developed to the east of the thrust belt during the Late
Cretaceous. Subsequent eastward thrust front migration inverted part of this basin,
sourcing an Oligocene basin to the east. This pattern of eastward-stepping depocentres
being uplifted as piggy back basins,,sourcing the next younger basin to the east, is still
active in the Recent Salar de Atacama hydrocarbon play, where an intrabasinal
foldbelt (Cordillera de la Sal),is uplifting on a thin-skinned rotating thrust sheet.
Although lying to the foreland side of thrust belts, these basins were formed on
actively thickening crust (nbw some 70 km); this questions the importance of thrust
loading for basin formation in this crustal segment.
Basin-fills are characterised by alluvial fan, aeolian and lacustrine facies
associations. Sediment dispersal patterns (and thus resenroir position/geometry) are
complex but predictable, reflecting the interplay of thrust system dynamics and
palaeoclimate.
233

TRIASSIC DEPOSITIONAL BASINS IN NORTHERN CHILE

C.M. Bell * and M. Suares **

* Department of Geology, Cheltenham and Gloucester College


of Higher Education, Cheltenham, GL50 2RH, United Kingdom.

** Servicio National de GeologZa y Minerfa, Avenida Santa


Maria 0104, Santiago, Chile.

Key words : Andes, Triassic, sedimentary basins,


depositional environments

Sedimentary basins with associated rhyolitic, andesitic and


basaltic rocks developed during Triassic times as a result
of extensional (possibly transtensional) tectonism in Chile
(Charrier, 1979; Suarez and others, 1985) and Argentina
(Stipanicic, 1983; Kokogian and Mantilla, in press; Uliana
and Biddle, 1988). Most of the sediments were deposited in
a continental environment but some marine strata have been
recorded from Chile. A NW-SE orientation for the basins
has been demonstrated in Argentina (Uliana and Biddle,
1988).

The Triassic marks a major change in the geological


evolution of Chile and Argentina. Paleozoic terrane
amalgamation and subduction-related magmatism and accretion
was followed by widespread non-erogenic silicic magmatism
of the Choiyoi Group during the late Carboniferous to early
Triassic (Bell, 1984, 1987; Kay and others, 1989; Ramos and
others, 1986). BY Jurassic times a well-defined
subduction-related island arc - back arc system,
characterising the Andean orogeny, had de,veloped parallel
to the prr%ent day coastline.

The present study includes a summary of the conclusions of


sedimentological, palaeogeographical and tectonic
investigations of four major but geographically isolated
occurrences of Triassic sediments between 26 and 29s in
northern Chile (Fig. 1). The San Felix Formation is Middle
or possibly Lower Triassic in age and the Cifuncho, La
Ternera and Dona Ines Chico Formations are attributed to
the uppermost Triassic (Charrier, 1979).
235

The Cifuncho and La Ternera Formations contain thick


successions of alluvial fan and ephemeral braided stream
deposits. These, together with the halite casts and playa
lake deposits of the Cifuncho Formation, suggest an arid to
semi-arid climate. However, the Triassic formations are
also characterised by an abundant fossil flora and thick
mudstone successions which are indicative of a more humid
climate.

The sedimentary succession of the San Felix Formation can


best be explained by initially rapid subsidence, with
subsequent turbiditic prodelta sedimentation, related to
steep slopes in a near shore but deep water marine
setting. The thick sequences of coarse-grained and
poorly-sorted alluvial sediments of the Cifuncho and La
Ternera Formations were deposited on alluvial fans and
braid plains in fault-bounded and actively subsiding
intermontane grabens or foredeep basins. By contrast the
Dona Ines Chico Formation was depositedon a stable coas ta1
plain or delta. Variations in thickness and sedimentary
facie8 in the Triassic successions indicate a remarkable
range of depositional environments and settings. The
sedimentation probably occurred in a number of tectonically
active and isolated basins between uplifted blocks.

The scattered and isolated nature of the Triassic exposures


in northern Chile mean that the position and shape of the
sedimentary basins, and the associated magmatic arc, are
poorly constrained. This problem is compounded by the
presence of major thrust and strike-slip fault sys terns
which have probably produced significant lateral
displacements between the exposures.

Calc-alkaline volcanic activity suggests that the Triassic


sedimentary basins resulted from intra-arc extension (or
transtension) associated with active subduction, during the
earliest stages of the Andean orogeny related to the
breakup of Gondwana.
I___.. , __..-_.

236

References

Bell, C.M. 1984. Deformation produced by the subduction of


a Palaeozoic turbidite sequence in northern Chile.
Journal Geological Society, London, 141, 339-347.

Bell, C.M. 1987. The Late Paleozoic evolution of the


Gondwana continental margin in northern Chile.
Gondwana Six: structure, tectonics and geophysics.
Geophysical Monograph 40, Amerrcan Geophysical Union,
261-270.

Charrier, R. 1979. El Trigsico en Chile y regiones


adyacentes de Argentina: una reconstruction
paleogeogrlfica y paleoclimgtica. Communicaciones, 26,
l-37.

Kay, S.M., Ramos, V.A., Mpodozis, C. and Sruoga, P. 1989.


Late Paleozoic to Jurassic silicic magmatism at the
Gondwana margin: Analogy to the Middle Proterozoic in
North America? Geology, 17, 324-328.

Kokogian, D.A. and Mantilla, O.H. In press. AnZlysis


estratigrgfico secuencial de la Cuenca Cuyana :
secuencias depositacionales continentales. In:
Simposio sobre cuencas sedimentarias de la Argent i:,
X0 Congreso GeolBgico Argentino.

Ramos, V.A., Jordan, T.E., Allmendinger, R.W., Mpodozis,


C Kay, S.M., Cortes, J.M. and Palma, M.A. 1986.
Piieozoic terranes of the Central Argentine - Chilean
Andes. Tectonics, 5, 855-880.

Stipanicic, P.N. 1983. The Triassic of Argentina and


Chile. In: Moullade, M. and Nairn, A.E.M. teds) The
PhanerozoG geology of the World. 11 Mesozoic, B,
181-199. Elsevier.

Sugrez, M., Naranjo, J. and Puig, A. 1985. Estratigrafia


de la Cordillera de la Costa, al aur de Taltal: etapas
iniciales de la evoluci6n andina. Revista Geol6gica de
Chile, 24, 19-28.

Uliana, M.A. and Biddle, K.T. 1988. Mesozoic - Cenozoic


paleogeographic and geodynamic evolution of southern
South America. Revista Brasileira de Geociencias, 18,
172-190.
238

Southern Subandean Andean

ER R S

Late Jurassic
Late Oligocene

lchoa Fm

__________________--- ---- ________-__--____


LY
Ravelo Fm -
e
-1 baaLlllll rlll I-- I
I I-,\ I- I

Tapecua Fm
Entre Rios Basalt

Ipagu;32~ _ km
-
I
EarlyTriassic I I200 m
\
\ Sawi Fm1-j

A:Simplified stratigraphic sections


(locallzeci in
Rgml-
of the Entre Rios area (ER) , Ravelo
and Sayari (R)
(S). 1:carbonates; 2:mudstones; 3:medium- to fine-grained
sandstones; 4:coarse to medium-grained sandstones;
S:conglomerates; 6:mos;tly alluvial facie5 (channels);
7:mostly eolian facie5 (dunes); 8:evaporites; 9:basalts.
B:Location of stratigraphic sections shown in A.Dotted
area5:known (or supposed) extension of the Middle
Triassic-Jurassic sequence discussed in text m Political
borders are shown by dashed lines.
.
..-..-.a -1-

239

latter is capped by a thick silcrete, on which lies in pro-


nounced disconformity the Late Oligocene Petaca conglomerate,
Sedimentary continuity characterises the Ipaguazu-Ichoa inter-
val, which must thus cover a continuous time span.

Central and northern Subandean belt

Eolian and minor fluvial deposits compose the Ichoa and Beu
Fms in these areas. Both units are disconformably overlain by
fossiliferous Late Cretaceous deposits, and post-date normal
faulting and block tiltings that affect units as young as
Permian.

Cordillera Oriental

Most of the Cordillera Real granitoids were emplaced at the


roots of a Middle to Late Triassic NW-trending rift system
(2). The Sayari and Tiquina Fms post-date Late Permian-Early
Triassic units and consist of fluvial, locally gypsiferous,
red beds; they do not extend outside 2 narrow NW-trending
strips defined by their outcrops, which probably represent
paleo-grabens. The Sayari Fm is overlain by the extensive
Ravelo Fm, which comprises a fluvial member followed by a
thick eolian member (fig. 1). The Ravelo Fm may be more than i
km-thick, contains several basalt flows and locally shor;s
paleoalteration at its top. It is usually truncated, in many
localities to complete erosion, by coarse red beds overlain by
Cretaceous units. This sharp unconformity is assigned to the
widespread Eimmeridgian Araucan event (3, 4). No eolian
sandstones are known from undisputed Cretaceous units in An-
dean Bolivia.

Correlations, age of the sequence and conclusions

It is clear (fig, 1) that the units described above belong


to the same sequence, which seems to cover the Middle
Triassic-?Oxfordian interval and is characterised by : (a)
initial rifting processes, with deposition of (gypsiferous)
red beds in narrow troughs; (b) extending fluvio-eolian sandy
sedimentation; (c) extensive erg development.
A similar evolution, albeit lacking the initial rifting, is
known from the Parana basin of Brazil, where the sequence is
assigned a well constrained Middle Triassic - Jurassic age
(5). Middle Triassic rifting is known from western Argentina,
Chile and Peru, where it is followed by Late Triassic and/or
Jurassic marine ingressions (6). These data can be summarised
cartoon cross-section of western Gondwanaland in the
za"rlt Jurassic (fig. 2). It shows the development of an exten-
sive desertic environment on the inclined craton that flanked
the subsident carbonate basin established on the Pacific mar-
gin.
.I_. . -me

240

w E

I ?h
carbonate ergs on cratonic areas
back-arc basin

aborted rifts
= dwachment 7

Tarapaci basin Parani


A/lip/an0 Porosi- Chaco basin basin

Fip.2. Cartoon cross-section across part of the Liassic


Western Gondwanaland.No vertical scale.

References

(1) T. Sempere et al., 9th Boliv. Geol. Cong., in press, 1990.

(2) D.J. Kontak et al., in Magmatism at a plate edge : the


Peruvian Andes, k'.S. Pitcher et al., eds., p. 36-46, 1986.

(3) T. Sempere et al., 6th Chil. Geol. Cons., v. 3, p. H37-63,


1988.

(4) E. Jaillard and T. Sempere, GSGP Symp. on Latin Amer.


Cret., p. Al-25, 1989.

(6) P.C. Soares, in Cuencas sedimentarias de1 Jurasico y


Creticico de Amzica de1 Sur, v. 1, p. 271-304, 1981.

(6) J.J. Zambrano, in Cuencas sedimentarias de1 Jurasico y


Cretacico de America de1 Sur, v. 1, p. 9-14, 1981.
241

Evolution tectono-sbdimentaire dans le Crbtack du


synclinal dOtavi-San Lucas (Bolivie centre-sud)

par C. MARTINEZ *, E. VARGAS ** et G. LAUBACHER *.


* Institut Fran+ de Recherche pour le D&eloppement en Cooperation (ORSTOM),
BP 5045, Montpellier cedex 01, et Laboratoire de GCologie des Bassins, USTL,
place E.Bataillon, 34095, Montpellier cedex 01.

** Universidad Mayor de San A&es, La Paz, Bolivie.

Abstract
In the southern and central Andes of Bolivia, cretaceous marine and continental strata are
preserved in several andean synclines. Extended and continue tectonic instability affect the
cretaceous sedimentation in many places. The Otavi-San Lucas syncline is a nice example
where important synsedimentary tectonic activity, of early to late Cretaceous age, was
evidenced.

Introduction
A p&s de 70 km au SSE de Potosi (fig.l), le synclinal dOtavi-San Lucas (2000 S et
6515 W) est situ6 dam la Cordillbre orientale entre les localit& dotavi, ?IlW, et de San
Lucas, B 1E. Orient6 NNW - SSE, avec une longueur
de 30 km et une largeur de 10 km, il est constituC
dune Cpaisse s&ie de terrains du &tact-Tertiaire
inftrieur, continentaux et marins, reposant en
discordance angulaire sur le PalCozoYque inf&ieur
(Vargas, 1984).

Dans la rtgion NW du synclinal, la sCrie


crttacte pr&ente des variations de facibs et
dCpaisseur, des discordances angulaires, des failles
scell6es et des niveaux volcaniques, qui soulignent &
petite et grande Cchelles lexistence des d&formations
tectoniques synskdimentaires. Ces dtformations
accompagnent la sddimentation depuis la base du
C&ta& jusqug la discordance du SCnonien (fig.2).

Evolution tectono-sddimentaire
Le C&act de Bolivie a fait Iobjet de travaux synthttiques de la part de Russo et Rodrigo
(1965), Kriz et Cherroni (1966), Reyes (1972), Sempere (1986) et Scmpere et al. (1988). Dans
ce demier travail, une succession shuentielle est proposCe qui sert actuellement de rkfdrence.
Nos observations dans le synclinal dOtavi-San Lucas, nous permettent de distinguer,
sous la discordance du SCnonien, 4 mCgasQuences (C2 g C5, fig.3), que nous avons tent6 de
corr6ler avec la succession de Sempere et al. :
1. La m6gasCquence de base (Cl), coimue au N de Potosi (formation Macha,
Sempere 1986, Martinez et Vargas 1988; ou formation Ravelo, Sempere 1988), na pas Ctt
observCe ici et les d6pp8tsmCsozoiques sont discordants sur le Paltozoique inf&ieur par une
sequence (C2a) de conglom&ats ef de g&s conglomQatiques rouge-viola& (formation Condo),
que surmontent des lutites et des g&s rouges (formation Kosmina). Des coultes basaltiques,
depaisseur variable, g&&alement al&es, sintercalent dans cette sauence.
Au-dessus, souvent en apparente continuitC stratigraphique, une autre sequence (C2b) est
reprksende presque uniquement par des g&s blancs 3 rods, en gros banes ?I chenaux et a
stratifications obliques. Elle est a mettre en paralltle avec les formations La Puerta et Sucre,
242

reprksentees dans la rkgion de Potosi par des facibs fluviatiles et deltaiques (Sempere et al.
1988). Dans la tertninaison N du synclinal, cette sequence est affectee par des accidents NW-
SE et E-W qui generent des horsts, des grabens et des basculements de blocs (fig.2).

Des le debut de la mCgasCquence C2, des failles dkametriques B hectomktriques,


intrafotmationnelles. sont successivement scellees; leur jeu controle la prCsence ou non des
coulees basaltiques et les brutales variations dtpaisseur des conglomerats, des gres et des
lutites de C2a. Leur activite setend jusqua limportante discontinuite qui &pare les
mCgaskquences C2 et C3, les gres C2b pouvant etre absents.

2. La mCgasCquence C3, que nous proposons dappeler formation Tambillo,


repose en discordance angulaire sur les differents termes de la megasequence C2 et les failles
intraformationnelles pr6cCdentes. Cette megasequence se depose aussi dans un contexte
dinstabilitb tectonique forte soulignte par des discordances successives, des onlaps, des
coultes andCsitiques qui sont, elles memes, remanites dans des conglomkrats grossiers. Ces
demiers sont surtnontCs de grbs roses a rouges, bien stratitiCs et ?I laminations obliques,
pouvant d6passer la centaine de metres dkpaisseur. Cette megasequence C3 peut Etre mise en
relation avec la mince sequence S3 que Sempere (1986) observe entre les formations La Puerta
et Tarapaya, quil ne mentionne plus dans sa dernibre synthese (Sempere et al.1988), mais qui
semble comcider avec une importante discontinuitt dans la serie du CretacC inferieur du bassin
de Potosi.
M.... ,,

243
244

3. La mCgast5quence C4, qui vient en concordance angulaire apparente sur les


grts rouges de la megas&prence C3, comporte plusieurs dizaines de metres de lutites et de silts
rouges (C4a), rapport& a la formation Tarapaya, et un horizon calcaire et greseux (C4b). de 10
a 20m depaisseur, que nous sommes tenter de corrkler, au moins provisoirement, avec la
formation MirafIores dige cdnomanien. Cette mkgasequence C4 ne semble complete que sur le
flanc ouest du synclinal.
Un important rejeu, plurihectometrique. probablement decrochant, des accidents
anterieurs BC4 nous parait responsable de labsence (erosion, non depot) dune partie de la C4
et de la mtgasequence C5 (formation Aroifilla) qui la surmonte SIXle flanc ouest mais que nous
navons pas observee a lest de Tambillo.

4. La discordance du SCnonien (C6: formation El Molino).


Dans la terminaison nord du synclinal, les grandes failles NNW-SSE et EW, qui
accompagnent la sedimentation cretade avec des rejeux cumules importants, naffectent que
peu ou pas le C&ace terminal. I1 est clair que la transgression senonienne sest effect&e sur
une rdgion ddjja tr&s strocturke.
-Lanaiyse par photos atriennes de lensemble du synclinal suggtre une discordance
gtnCralisCe du C&act terminal (C6). a la fois sur le PaltozoYaue et sur les diverses stauences
&tacCes anterieures. Cette discontinuid majeure a cachetk les failles, mais celle&i sont
partiellement reutilisees lors des phases andines de serrage, soit quelles rejouent en failles
inverses, soit quelles fixent des synclinaux pin&s.
Le coeur du synclinal est occupt par une formation g&o-lutitique rouge (formation
Santa Lucia, PalCoghe) et par une s&e conglomkratique dlge teniaire, sans plus de precision,
ce qui indique sans 6quivoque que la formation est aussi tertiaire.

Conclusions

Dans les bassins cr&acCs de Bolivie, il devient un fait ghkal que la skdimentation sest
effect&e dans un contexte dinstabilitk tectonique. La dtformation a dCbutCdks avant le dCp6t
des sequences basales du synclinal dOtavi-San Lucas et elle allie frequemment des
phenomtnes distensifs et compressifs (failles inverses, plis et bombements synsedimentaires).
cest le cas a Macha (Martinez & Vargas 1988) et ?ITica Tica, situ& respectivement a 150 km
au NNW et 120km a IW de Tambillo. Nous observons. dans le svnclinal dotavi-San Lucas.
que la dbformation, caract&isCe par des failles, des discordancks, des conglomkats et d;
volcanisme basique puis andtsitique, se dheloppe de faGon quasi continue pendant tout le
C&act infkieur et supkieur. Cest elle qui contr6le la ripartition locale des skquences et done
leur prksence ou leur absence. Les donnCes actuelles ne permettent pas encore de prkciker le
rkgime des contraintes correspondant. Cependant, la structuration observke ici est compatible
avec un senage approximativement NW-SE associe a une extension NE-SW.

RCf&ences

Kriz S.J. y CherroniC.. 1965 - Diagramas correlativos de formaciones cretacicas dcl suroeste dc Bolivia. Serv.
Geol. Boliv.. Hoja informativa II. La Paz.
Martinez C. & Vargas E., 1988 - Sur les deformations synddimentaires mesozoiquesde la region dc Macha-
Pocoata-Colquechaca(nord de Potosi, Cordillbre.orientale de Bolivie). Geodynamique 3. (1-2). 107-I 15.
Reyes EC., 1972- Correlaciones en el Cretacico de la cuenca andina de Bolivia, Peru y Chile. Rev. Tee. YPFB,
1, (2-3). 101-104.
Russo A. y Rodrigo L.A.. 1965 - Estratigrafm y paleogeogralla de1Grupo Puca en Bolivia. Bol. Inst. Bol. Petr.,
5 (3-t), 5-51, La Paz.
Sempere T.. 1966 - Conoibucion a la estrarigrafia de1 Mesozoic0 boliviano en el dominio andino. Publication
Mission ORSTOM. 1, 34 p, La Paz.
SempereT., Oiler J., Barrios L., 1988 - Evolution tectonosedimentaria de Bolivia durante el Creracico. Vto
Congr. Geol. Chileno. II, H35-H65.
245

CREI-~OUSANDEAFKY- IN NORTHERN CHKE EEIWEEN 219AND2ss.

l Tomiskv Bogdnnic, HRelnhafd Di5bel

l Depto . Geodmdas Univ. del Nor&i, Ca8illa 1230, Anbfagasta - Chile (at present lnrt tiir
Geologic dw FU Serlln

w Institut1QrGeokgie der Fmien Univemltit Berlin,AlMsteinstr. 34a, 1000 Berlin 33, West Oer-

Se pmentanloa ndtados de un lstudo en ltlotamlentcwcziAs_ ntakr y vok&nkoa del


Cmtbko - TetrAario InMor. EL&M dep6attua npmentan dibnnbr lstndm en la adividad
ma~~dd~lloQImarpenoonthsntal~riornoenutru~dslorAnder,con-
llgurandodonJnbrpnleogeoqaneordlbrentupurcadaMlomrrgma~.

The study ama ia located in the Central An&r, in the Chilean Pfu-Cordilkn (Siina de Mormo -
Siena del Medo) L&v&en 219 and 235 II Regbn of Antofagasta, Chile (Fig. 1). The sbdy la tta-
rod on verdlcalse&m8 and gedogkal mapping of the ama8 concerned This inveetigatkn ham
pemltted dssctiptionrand a rsdellnltkn of tie vertical and horizontalexbmion, age and mgknal
rlgnltkanca of the forrnatknrpRviously&scribed aa Cmtawous. The object of the investigati-
on war b dsraib ti uquences and analyu their rtmtigraphkal and palaOgeOgfaphlcal8ignM
MawiWn~~utknoonbaxtdIhlrpartof~Andea~areruldthtrtudy~ouruque~
cm could be dstiguished on the bask of Mokgy, rbatigraphk nhtionr and age.
I. .__... ____ _.,,...

246

The seqences, trum Kimmeridgianto Eocene,(tig.2) conaspcnd tc:

a) The Western Sequence (Kimmeridgianto BanvmhnlAptian?) occurs only on the wstem tlank
of the Siemr de Momno with tine grained 5000 m thick sechents (sandstonesand siltsbnes). It is
of lhvial+Maic characbr, math ln its Iowr pus and show pbocwnwts trcm E and NE. The
sequence develops amMnably ltcm Jufasslc marine sedments (Oxtordian amYor basal Khr-
meridglan)and is ccntcfmably cverhin by anderiter ot Vclcanic Sequeno 1 (see below). These
sedmentscompond b themldde and upper part ot the Chacarilla Formationot GALU a DING-
MAN, 1962.

Fig. 1: Loalicm of the invedigated area in tfm ReCordUa of Nattwn Chile


247

b) Vokanio Sequena 1 (AIbian to CenomanC


an)ocoueontyonthewsMmUanko?Uw
Sbrm ds Momno. Its 1000 to 1500 m thidc
vdeanbrocksarecomposeddiavabedsand
voNanl0 breedas. The seqwna developr
conmrmablytmm the Wesmm Sequewe and
is uncmnkmnabtywertain cy CUgocene-Mio
ten gravels (Skhal FormatIon, sensu MAK-
SAEV, 1978)an&rhasataultountactti
Me Eastern sequence (see bekw), L ags is
based on a WR tUVSr isodwn v&h maximum
of 104 + 19 Ma (ROGERS, 1985). Furhwmo-
n, ens of the autbrs (FL Dlibel) dated monzo
nkk skxks ot DueMada Amas which had In-
Uudsd Into Uwse vokank rocks (WAr-age In
bbur: 9g43.1 Ma) and dkes nlaw to U-w
se stocks Mat had h&dad hto lhe Wesbm
SeqUOfMX(Ar/Ar In homblende: 924229 ma).
On the basis of lhe stratigraphicrelations,the-
wvoloankmcluoanbeassumedtokmrparl
of the Gem, Eftpexa Formation (GALLI &
DINGMAN, 1982).

c) The Eastsm Sequence (Campanian? to


Early Eocene) outuop along the Sierra dsl
MeBo and is cornpored of coarse grained
1900 m U&k aUuvial-fanch~osits (congloma
Fig.2: GmnraNzadW~oahtlgqhic columnoof fhr Sierra da ra&!s)with pahoanents tram S and SW. The
hknno and Slum dol Medo in fho Re-Cwdllua of N-Chile. contact Doths undsriyingmarine Jurassicredi-
ments (Csllovian amYor Cxkrdian) is tied
by en emsicnal unconMmUty.The sequence is contonnabtyovenain by Vokank Sequence 2 (see
bskw). This Eastern Sequence cornsponds to the Tolar Formation (MAKSAEV, 1978) - lo
ww/middb partot Um PurilactisFormation(sensu CHARRIER & REUlTER, 1988).

d) Volcanic Sequence 2 (middblupper Eocene) outcrop in the Sbna dsl Medb and in Uw sow
~empartoftheSbnadrMorsno.Itiscomposedd1OOO01500mWdcblffand~Ucda
posits of caltxlkalhe cornPosition.This sequence develop w&wmab& from the Eastern Se-
quence and is ovettain unconkrmably ty Cilgocene - Mbcene gravels (Skhai Forma&m, MAK-
SAEV, 1978). its vokania rocks wsm dswd (D&el) with Uu AnAr - method h Motib snd hom-
blench (betwen 47.820.8 and SS.S_S.SMa). These results cormlaaswith the ta&metric ages or
HUETE et al. (1977) Mr similarrocks (WR K/Ar in MotiM, be-en 55.8+0.8 and 41.220.8 Ma). On
the basis of the straUgraphk mlaUons and radiometricdating the volcanic mcks can be said to
conespond to the kanche Fonaticn or MAKSAEV, 1978.

- On the basis of ths compodUona of sanda~es (Tttsngular-Diagramr,at&r DICKINSON, 1985) It


is posslbb b deduce the fotlowhg: the We&m Sequence b d wcbd erogenic Provenancei.e.
these dsMUcsedimentswere depositedduringa tw&onal period betwen hw &tonic regimes
(istandam to continentalmargin?).The EasMm Sequence show also a recycledongenk prova
nawa but with ths existence ot a consolkM0sdnUIgmatk am - Vdcanio Sequence 1 Ir associated
with Inbar-medials intrusive8v&h age8 d between 100 and 80 Ma (ROGERS, 1989). It corn
spnds to the end of a magmatic actMy associabd wiufththe effscts of tectonicmovemenb of lh0
248

Subhefqnbn 01 P~NvM Phase (senw MEt3ARD 1997) of Santoninnage.

2) Awan b Cenomanian- in lhe amwlt Chilean Pm-Cordiwa.

S) Campanian?b Eocene - in ti wtmnt Chilean PnCordilleta

-Thlr~ntUld~~~onot~~nntmclgmatlcurrrhowracydid~dcMnb.
Ewy qcie stub wilh very acUvevdcanfsm. the in9usicn of prubns that pduce alterationand
mberalkaUa~, acccmpanbd by wty acUvektcnb movementr, pm&ced along major faults (Ata-
cama Fault West Fissum, sensu REUTTER & SCHEUBER, 1988). Tecbnb and magmatb activity
finallytacome mlativetycalm and thbk continental8edimentary8ehr am &posibd

CHARRIER, R. & REUnER, K-J. (1958): la Forma&n PuMciis en el bard, ocddental del Se-
lards Atacama, 23%?%5 de kitud Sur, Chile. - Ccmmunicadorq 34 211. Santiago.

DICKINSON, W.R. (1985): Intefpmting provenancemlatbna fhnn de9ital modes of sandsbnw. -


IN: ZUFFA, Q.G. (ed.) Pfwe~ncll ol Amtke8. Durlmcht Rebel, 333-351. Holland.

OALLI, C. b DINOMAN, R.J. (1952): Cuatiguba Pi Aba, Matilla y Chaauilk Prwinda de


Tampa& -In& lnwst God. Chile, Carla Geol. Chile, N*. 7-10, 125 8. Sentiago.

HUETE. C.; MAKSAEV, V.; MOSCOSO, R.: ULRIKSEN, C. & VERCIARA, H. (19m: Anbaden-
hr gec&fcnoMgicce de nxas intwivaa de la CoMbm 6 bs An&r compmndidaentm Ir. Siena
de Mcmnc y ll RIO ka y ba 21* y 229 latihrd sur, II Regi6n. Chib. -Rev. Geol. Chile, 4: 35-41.
Santiago.

MAKSAEV, V. (1978): Cuatigub Chkiguay sector Ocddental &I Cuadangub Cerro Paipanas-
lnst Invest Beol. Carta &cl. Chile, Ne 31.55 8. Bentiago.

MAKSAEV. V. ; BORIC. R; ZENTILLI. M. & REYNOLDS, P. (1955a): Slgnifbado metak&nico de


datadones WAr, 4*Ar/% y 9azas de lisi6n de zonaa mineralizadasen el No* Grande de Chile.
-In: Actaa V Congr. Gaol. Chileno,Tomo I: 865-885. Santiago.

MEGARD, F. (1987): Cordkan An&r and MarginalAndes A mvbw of andean geobgy northof
9u A&a Elbow (1eP S). -lnz MONQER, J.W. I FRANCHETEAU, J. (eds.). Ciim-pacific Oroge-
nb Beib and Evdutbn of the Pack! Ocean Baain. Int Uthosphen Program Conk - Geo@namb
Seder, 8: 71-95. Washingbn D.C. - U.S.A.

RECITER, K3. & SCHEUBER, E. (1985): Relatbn betwen &tonica and magmatism in the An-
des of Nodwm Chib and aeacent amaa between 21* and 259 S. - In: A&a V Congr. Gaol. Chi-
bno, Tomo I: Aw353. Santiago.

ROGERS, 0. (1955): A geochemkal travefs@ awos8 the North Chiban Andes. -lJnpubl. Ph. D.
the&, Dept Eulh Sd. Opn Univ. MUkvvKeyner.333 s. London.
249

The Purllactls Group of Northern Chile:


Llnk between arc and backarc
durlng Late Cretaceous and Paleogene

Reynaldo Charrler* and Klaus-J. ReutterW

*Unlversldad de Chile, Departamento de Geologia y Geofislca, Casllla 13518,


Correo 21, Santiago de Chile

**Frele Unlversltllt Berlin, Instltut fur Geclogle, Altenstelnstr. 34a, D-1000


Berlin 33, FR Germany

Resumen

El Grupo Purllactls, de aproxlmadamente 3000 m de espesor y dellmltado


por dlscordanclas angulares, abarca la Fm. Tone1 (arenlscas, yeso), la Fm.
Purllactls (conglomerados, volcanltas), y la Fm. Yeslfera Superlor. Este
grupo se deposit6 en una cuenca de tras-at-co muy cercana al arco mag-
mAtlco del Cretaclco superior hasta el Eocene superlot-.

Key words: Paleogene, Salar de Atacama, magmatlc arc, backarc, orogen-


parallel strlke slip faults.

Introduction

The Purllactis Group, exposed along the escarpment at the western border
of the depresslon of the Salar de Atacama (Fig.11, forms a thick sequence
of red continental clastlcs with volcanic lntercalatlons. It rests uncon-
formably on the late Paleozoic Agua Dulce and El Bordo Fms., and Is
unconformably covered by the gravels of the Oligocene - Mlddle Miocene
Tambores Fm.. Its stratlgraphy, structure and paleogeologlcal relations are
under dlscusslon (Ramlrez & Gardeweg, 1882: Flint et al., 1989).

Stratlgraphy and Structure

The Purllactls Group (8.1.) Is a continuous sequence that can be sub-


dlvlded Into three llthostratlgraphlc unlts (Flg.2):
250

a) Tone1 Fm.: This lowermost formation of the Purllactls Group can well be
studied In a sectlon In the southern part of the area represented in Fig.
1 (lot. purl, pur2). There, It rests with a low angle unconformity on a
thick, dark grey vesicular lava assigned to the Agua Dulce Fm. The 20 m
thick basal member consists of a massive Sandy breccla with dark volcanic
llmestone fragments (5 m) at the base, and an alternatlon of fine brecclas,
grey mlcrltlc llmestones and flne red sandstones and siltstones. It ends
with a conglomeratlc level containing limestone fragments. The following
maln part, approximately 1500 m thick, is formed by alternating members
of thick gypslferous sandstones with large cross-bedding, red nodular
and gypslferous mudstones, red sandstones and thln conglomeratic layers.
Llmestones are Intercalated only wlth the lower part. The top of the Tone1
Fm. consists of 20 m of alternating orange coloured mudstones and
gypslffeus
251

pur3 and NE of pur2) gave 44.02 0.9 Ma and 43.8 +, 0.5 Ma Thus, a Middle
and Late Eocene age can be assigned to the Puriiactls Fm. s.str. The
question arises whether the underlylng Tonei Formation has Its base still
In the Upper Cretaceous. In thls respect, It is tempting to correlate the
basal llmestones of the Tonei Fm. wlth the Yacoralte Fm. In NW-Argentina
If this assumtion prooves to be true, the Purllactls Group corresponds to
the Baibuena and Sta Barbara Sub-Groups of the Salta Group (Marquillas
6 Salfity, 19881, notwithstandlng some differences In facies.

Eastward thrusting of the late Paleozoic substrate over the Puriiactls


Group determines Its structures. In the northern part, the Paleozoic rocks
were pushed upwards In such a way that the basal members of the Purl-
iactis Group were uplifted and turned up to an upright posltlon, while the
upper members are now found to the east In topographically deeper posl-
tlons. In the southern part of the study area, the Paieozolc rocks were
thrust over upper members of the Purilactis Group. Eastwardly vet-glng
folds and several secondary thrusts run through the group Itself. No
cleavage was developed. Local dlapirism can be observed in gypsum-rich
parts of the Tonei-Fm. An important feature are N-S directed dextral
strlke SIID faults which are revealed bv vertical folds which were deve-
loped In upturned strata (Fig.1). These and the shortening structures did
not affect the Tambores Fm. and thus are prior to the Mid Oligocene.

Conclusions

The Purllactls basln was sltuated immedlately to the East of the magmatic
arc of that tlme whose volcanic products (Chile-Aiemanla Fm) were de-
poslted, without Intervening sediments, directly upon a probably elevated
area that had underwent tectonics during the Late Cretaceous. The paleo-
geographic posltion of the Purilactis Group between arc and backarc is
underllned by the interdlgitation of sediments and andesltlc volcanic%
which are completely lacking in the sediments of the Salta Group In
Argentina According to Reutter & Scheuber (19891, dextrai, orogen-par-
aiiei strlke slip faults are a characterlstlc feature of the tectonics, of the
Late Cretaceous - Paleogene magmatlc arc which then was situated In the
Precordiiiera of this region. The presence of such structures In the Puri-
iactis area does connect it to the magmatloc arc, while the llkewlse
present thrusts and folds facing to the east are a feature of backarc
tectonics.

References

DCBEL, R. (1989): Geochemie und Geochronologie alttertlllrer Vulkanlte aus


der Prakordliiere Nordchlies zwischen 21 und 2330 S.- Inaug.-Diss.
Fachberelch Geowlss., 178 p., Frele Unlv. Berlin.

FLINT, S.S., HARTLEY, A. J., REX, D. C., GUISE, P. & TURNER, P., 1989.
Geochronoiogy of the Purllactls Formation, Northern Chile: an insight into
Late Cretaceous/Eariy Tertiary basin dynamics of the Central Andes.- Re-
vista gsoi. Chlie, 16/2, 241-246, Santlago.
-_....-..

252

MARQILLAS, R.A. & SALFITY, J.A.: (1988): Tectonic framework and corre-
lations of the Cretaceous-Eocene Salta Group, Argentina.- Lecture Notes
Earth Scl., 17, 119-136, Springer, Berlin.

RAMIREZ, C. F. & GARDEWEG, M. (1982): Hoja Toconao.- Serv. Nac. Geol. MI-
ner/a, Carta geol. Chile, n. 54, 122 p., Santlago.

REUTTER, K.-J. & SCHEUBER, E. (1989): Relatlon between tectonics and


magmatism In the Andes of northern Chile and adjacent areas between 21
and 25 S.- 50 Congr. geol. chil., 1, A 346 - A 363, Bantlago.

:;:
. .... ... ..
... . .
. . . . z:
. ,.. .
... .
ixi;
IX&
:::w:
:::cs:.

-*-t=s+wAii axes 9
x FatAt and thrust?
l OamwKeat A
mitt foldsA

m: Stratlgraphic column
and subdlvlslon of the
Purllactls Group near the
southern border of the
study area (Flg.1).

EkLl: Geologlcal and


structural sketch map of
the study area at the
western border of the
Salar de Atacama near 23
S.
253

EVOLUTION OF THE CORDILLERA DE LA SAL, NORTHERN CHILE

Eberhard Wilkes and Konrad Gorier

Institut fur Geologie, Freie Universitat Berlin, Altensteinstr. 34A, 1000


Berlin 33, W-Germany

Resumen

En la depresi6n de1 Salar de Atacama, la sedimentacibn es de abanicos a-


luviales y playas con intercalaciones piroclisticas daciticas y rio-
daciticas desde el Oligocene. El origen de la cuenca y la deformaci6n
sinsedimentaria presente en la Cordillera de la Sal se deben a una tectd-
nica de deplazamiento sinistral en fallas paralelas al rumbo de1 orbgeno.

Key Words: Andes, Salar de Atacama depression, Cordillera de la Sal, con-


tinental red beds, transpression.

Geographical and geological setting

The narrow Cordillera de la Sal (CDS) fold belt is situated in Northern


Chile (II. region) within the endorheic "Salar de Atacama depression",
which is part of the Preandean depression, a morphological low between
the Chilean Precordillera and the Western Cordillera of the Andes
(fig. 1). At the present time the Western Cordillera represents the west-
ern part of the modern magmatic arc; the position of the Preandean de-
pression is therefore an unusual one between the magmatic arc and the
forearc region. Nevertheless it is considered an intraarc basin. Its ac-
tual marginal position with respect to the modern magmatic arc is prob-
ably due to the effect of the eastwards shifting of the position of the
magmatic arc in time (REUTTER et al. 1988).

Stratigraphy

The CDS is mainly built up by red beds and evaporites of the Oligo-
/Miocene San Pedro Formation (SPF) with an exposed thickness of ~3,000 m.
.I *-- .... _ . . .--. ..-,e~.

254

-I-
I

2
;
>
.
: ;
= 5

Fig. 1 Location map and main structural units; working area hatched.
c
CONO- IAOIOME- TECTONIC EVENTS
NC DE*.: RIG DATA
OSITS

24.9: 1 M.
*a** M.

JS.S t 3 M.

Fig. 2 Stratigraphic data of the San Pedro Fm.; radiometric data


according to MARINOVIC & LAHSEN (1984) and RAMIREZ & GARDEWEG (1982).
Lithology: 1: halite; 2: mudstone; 3: siltstone; 4: sandstone; 5:
conglomerate; 6: tuff.
_.I .I .__-_.

255

The stratigraphic section presented in fig. 2 has been measured in the


CDS in the Valle de la Luna area at about 230 S. In contrast at 230 30' S
seismic investigation (TOWNSEND, 1988) indicate only 450 m.
The stratigraphic record in the CDS and its surroundings is characterized
by rapid lateral changes of thickness and lithology. For this reason and
owing to a complicated deformational style a reliable lithostratigraphic
correlation between the different portions (see FLINT, 1985) of the about
150 km long outcrop area of the SPF seems to be impossible.
Within the SPF frequent intraformational breccias, slumping horizons,
spectacular increases of thickness over short distances and angular un-
conformities indicate synsedimentary tectonics. At Oligocene time an in-
tensive dacitic to rhyodacitic volcanism starts. Towards the west the
evaporites of the SPF are interfingering with conglomerates and sand-
stones of the Tambores Fm.
The San Bartolo Group (0 - ~100 m), mostly unconformable, on top of both
San Pedro- and Tambores Fm., is built up by a sequence of Upper Miocene
ignimbrites with intercalated conglomerates and sandstones.
The Vilama Fm. (0 - 80 m) of a Late Miocene - Pleistocene age, mostly
concordant on top of the San Bartolo Group, contains conglomerates, sand-
stones and pyroclastics.

Facies Interpretation

The sediments of the SPF, San Bartolo Group and Vilama Fm. were deposited
under similar arid conditions within alluvial fan/plays systems, strongly
influenced by volcanic activity. Owing to the very low bromine content in
the halites a marine origin of the salts can be excluded.

Tectonics

The most evident tectonic elements are:


- NNE/SSW trending folds with doubly plunging axes. The fold axes display
a sinoidal bending, locally evolving towards an "en echelon" grouping.
SE-vergency prevails, but there is also a tendency towards divergency,
and recumbent folds occur at both margins of the CDS.
- A very complex pattern of near-vertical small-scale faults (normal, re-
verse and strike slip, mostly sinistral)
- NNE/SSW trending low-angle reverse faults, mostly with an ESE thrust
direction, but also the opposite heading is present. Such low-angle re-
verse faults, sometimes with overthrusts of more than 100 m on top of
loose gravels are common outside the CDS near the margins of the Salar de
Atacama depression (HOOPER & FLINT, 1986). They evolve over short dis-
tances from monoclinal folds.
- Vertical tension fissures, well preserved especially in the
conglomeratic Tambores Fm. (N/S) and the Ignimbrites of the San Bartolo
Group (NW/SE).
There is good evidence for repeated tectonic syndepositional movements
from Oligocene/Miocene time up to the present.
256

Palaeogeographicand geodynamic interpretation

The Salar de Atacama depression has existed at least since Late Oligocene
time as an endorheic evaporative basin. It has been fed mainly by allu-
vial fans deriving first from W to NW and for the uppermost portions of
the SPF from N, whereas a transport from E has not been observed. This
trend continues to Plio-/Pleistocene. As a consequence, in this region a
very young uplift of the Western Cordillera to its present high seems
most probable.
The origin and the observed sedimentary and tectonic pattern of the Salar
de Atacama depression fit best with the assumption of sinistral strike
slip movements combined with compression (HOOPER & FLINT, 1986; FORSYTHE
& TOSKOS, 1988; WILKES & GI)RLER, 1988). The decreasing tendency of total
thickness, age and intensity of deformation of sediments from N towards S
could be a hint to the development of a pull apart basin. A
transpressional regime is documented in the "flower structure" of the
CDS.

Literaturecited:

FLINT, S. (1985): Alluvial fan and Playa sedimentation in an Andean arid


closed basin: the Pacencia Group, Antofagasta Province, Chile.-
J. Geol. Sot. London, 142: 533-546, London.
FORSYTHE, R.D. & TOSKOS, T. (1988): Late Cenozoic tectonics along the
southwestern border of the Altiplano/Puna, Salar de Atacama
Depression, Antofagasta Region, Chile.- VII. Congr. Lati-
namericano de Geologia Abstracts: p 315, Belem.
HOOPER, B. & FLINT, S. (1987): Miocene-Recent Tectonic Evolution of the
San Bartolo Area, Northern Chile.- Zbl. Geol. Palaont. Teil I,
L&: 967-981, Stuttgart.
MARINOVIC, N. & LAHSEN, A. (1984): Hoja Calama, Regi6n de Antofagasta.-
Serv. Nac. Geol. Min.: Carta geol. Chile No. 58, pp. 144,
Santiago de Chile.
RAMIREZ, C.F. & GARDEWEG, M.P. (1982): Hoja Toconao, Regi6n de
Antofagasta.- Serv. Nac. Geol. Min.: Carta geol. Chile No. 54,
pp. 121, Santiago de Chile.
REUTTER, GI)TZE H -J SCHEUBER, E., SCHWAB, K.,
~"C&~RZ,~~GES~'~,
'SWIGGER, 'P. '(1988): Structures and crustal
development of the Central Andes between 210 and 250 S.- in
Bahlburg, H., Breitkreuz, Ch. & Giese, P. (eds.): "The Southern
Central Andes": Lecture Notes in Earth Sciences, u: 231-261,
Berlin.
TOWNSEND, F. (1988): Exploration petrolera en la cuenca de1 Salar de
Atacama, Region de Antofagasta, Chile.- Vertiente, 4: 45-55,

WILKES, E. & GGRLER, K. (1988): Sedimentary and Structural Evolution of


the Cordillera de la Sal, II. Region, Chile.- V. Congr. Geol.
Chileno, 1: pp A173-A188, Santiago de Chile.
257

SD-Y-DE sEDwmaoIy EN LA cum NEsozDlcATRAsARco


DELNDmEDEDHlLE

AxeI van Hillebmndt PeberPtinz, Haru-Gehvd Wik

Instht filr Geobgb und PaknaDloaia, Technbche UniversitiitBerlin, Sekr. EB 10, Emrbfleuter-
PIatzl,D-1OOOEerlin10

LaImfeaugaoi6nrderandk5eneinuuwdel~
grama da InvastIgadh alamAn MoMidad da mar
ganer actho8 d6iaantimntas: h ha Q erb tfa-
bajoes d esbrdk hmshfo dr cad bdaa losa&a-
mientos masozoiax marlnor (lig. 1) desde el punt0
6 vista tk4ratIgfUko en ll ha hvdgado
(GR6SCHKE et al. 1988). h lvduc46n ~aleogec+
gw dal ha purdr 8ar nrumkia dol sigubnta
modo: la 8echantaMn nwina mesozoica 18 inicia
lnerbcrrncaenaIN6rieo,taloomooarm~P~
nSy CdomMa (QNER lsS0). h formad& ti esh
cuancara@icaatn~depmce808~dlata-
d6nnhchadosaladirgnga~cbPangaa.Ak
latgo&lJurMcohtmtug46npmgmuhadaeI
EmbraUAdakrUmf@s6laadualPreaxdilbm
chknqkgandoasuMansi6nm6xhnadvanlleI
Oxkxdho (fig. 2a). En ll Jwhico rupahr @or&
oxfadho) deawa la innurnch marina deposItark
dou pott3naU8erbr Isrmrb88.
FhJ.1: tlbkaddn dd ha inve6tigada
.-... ,. ,_..^l. ._

258

i I

V
V

V
VV &?.l...

c,

11

t,

-L
Fii 3: Moddo pJeogw@f~co m al JurAaica aupatior (Oxfcdi~o)

El volcanismoasodado a la subducdon eat& en nlad6n oon la formad6n da la cuenca URISarw.


Loa dkmnbes gmcbs de aedtmentad6n
.._-- _ ._ ..,

261

SEDIMENTOLOGY OF THE CERCAPUQUIO AND CHAUCHA FORMATION8


(CENTRAL-PERU)

Silvia Rosaa & Lluis Fontbot4

* Mineralogisch-Petrographischea Institut der Universitat. Im


Neuenheimer Feld 236, Postfach 104040, D6900 Heidelberg.

Sedimentologic investigations on the Cercapusuio and


Chaucha Formations in the Azulcocha area (central Peru,
1235s. 754OW. Fig. 11 are presented. Both the Cercapusuio
and the Chaucha Format ions are characterized by
sedimentological cyclicity (Fig. 2 and 31. It may be related
to the first uplift phases of the Marafion Geanticline.
The Cercrpuquio Forma-
tion consists of braided ri-
ver deposits. It is subdivi-
ded into four cycles or se-
quences (Fig. 21. Within each
sequence a granulometry de-
crease from bottom to top is
observed, whereas the forma-
tion as a whole shows a gra-
nulometry increase to the
top. Sorting and skewness va-
rig. 1. Location of the lues indicate fluviatile
Azul cocha area transport. This is consistent
with facies analysis.
The mineralogical analysis of the Cercapusuio sandstones
reveals a fairly mature composition. The Brazilian Craton
appears to be the most probable source region. This indicates
that during deposition communication between the basin and
the Brazilian Craton existed.
The Chaucha Formation was deposited in a peritidal
carbonate environment. Three upwards shallowing cycles are
distinguished (Fig. 31. The first one comprises intertidal
and supratidal sediments, with characteristics of Sabkha-
type deposition (including formation of early diagenetic
dolomite and evaporitic minerals) and with are overlayed by a
thick red siltstone. The second cycle evolves from subtidal
to supratidal environment and ends with red siltstone too.
The third cycle is subdivided into three subcyc les
characterized each of them by evolution from subtidal (with
barrier f aciesl to supratidal environments. The cycle
finishes also with a thick red siltstone layer. The top of
._._.._I, .._ . .._.

262

I. ordc arder
. d

Fig. 2. Facieg analysis of the Cwcapuqulo Formatton


Cl .ClsYstone G..GranulometrY -- .Clsy gal I
RC !31I t.atona
__._____--.-_ -4.Erosion eurfsce h .Lsaf remain
Gl.Very fins sandstone H.Horizontal bedding P .Trunk remain
GZ.Fine sandstone d.Through cross-bedding c .con1
Gt.Medium sandstone *.Ripple marks 7 .Bioturbation
SS.Coarse sandstone d.Flaser bedding ~7 .Pyrits
tlc.Microconglomerate )* .Overturn
..- . e--,^

263

. 3. Fecims lnalysim ot thm Chauchr Formation


. Limestone .Wackestone m..Wavy lamination
.Sandy limestone :** . Packstone (calcrete)
.Dolomitlc limestone G:: *Grainstone e..Algal lamination
.Sandy dol. limestone SbT.Subtidal 0 ..Pseudom. after ha-
.Dol. brecc. lime- It. .Intertidal lite
stone SpT.Supratidal 0~ .Pseudom. after an-
.Calcareous dolomite SB..Submarine barrier hvdrite or gypsum
.Sandr talc. dolomite L . . .Lagoon t . .PelecyPod
.Calc. brecc. dolo- ~..Erosion surface B ..GastroPod
mite -..Horizontal bedding ;r ..Bioturbation
m .Red si ltetone (Terra Y*..Through cross-bed- 0 . .Ooid
-rossa?) ding 0 . .Pellet
n . ..Hudstone Y..Oblique bedding
264

each cycle is characterized by progressive dolomitization.


occurrence of calcrete layers, detritic dolomite. and/or
reworked carbonate sediments (Fig. 31. All these features
indicate that the cycles culminate in emersion. The red
siltstones are easily wheathered and poorly exposed. making
it difficult to determine their deposition environment.
CEDILLO (1988) investigated similar facies 45 km to the
southeast in the Cercapuquio region and interpreted them as
terra-rossa . This interpretation is compatible with the
emersion sequences recognized below each siltstone layer and
with the ocurrence of two small channels filled with micrite
and elastic material within the red siltstones at the end of
the second cycle, and which are interpreted as supratidal
channe 1s.
Lithogeochemical investigations on the carbonate rocks
of the Chaucha Formation indicate that trace e 1ement
contents are normal for carbonate rocks (Tab. 11. Sr
displays, as expected, lower values in dolomitic layers (loo-
160 ppml than in limestones (UP to 300 ppm) . Mn correlates
well with dolomitization (ROSAS, 19891 . Zn (14 to 130 Pam)
and Pb (8 to 30 ppm) values are also normal except in a
sample of the basal dolomite (Zn - SO0 ppm) located close to
a barite-bearing layer (BaSO+- 54000 ppm).
TAJ.1. G2ocHEtI13TRY Or THI cxAucHA rnnnATIoN

HNcc3 ?D w _..
slnl
~f~l lipI fN1
2084 10 121 99.33
632 12 20
I92 12 27 ::.R
y: 1: 1:; 9a:rc

969 13 II ::: :t
lS80 I1 21 99. lb
lOO.bS
:f.it
91:3
100.2 #
w
91:sr
:P: t: 1: ::A. ::
ssa 13 16 mph:
$259 19 IS
4886 22 b2 g:::
3b64 13 14
4993 13 21 99109
lb19 16 1s 1;). ;g
212b 21 15
3698 20 100 91:os
9613 30 39
2910 12 18 ::+
9428 I1 20 19: b2

The ape of the Cercapuquio and Chaucha formations is


still controversial. They lie between the Condors i nga
Formation (UP to Toarcianl and the Goyllarisquizga Group
(Neocomian) . In the absence of explicit paleontological
f indinss a Bajocian age is favored on the basis of
lithological correlation with carbonate platform sediments of
Bajocian age in the Huancavelica region (Chunumayo Formation)
and in the Arequipa region (Socosani Formation). Bajocian
sediments are observed also in other regions whereas in the
time span between Bajocian and TitoniaWBerriasian. no
carbonate sedimentation is known in Peru.

CEDILLO,E. (1988) Dissertation Univ. Heidelberq, 201 p.


RIMS, S. (19891Diplorarbeit Univ. Heidelberg, 126 p.
265

THE KIMMERIDGIAN (?I - EARLY VALANGINIAN TECTONIC EVENTS ON THE


PERUVIAN MARGIN

* . Victor CARLOTTO (21, Javier JACAY (3) and


ES~I~"JAP~~D'~~~.
1 Univ. Nac. San Agustin, casilla 1203, Arequipa, Peru. _
3 i;rn;. Nac. San Antonio ,departgmento geologia
II Franc&s de Estudios Andinos, casilla 18-?!!??'L?~?i8,
Peru. *
(4) ORSTOM, et Institut Dolomieu, 15 rue Maurice-Gignoux, 38031
Grenoble Cedex, France.

el Jur6sico m&s

Atl8ntico en
Bstalatitud.

Introduction.
the late Jurassic "Nevadan oro eny" has been invo-
ked zi p:~~lfi~;e$~; unconformities observed % elow the Neocomian
sandstones. the importance and the age of these eve;;:
are still This paper is an attempt to precise
chronology significance of the Kimmeridgian to earl
;F;;s%;nian events, through stratigraphic and sedimentologica 1

Geological setting.
During middle Jurassic times, the Peruvian margin underwent
widespread tectonic events that resulted in.a contrasted paleo-
%~;%f;: :ic pattern (Jaillard et al. 19901.. Since this time
P to distinguish fromA 7Z E. (1) a narrow coastal '$oZ
which is badly known as far; (2) a Western area characterized b
marine sedimentation ( resently Western Cordiller8); (3) an Axja y
Cordillera which acte % as a positive swell (Mar-anon eanticline
SW Altiplano) and (4) an Eastern domain' whit % received
$i?nly continental deposits (Oriente and NE Altiplano).

Geological evolution.
1. From Kimmeridqian (7) to early Tithonian.
Kimperidgian deposits have not been recognised in northern
PestI.
southwestern they are re resented b _shalow '
istones and shaleEei%&bra Fm Vicen e e et al. ~~82) and ~~r~~~
part of the eastern_oiiZ; by alluvial
: Huambutio Fms, Klinck --
et al. 1986, Carlot-
often capped by marine, ,partli
thonian ag;g8aGramadal,
ie et al. Batty & !i?ifard
robably.correlate i?itri-the Simbal Fm of northern
Jacav 1989). Thev iare coeval with a major eusta-
ti$h~~a~l.evgl rise (Iiiqet &. 1 987).
eriod is marked-1Sv nrim&-ous-synsedimentary tectonic fea-
tures w!*ich exprcass a NW-SE-trendin extensional re irbe in the
Axial Cordillera ([Batty 6iJaillard 1889, Carlotto 19%9) .
2. From middle to latest Tithonian.
After Tithonian times lar e areas of southwestern
Peru emerge~a~~~tt & Jaillard 1989) %ut marine
shales of late Ti-41honian to Berriasian age are '1oZEByY R!Z
(Bellido 195 Geyer 1983).
al Cordillera, undated, continental to tidal fine-
In the Ax f#
266

calcareous beds

of late Tithonian times,


was created (Punta Morenz

3. Berrlasian times.
on of the undated tidal to
Iiuambutio lower
that they only
he Berriasian.
active up to
the latest Berriasian (Wiedmann 1981). It is coeval with the
deposition of widespread marine (Bellido 1956) to deltaic sandy
black shales (0 on Fm, Wilson 1963).
In northern $ eru, shallow marine sandstones abruptly overlie
Tithonian to Liassic deposits (Tinajones Fm, Cobbing et al. 1981)
and are associated with extensional tectonic features7JZillard &
Jacay 1989).
4. The latest Berriasian-earliest Valanainian disconformity.
Near the Berriasian-Valan inian boundary, the contrasted
paleogeogra hit pattern is over 9 ain by wide8 read, East-der$;;g,
clean sands on 8 (Chimu Goyllarisquizga, SaP to de1 Fraile -
huani & Iian!ant Fms) The sands disconformably overlie ro&ks
Paleozoic to Berriasian age, Paleocurrent measurements indicateOd
westward transport in the eastern areas, and a SE-ward distribu-
tion along the western areas. Depositional environments vary from
fluvial to tidal.
These Valanginian sandstones conseal the late Jurassic tecto-
nits and level the inherited contrasted paleotopography.

Conclusions.
Between KimFerid ian (?I and Valanginian times the Peruvian
margin recorded impor ? ant sedimentary discontinuities (Jaillard &
Sem ere 19891, among which two are major tectonic events (Middle
Tit Ronian and Berriasian-Valanginian).
The Kimmerid ian (? event main1 concerns the southern
% een re 1ated to the K raucan phase of Chile (8zfb$b
?il a?$8!asBatt h Jaillard 1989)
'Liz' middle Titx onian tectonic event affects,the whole margin.
ihv;Ige$;er%Zeted, as resultinq from. a drastic change inwhi;%
rection of the Pa eopacific oceanic
would have triggered the collision of allochtonousP$~E~~nes in
northern Peru (and farther north) and an extensional tectonic
regime in the rest of the Peruvian'margin (Jaillard bl Jacay 1989,
Jaillard et al. 1990).
The di55oRinuit of the Tithonian-Berriasian boundary
main1 concerns nor Yhern Peru and can be related to the COW
siona 1 processes (Jaillard L Jacay 1989)
These three events main1 affect the western part of the Peru-
vian and are rela f ed to geodynamic events occurring in
the weitz%?!8Paleopacific plate.
In contrast, the latest Berriasian-earliestValanginian tefn;
is marked by the creation of a wide easterly source-area
rovided the east-derived sands. It can be ln;PrtE;edats &
fir t manifestation of the southern Atlantic
latftude.

The late Jurassic to Valanginian sedimentation in the


PHL margin.
Lithology: evaporites; limestones.
sandstones: 5:'ionglomerates; 6*2bartl volcanosedime$$!~sdep~~
sits. 7: mainly volcanic flows. 8: olis X olites.
Sediment;ry:~;tz;;onmentst~~x~t column): A: alluvial
fluvial; : E: lagoon: F: shallow sizfi; gi
~T~~~LEP~~i~~~e~~~biSi":.l. farly middle late: 0: Qxfordian;
; T: Tithonianj B: Beiriasiani V: Valanginian.
3LAOROYA 7 cuzco 1nPuNo

5 HUAYTARA
268

References.

-Maqmatism Zt a Plate ed e
and IIalSti5aPrmeW * the
0rRK

:a-dei cijrsc:medio de1 rio Huaytara,

_..I.. .

yer,_ olF,.. 1983.. Obgrt$t~onische Ammoniten-Fauna van Peru.


lronclggy_gf fluctuating sea levels

_ _. 30. Geodynamic evolution of the Northern


urdes
-__q dwtng . early
_-A__ to
A- middle
----- Mesozoic times: a Tethy-

LGJ IIIIL.=s4A~TL AI-L. , y.

EFatigrarla ae laosta de Lima.

-__ -_-.
Sam Bre T. e 1988. Evoluci6n tectosedimentaf$; de,~~l~;;a
duran ee ei Cre &k 5to Cons. geol. Chileno, , - ,
t- . ..

269

MESOZOIC EXTENSION AND CRUSTAL THICKENING


IN THE PERUVIAN ANDES.

Etienne Jaillard. IFEA (Lima, Paris) & ORSTOM, UR 1H (Paris);


present address: Laboratoire de eologie alpine,
Institut Dolomieu w aurice-Gignoux
38031 &;eizsle Cedex, France:

Abat act: The subsidence history of the


f shows the western areas to be floore
:K% overlginsuk thick sedimentary wedges.
shortening x structures can ex lain a significant part of
the present-day observed crustal thicR ening.

Introduction.
Most of the studies carried out in the Central Andes were
concerned with the Cretaceous-Tertiar with the Andean
or with the present-day processes. Iiowever,
1s known so far about evolution of
margin of the Central Andes The scope of this nk:t
is to analyse the behaviour of the Peruvian active margin cId;iF%
the pre-erogenic period (i.e. Jurassic and Cretaceous),
examine its effects during the Tertiary orogeny.

Geological setting.
Since the latest Jurassic, the Peruvian mar in was subjec;Tit;o
the. nearly continuous subduction of the Fara9 lon oceanic
Eg;;:%lthis period, the Peruvian margin can be divided into: l),a
zone which is bad1 known and is not discussed in
2) a western basin w5 ich received a thick eastward-tk?i?
!?:griedge of main1 marine de osits durin the Cretaceous 3) an
axial positive Cor 5 illera wit R condensed w esozoic sedimentation;
and 4) an eastern area, in-filled b an eastward-thinning wedge
of marine to continental Cretaceous 3 eposlts.

Method and limits.


This study is based on several subsidence curves obtained
T:on%_7:woS) large-scale palaeogeographic profiles,
and southern Peru (= 14-17'
northern
respe;!ivel The
curves have been computed using method and parameters of 8. clater
and Christie (1980). However some uncertainties arise from local
unaccurate stratigraphic a es, and from uncom lete sections (ba-
sal decollements and/or su% sequent erosions). R oreover, published
;glzewhich have been used for some sections are often estimated.
, the presented curves must be considered as approximated.

Subsidence patterns.
The subsidence curves exhibit three patterns according to
their palaeogeographic setting.
~~~~s~~~gE~~;:'e~~8~~easf the curve= show an upward roncavii;
ests an extensionjzhermal-controlled subsiden-
Tectonic subsidence can reach as much
latest,Jurassic-Cretaceous.times. In northern Peru,
it $lowed t e deposition of a thick sed!mentary wedge that reach
toward the West. and can be consldered pro ;~sslvsl
thinning eastwards, so that it represents 1 kma:ear P axis 1
270

Cordillera. In sout em Peru, the same pattern is observed, with


a 4 to 1 km-thick setlimentary wedge.
- The axial positive swell recorded a linear tectonic subsiden-
ce curve which does not exceed 500 m. Thickness of the Kimmerid-
ian-Cretaceous sedimentary cover varies from 1 to 0,2 km. Though
is palaegeographic feature is clearly inhgrited.from the com-
earlier evo ution of.the Peruvian margin (Jaillard & al.
its tectonic si ;;f;cance remains unclear.
the eastern % the subsidence curves also show
linear sha e which indicates continuous flexion of th;
substrate tE rough time, whithoutaevidences of significant exten-
sion. They are interpreted as re~~~~~; from the westward tilt'ng
of the South American craton 8 to the oceanic subduct it
on
Mitrovica c a. 1989). Tectonic subsidence is less than 1200 m
or the considered period, and the sediment accumulation results
in an eastward tappering wedge that reaches 3 km toward the West.

Palaegeographic control of the Andean thrust sheets.

Mesozoic crustal thinning, sediment accumulation and preeent-


day crustal structure.

In Southern Peru, gravity anomalies indicate a veryt$pogk$;E


thickening beneath the western area (up to = 65 km),
Cordillera and beneath the western part of the eastern domain (=
55 km)(Fukao et al. 1989).
271

In the western area,of southern Peru, structural,data are too


~~;~~~stoc~;;~~dan estimation of the average shortening. However,
on in the eastern areas of Bolivia indicate a
very important shortening (e. Baby et al. 1989). Assuming a
shortening of 50 % in the wes 8.
ern zone, the sedimentary
would become 8 to 2 Km-thick from W to E, thus representingw?2gg
of the observed crustal thickness (i.e. 25 % of the observed
thickening).
Hence, thou h the density of the deformed sedimentary wedge
should be 51ig % tly different, neither the crustal extension
the associated sedimentary accumulation can be ignored when dt:E
ling with the Andean crustal thickening.

Conclusions.
history of the Andean mar in shows that
the crusta and the associated sedimen e accumulation
art in the structural style during the
he observed thickening of the continental
the inversion of the crustal extens;gaal
and the related shortening of the superimposed
wedge can ex lain the whole observed crustal thickness,
:iZ:::: in southern B eru, it acted as a minor but no negligible
parameter.

References.
I _ _- __... ,_... .- ^. . .

272

SUBSIDENCE HISTORY,CRUSTALTHINNING AN0 ANDEAN


SHORTENING IN THE NORTH PERUVIAN MARGIN
A. TECTONIC SUBSIDENCE CURVES
8. RECONSTRUCTION OF THE MARGIN BEFORE ANDEAN SHORTENING
C. PRESENT-DAY STRUCTURE (crustal thickness after Fukao et ai. 1989)

0A [earC.lnldCrllattc
im
cr laaiqr.
TO

I
; COASTAL ZONE WESTERN BASIN EASTERN BASIN
/z\! x \ /
J3Ji /
I
.-_._,_-.-

273

LAS CUENCAS CONTINENTALES DE LOS ANDES CENTRALES. RELACIONES


CON LA EVOLUCION GBODINAMICA ANDINA.

por Rend MAROCCO * y Jean DELFAUD **

* ORSTOM, 213 Rue Lafayette, 75480 PARIS CEDEX 10 (France);


** Unlversitd de Pau et des Pays de l'Adour, Avenue de
l'unlverslt~, 64000 PAU (France).

ABSTRACT
Dans les Andes Centrales, ?Ipartir du debut de la tecto-
nique andine (Santonien), se constituent une serie de bassins
contlnentaux dont la dynamique est dtroltement lihe a
l'dvolutlon tectonique et magmatlque CrBtach-CenozoYque des
Andes.
KEY WORDS : Continental sedimentation, synsedimentary
tectonics, Central Andes, Intramontane basins, Fore-land
basins.

INTRODUCCION

En el Santoniano, bajo el efecto de la fase peruana,


primera manifestacidn importante de la tecthica andlna, 10s
Andes Centrales lnician su emersih. A partir de este evento,
el regimen tecthico pasa de distensivo a compresivo. En las
zonas superflciales de la corteza, dicha compresi6n se
traduce por un continuum tectbnico que produce, segun las
condiclones locales : fracturaci6n, deformacl6n ddctil o
formaci6n de depresiones actlvas -cuencas sedimentarias- con
relleno continental.
De1 CretkiCO superior a fines de1 Cenoz6ico, la
morfologia de la zona andina evoluciona bajo el efecto de1
continuum tect6nico y de 10s saltos de1 arco magmatlco we
pasa de la actual zona costanera (Cretacico sup. ) a la
Cordlllera Occidental. La deformaci6n comienza al santoniano
en la parte oeste; progreslvamente alcanza las zonas m&s
orientales. Un period0 de calma parece existlr durante el
Oligocene inferior.
Las cuencas continentales formadas se califican en
funcidn de sus posiciones respect0 al arco magmatico. Asi,
existen cuencas de ante-arc0 y de tras-arco. Dentro de las
cuencas de tras-arco convlene distinguir: a)las cuencas de
ante-pays ublcadas al frente de las unidades tect6nicas we
cabalgan su ante-pais oriental; b)las cuencas intramontanas,
ubicadas en el seno de la cordillera en curso de formaci6n,
controladas por fallas
274

LOS saltos de1 arco magmdtico y la progresi6n hacia el


este de la deformacidn determina una modificaci6n de la posi-
ci6n de 10s diferentes tipos de cuencas segiln el period0
considerado.

LAS CUENCAS DE ANTE-ARC0

Son cuencas de poca actividad tect6nica; se ubican en el


borde occidental de1 arco magmitico. El relleno, fluvial y
lpcustre, estd alimentado por la Cordillera Occidental en via
de edificacibn. Solo se conocen las cuencas de ante-arc0
posteriores al Eocene. La m&s importante (la m&s subsidente)
es la Cuenca Moquegua de1 sur de1 Peril y norte de Chile. Es
un medio-graben con relleno fluvial lacustre
granocreciente, limitado al este por una cola de falla
dextral-inversa. El funcionamiento de la cuenca comienza al
Oligocene inf. (ta1 vez en el Eocene). La actividad tect6nica
y volcknica sinsedimentaria se hate nctar a partir de1
Oligocene sup. (28 ma parox.).
La Cuenca Moquegua, practicamente al nivel de1 mar, fue
invadida por una transgresibn marina al Oligocene sup. (26-27
ma aprox.). Hacia el norte el ante-arc0 estd ocupado por
cuencas marinas litorales, de poca subsidencia en el Peril
Central, m&s activa hacia el norte.
De1 Miocene al Actual, una potente serie volc6nica se
acumula en el ante-arco; las intercalaciones sedimentarias
(fluvial) son siempre de poco espesor.

LAS CUENCAS INTRAMONTANAS

A menudo, las cuencas continentales antiguas ubicadas en


la Cordillera de 10s Andes Centrales, fueron calificadas coma
"intramontanas". Esta calificacidn, coherente con un modelo
andino de poco acortamiento, no vale en la actualidad. Ahora
sabemos que 10s Andes han sufrido un acortamiento importante
debido al juego de grandes superficies de despegue; la
deformaci6n y 10s relieves creados han progresado de1 oeste
hacia el este. De ta1 modo que gr&n parte de las cuencas
continentales andinas son cuencas de ante-pais formadas al
frente de las masas cabalgantes y que migraron hacia el este
con las unidades tectbnicas.
Las cuencas intramontanas S.S. se reparten a lo largo de
10s Andes, por ejemplo: Cuenca Koniri (Bolivia); Cuencas de
Rumichaca, Cajabamba, Namora (Perb); Cuencas de Vilcabamba,
Nab6n, Cuenca, Riobamba, etc. (Ecuador). Todas estas cuencas
se formaron entre fines de1 Oligocene y el Pliocene sup.; son
todas muy subsidentes y el volcanismo es muy importante. La
sedimentacibn, siempre controlada por la tectbnica, se
caracteriza por series granocrecientes. La sedimentacidn
varia mucho de una a ctra regi6n debido a las condiciones
locales (clima, topografia, etc.). En el norte (Ecuador), las
cuencas son lacustres profundas con megaturbiditas pasando
hacia arriba a cones aluviales que marcan la "clausura" de la
cuenca. En el sur (Bolivia), el relleno de las cuencas,
fluvial, pasa de distal a proximal.
275

Todas las cuencas intramontanas estudiadas evolucionaron


en regimen compresivo. Estuvieron controladas por SL~S bordes
fallados. La genesis de estas cuencas esta ligada al juego de
fallas al techo de unidades tectonicas en movimiento hacia el
este.

LAS CUENCAS DE ANTE-PAIS

Se formaron al frente de las unidades tectonicas cabal-


gantes que se movilizaron primer0 al Santoniano (fase
peruana) en las partes occidentales de1 sistema andino, y
despues a partir de1 Oligocene superior en las zonas m&s
orientales. son medio-grabens limitados al oeste por las
fallas inversas de1 frente cabalgante; al este 10s sedimentos
descansan sobre el ante-pais no todavia deformado.
Las ma.5 antiguas cuencas de ante-pais reconocidas son
las cuencas de Capas Rojas de1 cretacico terminal (Bclivia,
Peru). La m&s reciente es la Cuenca Subandina. La
granocrecencia de las series rellenando estas cuencas y la
importancia de la tect6nica compresiva sinsedimentaria,
traducen la influencia creciente sobre la sedimentation de1
avance de las unidades cabalgantes.

CONCLUSIONES
Las cuencas continentales andinas estdn ligadas muy
estrechamente a la evolution de 10s Andes desde el Cretacico
superior. La din&mica de1 relleno refleja el desarollo de 10s
eventos tectono-magmdticos que se produjeron en las zonas de
aportes y en la misma cuenca.
HAS alla de1 inter& estrictamente sedimentologico
(establecimiento de nuevos modelos sedimentarios) o aplicado
(evaluation de 10s recursos minerales contenidos), el estudio
de las cuencas continentales andinas es un complemente
indispensable para la comprensidn de la genesis de la
estructura de 10s Andes Centrales.
277

LES BASSINS TERTIAIRES DE LALTIPLANO SONT-ILS DES BASSINS


FLEXURAUX INTRACHAINE 7

par Claude MARTINEZ l et Michel SEGURET l*.

in&W Fran@ de Recherche pour le DBveloppement en Coophtion (ORSTOM),


BP 5045, Montpettier cedex 1 .
a Laboratoirede Ghtogie des Baeeine, ptace Eughe Bataillon,
34060 Montpellier cedex 2 .

Abetreet:The Altlpisno of northern 9ollvfr, that extend between the Wootern and the Eastern Cordilleras. is a
StrOngly whetdent basin. its geneatr Is sccomprnled by quasi-continuous compressive tectonks during Eocene to
PiiocOnO. PrOSWtly. no model WpYn8 In I statidactory manner this Alptano basin .

Entre les Cordilleres orientale et occidentale des Andes centrales, le Tertiaire de IAitipiano
nord-bolivien correspond P une sedimentation continentate dage eocene a Pliocene dont iepaisseur
cumulee depasse 20 km.
Pendant longtemps, et jusqua recemment, (Lavenu et Marocco, 1964), le bassin de IAltiplano
Btait interpret6 comme resultant dune succession de phases de distension, entrainant subsidence,
sedimentation et volcanisme, separees par des episodes de compression (phases tectoniques andines a :
40; 30; 7 et 2.5 Ma).

Reprenant ietude de la partie aeptentrionale de IAitiplano bolivien (entre 1630 et 18.30 de


latitude sud), nous avons effectue une cartographic preliminaire de deux secteurs: un secteur sit&
immediatement au Sud du hc Titicaca, occupe par des terrains aliant du C&ace terminal au Miocene
moyen, et un secteur plus meridional sur ie ftanc occidental du grand synciinal du centre de IAltipiano,
impiiquant des terrains dage miocene moyen a Pliocene.

SECTEUR SUD DU lAC TlTlCACA

Les dep6ts marins inter B supratidaux du Campano-Maastrichtien sont surmontes par une serie
continentale axorque de 5WC m depaisseur . Elle comprend deux groupes de formations:
Le premier groupe (~wanaku) comprend de la base au sommet:
la) Des gres grossiers, massifs. blancs. deposes dans un environnement fluviatile en tresse.
1b) Des pelites rouges, fiuviatiles de plaine dinondation et des gres moyens fluviatiles (de
type indetermine).
lc) Des gres grossiers, massifs, blancs, a stratification en auge et oblique plane de depdts
fluviatiles en tresse.
Id) Des pelites rouges et grises incfuant des barres greseuses de meandres.
Le deuxieme groupe (Coniri), de 2000 m depaisseur, comprend differentes formations (2a,
2b, 2c, 2d) conglomeratiques de &es ailuviaux dont la base est g&ewe chez les inferieures. Les
elements proviennent du Paleozoique (quartsites, g&s, caicaires) et. plus rarement. du Precambrien (
granites). Les conglomerate, bien developpes le long de la faiiie de Coniri sur la bordure orientale.
passent P des facies plus fins au centre du bassin.
Plus au Sud, dans la region de Corocoro, ie centre du bassin est occupe par des peiites, gres et
conglomerats dun troisieme groupe dont Iepaisseur es1 evaluee a 6900 m (Lavenu et Marocco. 1964).

LYge de ces series est probiematique.. Les groupes 1 et 2 pourraient se situer, respectivement,
dans ie Paleocene - Eocene superieur et dans I Eocene superieur - Miocene inferieur. Par continuite
physique avec des series datees plus meridionales, le groupe 3 est attribue au Miocene moyen.

Aucun indice de distension na 618 decele dans ce secteur septentrionai. Les nombreuses
discordances anguiaires entre formations et ies discordances progressives internes aux formations
montrent, au contraire, que la structuration en plis, faiiles et chevauchements est contemporaine de la
Mimentation. Aucune discordance nette na 818 mise en evidence dans le groupe 1. Par-centre. les
differentes formations 2a, 2b, 2c et 2d sont, dune part, discordantes sur la precedente et, dautre
part, affectees de nombreuses discordances intra-formationnelles plus ou moins progressives. La failte
de Coniri qui &pare. a IEst. le bassin du socle paieozoique es1 cachetee successivement par 2b, 2c, 2d.
_- --,..

279

A Corocoro, le Miocene du groupe 3 est discordant sur des couches du groupe 2 renversees. AU front du
chevauchement de Corocoro, ce Miocene a et6 plisse au fur et B mesure de son depot, si bien que des
couches de plus en plus Blevees du Miocene repose en onlap sur le groupe 2.

SECTEUR DE cALnPLAN0 CENTRAL

II correspond a un vaste synclinal de 200 km de long, daxe NNW-SSE. a coeur pliocene affectant
une serie miocene de 10 km de puissance (Meyer et Murillo, 1961). dont la base nest pas connue.

Sur le flanc W, les premiers depdts obsewables sont des gres fins (2). rouges, de systeme en
tresse. surmontes de conglomerats (3a). Ces derniers remanient des granites, pegmatites et gneiss
PrOVenant dun SOCle precambrien que nous situons sous les roches volcaniques de la Cordillere
occidentale. Cette serie, qut passe vers le haut a des Ores rouges et verts (3b). conglomeratiques.
contenant des niveaux volcaniques (basaltes. dacites, cinerites, ignimbrites) semble lequivalent des
ores grossiers a lentilles conglomeratiques. dage miocene moyen, du flanc oriental. Comme eux, elle
eSt surmontee par des gres fins (3c), des lutites et des argiles rouges a intercalations de gypse et de
cin6rttes puis par des sables grouien (3d) et des conglom6rats a galets de volcanisme. ddge miocene
superieur.
Les series des deux ftancs ont des Bpaisseurs a peu pres identiques (10 km) mais dans le flanc
occidental des discordances internes sont observees: les conglomerats et les intercalations volcano-
sedimentaires ou volcaniques y sent beaucoup plus nombreuses en raison de la proximite des appareils
volcaniques se mettant en Place. B cette Opaque. sur le massif precambrien de la Cordillere occidentale.
en voie de surrection.

Le flanc oriental apparait comme un vaste monoclinal de Miocene. Cependant. les discordances
progressives y sont possibles, sinon probables. en raison de celles observees plus au nord (pres de
Corocoro).

Le Pliocene (4) discordant sur le Miocene. es1 lui-meme affect6 par le plissement.

DlSCUSSlON

Cette etude preliminaire confirme que la subsidence cenoroique es1 considerable dans IAltiplano.
Aucune distension nest a lorigine de cette subsidence. Au contraire, la sedimentation est
contemporaine de deformations en compression plus ou moins continues depuis IEocene jusquau
Pliocene. Cepaisseur cumulee du Tertiaire depasse 20 km. La migration Bventuelle des depocentres au
tours du temps nest pas encore connue mais la relative symetrie du grand synclinal de IAltiplano
montre que la subsidence du Mio-Pliocene es1 au minimum de 10 km. La distension (alternant avec des
phases de compression), classiquement Bvoquee dans cette zone, ne nous paiait reposer sur aucun fait.

Les modeles disponibles de bawin ne permettent pas dinterpreter une telle subsidence dans un
tel contexte de chaine cordilleraine lice a une subduction:
Voulant eppliquer un mod&e de type MC Kenrie, on pourrait envisager que la subsidence tertiaire
correspond a une subsidence post-rift (thermique et par surcharge sedimentaire) dun rifting anterieur.
par exemple dage c&ace. Rien netaye une telle hypothese et la compression contemporaine de la
sedimentation ne serait pas expliquee dans ce mod&e.
Un modele flexural davant-chaine ne peut a levidence Btre invoque.
Reste la possibilite dun mod&e de bassin flexural intrachaine. Dans un tel modele. la subsidence
resulterait dune flexure de la lithosphere sous leffet de la surcharge creee dans Ies cordilleres
orientale et occidentale par lepaississement crustal (chevauchements et misc en place dappareils
votcaniques). La compression syn-s6dimentaire resulterait de la propagation des chevauchements. Pour
Btre argument6 et quantifie. un tel modele necessite la connaissance de parametres tels que: altitudes
de la surface de depdt au tours du temps et importance des soulevements recents de IAltiplano,
profondeur de Iinterface so&/sediments sous le bassin, profondeur du Moho, etc...

De ce fait, le probleme geodynamique majeur pose par les bassins de IAltiplano ne pourra etre
resolu que par une etude integrant geologic et geophysique.
280
281

FOREARC BASIN EVOLUTION AND PLATE KINEMATICS:


THE NORTHERN ANDEAN FOREARC

C.D. Williams, LG. Stlmpsonl & M.C. Daly2

1 Department of Geology. Unlverslty of Keele. Keele. ST5 5BG, U.K.


* Structural Studies Group, B.P. Exploration. Britannic House, London, U.K.

Active forearc basins provide the best opportunity to study the interaction

between plate movements and forearc tectonics. An insight into the plate

tectonic controls on arcs may be gained by the comparison of

well-constrained basin histories, selsmotectonlc analysis of the subduction

system and well-constrained plate klnematlc models.

The continental forearc basin stratlgraphies in Ecuador and northern Chile

preserve a history of contractional, extensional and strike-slip tectonics.

The Tertiary evolution of the forearc shows a very close correlation with

the changing convergence rate of the Farallon, and later Nazca, oceanic

plate with continental South America. The Incalc orogeny (3 40 Ma) occurs

at a tlme of a major change in Pacific plate motion. The angle of

convergence as well as the rate also appears to have a controlling effect on

forearc tectonics and the subduction of buoyant aseismlc ocean ridges

correlate with minor tectonic events such as the Quechuan orogeny (5 6 Ma).

A fast convergence rate will result in the subduction of thermally young,

relatively buoyant oceanic lithosphere. This produces increased interplate

coupling, a compressional forearc and basin inversion. Slow convergence

results in the subduction of thermally older lithosphere. trench rollback,

decreased coupling between the plates, extension in the forearc and basin

formation.
283

NOUVELLES DATATIONS K/AI SUR DES ROCHES VOLCANIQUES


TERTIAIRES ET QUATERN,AIRSS DES BASSINS CONTINENTAUX
INTRACORDILLERAINS D'EQUATEUR

M.G. BONHOMHE ', A.LAVENU l*, C. NO.BLET l**, F. DUGAS l*,


A. EGUEZ *+;* et G. VIVIER l

l Inst. Dolomieu, URA 69 CNRS, F-38031 Grenoble


l * ORSTOH, Ap 6596 CCI, Quito, Ecua,dor
l t 213, rue Lafayette, F-75400 Paris Cedes 10
l ** Botquelen, P-56610 Arradon
l *** ESCUela Polftecnica National, Ap 2759, Quito, Ecuador

Nous prisentons de nouvelles datations radiochronologiques


K/Ar concernant les d(p.6ts tertiaires et quaternaires des
bassins intracordillirains de la D(pression Interandine et
du Sud de 1lRquateur.
L'utilisation stratigraphique de ces donnies a nhcessiti 1'
analyse systimatique de8 roches et de8 feldspaths. Cette
nithode permet de mettre en (vidence, si ndcessaire, soit
la presence de matiriel dhtritique dans les tufs, soit 1'
altkration hydrothermale de certain8 composants des roches
(tudikes. A partir des r(sultat8 obtenus, nous avons pu
determiner les donncies stratigraphiques suivantes.
Dan8 le Sud de l'Equateur, le remplissage du bassin de
Cuenca est a'b;ge miocane infhrieur B miocine suphrieur.
Des pointements andhsitiques indiquent deucpCriode8 s&pa-
r&es de magmatisme, l'une oligocine infirieur, l'autre mio-
c3ne sup(rieur.
Plus au Nerd, dans la rkgion dAlaUSi, extremiti mhridio-
nale de la Dipression Interandine, 1'Pge oligocine infi-
rieur d'un volcanisme calco-alcalin est mieux dbfini. Ce
volcanisme est post(rieur a l'accrdtion de l'arc nacuchi
et 1'8ge oligo-mioc&ne de la Formation Alausi est confirmi.
Dan8 le centre de la Depression Interandine, dans le bassin
de Lacatunga-Riobamba, le remplissage rhcent ( Formation
Latacunga 1 est pliistoc&ne. 11 reposerait sur les volcani-
tes plioc&nes de la Formation Sicalpa. La Cordillire Orien-
tale est recouverte par les Bpanchementn de la Formation
Pisayambo dont 1'Pge miocbne est confirm6.
,- . -- -._
285

LA CUENCA INTRAMONTANA EN COMPRESION DE VILCABAMBA (SUR DEL ECUADOR).


ANALISIS TECTONOSEDIMENTARIO.

MAROCCO, R., LAVENU, A. y NOBLET, C.


(Convenio EPN-IPGH-CLIRSEN-ORSTOM)

l ORSTOM, Aptdo 6596 CCL QUITO, Ecuador.

l*Botquelen, 56610 ARRADON, France.

Abstract

The continental (lacustrine, lluvial. volcanic deposits) filling up of the Vilcabamba


lntermontane Basin ( Southern Ecuador) was controled by a NNW-SSE trending fault system. The
compressive regime of the basin evolution is linked with the general neogene geodynamic of the western
edge of South America Plate.

Key words: Ecuador, Neogene, lntermontane Basin, Synsedimentary tectonics.

IntroduccMn

Durante el Ne6geno, en 10s Andes ecuatorianos, se instaia una serie de cuencas


intramontanas con relieno lacustre turbidftico, fiuvial y volcanico, locaiizadas en ei Caiiej6n
Interandino, depresi6n tectdnica casi continua que, del norte al sur dei pals, separa las
Cordiiieras Occidental y Oriental. Estas cuencas intramontanas tienen varias caracterlsticas
comunes: tect6nica y voicanismo contemporaneos de la sedimentaci6n, din&mica controlada por el
juego de grandes fallas N 170 y N 40. Al sur del paralelo 3 40 ias cuencas ne6genas esten
iigadas al juego de faiias N170; coma la de Vilcabamba, ubicada entre 400 y 425 S y 7910
y 7922 W (fig.1).

El relleno sedlmentarlo

El relieno de la cuenca de Vilcabamba (fig.2) esta constituido, de abajo hacia arriba, por
ias formaciones Loma Bianca (lavas y piroclasticos) del Oiigoceno sup.(?), Trigal (arciiias) dei
Miocene, San Cayetano (areniscas) del Miocene sup., Quiiioliacu (conglomerados). El paso de una
a otra de las facies es progresivo horizontal y verticalmente.

En ei sur (ftgZA), la serie, conglomeradica, estrato y granocreclente, potente de 1000


m, presenta una organisaridn secuencial tlpica de 10s abanicos aluviaies. Los aportes
sedimentarios van hacia el WNW. Al norte (fig.28), la serie es lacustre con base fluviai
(secuencia I); lacustre turbidftica y fluviai distal (secuencia II); por fin conglomeradica, de
ambiente fiuvial proximal (secuencia iii). Los aportes sedimentarios van hacia ei W y ei NNW.
Las megaturbiiitas de la base de ia secuencia Ii indican el inicio de la inestabilidad de la cuenca.
286

Fig.1 . Mapa geol6gico de las


cuencas de Vilcabamba y Loja .

1: volchnico (fm Loma Blanca);


2: conglomerados de abanicos
aluviales (fm Quillollacu);
3: fluviatil (fm San Cayetano);
4: turbiditas lacustres (fm San
Gayetano;
5: lacustre fino (fm Trigal);
8: terrenos ante-ne6genos.
___.... * . ,., ---,--_

287

000~000
100

Ilki-
:::::.: :

.:, ;:.,.;

. :: . . .* . *. . .

II
.
. *

Fig. 2 . Columnas estratigrbicas en la cuenca de Vilcabamba.


A: sur de la cuenca; B: norte de la cuenca; 1: conglomerados; 2: areniscas
flUViale8 y lacustres ; 3: turbiditas; 4: oalizas lacustres; 5: carbh;
6: debris flow; 7: lutitas, limolitas; 6: evaporitas .
288

En las dos secciones, cuya base esta constituida por las volcanitas Loma Blanca, 10s
ambientes sedimentarios evolucionan, de abajo hacia arriba, de lo distal a lo proximal. Los
driekantef son frecuentes en los conglomerados de los abanicos aluviales; un clima arid0 debfa
reinar en esta regi6n durante el Miocene superior.

Tectonlca

Un sistema de fallas N-S a NNW-SSE, que control6 la dinamica de la cuenca, limita 10s
afloramientos nebgenos. El control es evidente en la pane sur de la cuenca donde 10s abanicos
aluvlales se superponen.

El analisis de las fallas muestra que jugaron en dos tiempos. El primer evento
(compresi6n N 30), contemporaneo de la sedimentaci&r, determina un juego dextral-inverso
de las fallas. El segundo evento (compresibn E-W) provoca un juego inverso de las mismas
fallas asl coma pliegues N-S. Este dltimo evento es contemporaneo de la sedimentaci6n de la
parte superior de 10s cones aluviales y de la generalizaci6n de la sedimentaci6n gruesa al
conjunto de la cuenca. La discordancia angular del tercio inferior de 10scones es otra prueba de
la actividad tect6nica compresiva durante la sedimentacibn.

Conclusiones

La evolucidn de la cuenca de Vilcabamba comienra en el Oligocene con la formaci6n de una


depresi6n tect6nica, orientada NNW-SSE, que entrampa rotas volc&kas. Despues, se instala en
la depresi6n una red fluvial que evoluciona rapidamente hacia un lago alimentado desde el sur. Es
la fase de abenura de la cuenca.

La llegada de megaturblditas en el lago, la prcgradaci6n sobre 10sdep6sitos lacustres de


un slstema fluvial distal, proximal y, por fin de 10s abanicos aluviales corresponden al cierre de
la cuenca en un regimen de compresidn N 30 y, posteriormente, E-W.

La cuenca de Vilcabamba muestra una evoluci6n analoga a la de las otras cuencas ne6genas
del sur del Ecuador. El movimiento hacia el none del bloque oce&kc costero, induce un regimen
compresivo contlnuo durante el Ne6geno en el horde oeste de la Placa America del Sur. La
compresi6n provoca un juego dextral e inverso de las fallas NNW-SSE, a ,lo large de las cuales se
forman cuencas controladas por un continuum tect6nico.
289

MAGMATISME
291

JURASSIC BIMODAL MAGMATISH IN NE PALMER LAND,


ANTARCTIC PENINSULA: GEOCHEMISTRY AND TECTONIC SETTING.

H.E. Wever & B.C. Storey.

British Antarctic Survey, Natural Environment Research Council, High


Cross, Madingley Road, Cambridge CB3 OET, UK.

Resume

On rapporte les premieres recherches d'une association blmodal de


basaltes tholeiitiques et des roches acides de la Peninsule Antarctique
qui, pendant les temps Mesozoyque, dtait continue avec les Andes de
1'Amerique du Sud. Des resultats geochimiques suggerent clueces roches
se sont formees dans un basin marginal situ6 en arriere d'un arc actif
continental.

Key words: tholeiite, back-arc basin, ensialic, crustal melt.

During Jurassic-Cretaceous times extensional marginal basins and


contemporaneous silicic to bimodal volcanism were common along the
continental margin of South America. The extent to which these phenomena
occur along the once contiguous proto-Pacific margin of Antarctica has
not clearly been established. There are few constraints on the
petrogenesis of the silicic volcanic rocks of southern South America, but
they are generally attributed to widespread crustal anatexis during
intra-continental extension associated with the fragmentation of
Gondwanaland. There is even more uncertainty about the tectonic setting
of the extended marginal basins, in particular to what extent their
formation is related to contemporaneous subduction along the Pacific
plate margin.

In this study we present new geochemical and isotopic data of a Jurassic


bimodal association of tholeiitic greenstones and silicic metavolcanic
rocks from the Antarctic Peninsula, which have some bearing on these
problems. The Antarctic Peninsula represents an "Andean" magmatic arc
system, juxtaposed between the Pacific Ocean and the Weddell Sea, and is
dominated by talc-alkaline magmatic rocks related to active subduction
during most of the Mesozoic. The prominent bimodal association occurs in
an previously unmapped region in NE Palmer Land, situated along the
inboard margin of the magmatic arc and the western edge of the Weddell
Sea.
292

The mafic greenstones are divided into three main sub-groups, which are
indistlnct In the field but have geochemlcal characteristics of IAT
(group I), E-MORB (group II), and transitional talc-alkaline-tholeiites
(group III). Groups I and II have a similar range in E15Nd (between
-1.2 and 3.71, which suggests they may have been produced from a similar
sllghtly heterogenous source. The rocks of group III are derived from a
slightly higher d50~d mantle source (+ 5.5).

Trace element concentrations of the sllicic metavolcanic rocks show a


distinct "within-plate" affinity. These rocks are peraluminous and have
variable but low d50~d (between -2.2 and -8.4) in comparison with the
greenstones. This indicates that they contain a significant proportion
of crustal material and have not been fractionated from the mafic
greenstones.

The Isotopic and geochemlcal characteristics of the greenstones of group


I and group III are uncoamnon for active Andean plate margins with a thick
continental crust. Their close temporal and spatial association with
E-HORB (group II), which have similar LILE-enrichment (Th/Ta ratios 2.2-
3.51, suggests that the greenstones formed in a supra-subduction zone
setting and were emplaced in a back-arc basin. This basin was broadly
contemporaneous wlth the formation of the Andean rotas verdes. However,
in a recent geochemical study on the "rotas verdes" from South Georgia
Storey & Alabaster (this volume) propose that these rocks formed In a
Gulf of California type setting, hence the back-arc basin in NE Palmer
Land cannot be regarded as a southwards extension of the "rotas verdes"
basin.

The silicic metavolcanic rocks from NE Palmer Land are contemporaneous


with the break-up of Gondwanaland and may to some extent be related to a
similar cause. They are not derived from the associated tholeiitic
greenstones, they have "whithin-plate" affinity, and formed by extensive
remobilization of crustal basement. This is consistent with their
formation in an ensialic back-arc basin where, during early stages of
continental lithospheric attenuation, the ascent of mafic mantle derived
magma may have caused widespread crustal melting.

In conclusion, we suggest that the bimodal association in NE Palmer Land


was produced in an extensional marginal basin located behfnd an active
continental arc system, i.e. a classic ensialic back-arc basin. By
anology, some of the sllicic to bimodal volcanism along the Pacific-plate
margin of South America may have formed In a similar way, which implies
that this type of magmatism may be more common for Andean orogenesls than
previously recognised.
293

THE PLUTONIC ROCKS OF THE SOUTHERN GERLACHE STRAIT, ANTARCTICA:


GEOCHRONOLOGY, GEOCHEMISTRY AND MINERALOGY

Parada Qki Jem SapWe, Ada Rlcanb. Guerm Rslsp. Muniiaga Francisco*. Berg Karstan.

l Cepartamenb da Geokafa Y Geoffsiy Unkersided da Chib. Casitfa 13518, cons0 21, Sawlago. Chite.
* Maratofre 6 GaobgfPekobgfe. FacutaOdes Scbnma et Techniques. 23. rue du Docteur Paul Mbhebn ,4023 .%it
Elienrm. tXdex, Frano.

Reaumen

Las mcas ptut6nk.a~estudiadas se cam&&an por una dsminu&n hwia el 0~s.~ ds tar adadas (Cretktm inferic+
Mibcw~~). Gaoqlmicamen te son simiiares am-qua sn tos granitofdes CretAdt se datecta una ten&n& de aumentar at StO,
hacfa et es% Esknaknes de h pmfunddad & la hmnti y da mphmmienb rugiers VI aumentc hada et oesw.

Key wor&: F%Jtism, An- geochmrmfogy. gecdwmtsby.

Introduction

The inbusi~ rocks of the G&ache Strait induda gabrros. diirites. tonafites. granodiirites and granites (Akro5n et at.
1976). which form part01 the extensiw plutonism developed in Ihe Antarctic Peninsula as a pmductof a continwus subducbn
since Early Mesozoic to Lals Tertiary (cf. Parkhurst 1962). Prevfour studier dstinguished pru and post-vokanic rocks ptutons
(West, 1974) and reported radiietric eges that indicate Cretaceous ages for the Dance Coast rocks and Tertiary ages for the
Wrenke Is. andSouthern Anversls. (Scott, 1965; Par&hunt, 1482: British AntucScSurvey. 1964). Thusaweshvarddecreasing
ages of tie pfulonio mdts is insinuated.

The aim of this study is to get a better undarstandrg of the poorly known plutonic history of tie Geriache Strait, based
ongeochronotogical. ge~emicalandmineralogicaldalaofalimitednumberof samples. ThisstudywastinmcedbythetNACH
grant 8.17, and is a conbibu%n to the IGCP Profact 249 An&m Msgmatism.

Geochronology

Three previous K-Ar ages of plutonic rocks 01he Dan00 Coast range from 117 to 94 Ma (cf. Scott. 1965, Sriksh Antarctic
Survey. 1964) and two Rt&r iscchrm ages gave of 131 f4 and 114 fll Ma (Pankhurst. 1982). A new biotite K-Ar age of 113
f3 Ma was obtained in a hornblende - biotke gram&rite of tie SE extreme of the Anvord bay (fig. 1).

Mineral and whole rock K-Ar ages in the 54-49 Ma range reported by the British Antarctic Survey (1964) gedogfcat map
in the Doumerlslardwereconf!kmedby twonewbiititeK-Aragesof 50.2f1.4 and55.7ii.6obtainedatSE Doumerlslandand
Lockroy Portin the Wmncke lstandreapecttvety (fig. 1). EarfyTemarybiitite K-Aragesof66i2and54.2f1.4 Maweruobtained
at IheNEextmeof Anvers Island. Atonatitecoltectedatthe Pafmerbasegaveabiotita K-Arageof 20.4 fl Ma,whfchissimiiar
b the previous K-k agas obtained h the surroundng areas.

Wtm the exwptkan of a Late Cretawous Rb-Sr age of Ihe Cape Monaco granites (cf. Pankhurst, 1962) on the extreme
SWof Anven Isfand. awestwardmigrationoftheptutonism withtimeisconfimted. Thelimitbetwwn~MesoroicandCemroic
plutonic rocks was roughly defined in fig. 1,

Geochemistry

Wh. the Mesozoic and Cenozoic intrusive rocks exhibit a wida Sf, range, fall in the VAG f&d of Pearce et al. (1964).
and have cabalkaline characteristics. The Cretaceous Dance Coast granitoids show and overatl in-ease in SiO, to meeast,
and the intarmedate Cenozoic rocks have the highest REE abundances and the bwert LaNb ratios.

Takenl~ether~eMesotoicandCanozo~samples,theYcontentsexhibitamaximumat~eintermediatemmposibns
whereas Nb and Zr show. respectively, a well and poorly defined positive correlaticn with S10r. The behaviour of Y. Nb and Zr.
and the presence of negative Eu anomalies in the intermediate and felsic rocks, suggest adenvation froma mc+abarb magma
by dinopyroxane + piagiodase removal.
.-. . __ .__ _.,__jl_ --a.. _-

294

Mall0 Mineralogy

Most of SW analyred pyroxener au Ca-rid! dfnopyroxene:orU~opyroxanewas only found in a Cretaceous granitoid


sampb. Udng he geobaromelw based on tie mole fraction Tschermaks of dinopyroxena in equilibriw with plagiodase and
guvtz (EUia. 1980). Um prearutw omdbns of he Ter3ar-ymagma satme. or pause region. in- 8Jgleweclfrcm4lDe.bu.

ANVERS Is.

Crotacrour plutonic roctr Tortiory plutonic rocks

$0 5 0 , (0 I, 20 2s 111
L ! : : : : : 1

Fg. 1. birtribubn of ho plutonicrocha in the Gedache Strait. modifiedaher Alarc6n et d. (1976). Numben indiilo
radffmetrk ages: endosed numbers are 2m ages obtained in ais study.

Moat of ti anafyred amphibolesaro magnesiohornbfende. The remainder ones correspond lo ferm-hombfendaand


actfnolilichornblende. The pressure condilims of the hornblends aysfaflizalfon. using the Hammarsbom and Zens (1oeS)
geobarome~r. twghfy increase lo the west from 1 lo 2.7 kbar.

No variationshave been detected in the biotitecompositions.The anafyzed biitilec have mde fractionannim (X_) of
0.4 - 0.5 with the exceplimr of an Early Tertiary sampk Uralhas X, equaf ID 0.6. In addbn. no vatiaticnr have been found
in the oxygen fugacitycondlions (lo-) f~ tie biotibscryslaftization.

Conclualone

The pfutonicdevelopmentof the Gerlache Strait is characterked by:


1)Awestwarddecreasingagesfrom Early CretaceourtoMiomne. Twononplutonicinlsrvalr~and~X1~emer(lehom
the avaitabfe geochrondogicd data.
2) Caloaikaline productaexhibitinga wide ccmpositfmd range, and on overall eastward Sii2 in the Cretawous Dana, Coast
graniloids. Tha geochemicd rimikritiea between rocks with similar Si02 but different ages, suggest a mmmm sour- md
simifarmagmak evdulion.
..1- . .II__^

295

Refwonooe

Afaroh. E.. Ambrur. J.. Cicay. L.. and Viiin. C. (1976). Geobgla del Estrecho de Qedadw enbv br pwalebr 64 y 5!Y Sv,
AntMia Chife~. Serb Cientffiiu. InstibllDAnt6rtbo Chileno. v. 1. pp. I-51.
.----

297

TECTONIC AND CEOCHEMICAL SIGNIFICANCE OF TERTIARY PATAGONIAN


BASALT& (40-50S) ARGENTINA

S. Mahlburg Kay*. A.A. Ardolino***, J.M. Cartes***. M. Franchi*** and V.A. Ramos**

l INSTOC, Snee Hall, Cornell University, Ithaca, NY 14853


l * Depto. de Geologia, Universidad de Buenos Aires, 1428 Buenos Aires, Argentina

l ** Servicio Geol6gico National, Av. Santa Fe 1548. 1060 Buenos Aires, Argentina

Abstract

Basaltic volcanism in the Patagonian plateau occurs with distinct events at different times and
places. In the south, voluminous Eocene basalts and Late Miocene-Pliocene basalts with no arc-
like chemical affinities are related ridge subduction, while to the north the Eocene basalts are
arc related. The significance of voluminous Late Oligocene-Early Miocene plateau volcanism
is unclear, but may be related to global events with melting concentrated in this region because
of the geometry of the Eocene subduction zone. Plio-Pleistocene basalts are related to
extensional back-arc activity.

Key words: Patagonia, Tertiary, basal& ridge subduction, plateau, back-arc volcanism.

Iotroductioo

Basaltic volcanism in the Patagonian Plateau (36-52S) in southern South America


occurred discontinuously from the late Cretaceous to the Recent. Important episodes took place
at different places and times with no one cross section of Patagonian being representative of
all the activity (Fig. la-d). The most important episodes occurred when there was minimal
volcanism in the arc to the west. Despite the extended volcanism, evidence for important
continental extension is lacking. Tectonic explanations for the different volcanic episodes
vary, as does the influence of an arc-like component in the source region (Fig. 2). However,
a common factor is that throughout this time the ocean crust subducted along this margin was
relatively young and oceanic ridges were nearby and at times subducted.

Paleocene-Eocene

Paleocene to Eocene volcanism in central and southern Patagonia is dominated by arc


volcanic activity north of 43S and extensive basaltic activity in the extra-Andean region to
the south. Volcanic rocks between 41-43S (Huitrera Fm) have a subdued arc geochemical
signature (Rapela and Kay, 1988) compared to modern SVZ (southern volcanic zone) arc rocks
suggesting that they are back-arc to the main arc to the west or geochemically transitional with
the basalts with intraplate signatures to the south (Posadas Fm). Voluminous silicic volcanic
rocks (also Huitrera Fm.) with a subdued arc geochemical signature to the east of the arc
between 41-43S could have resulted from melting associated with basaltic underplating of the
crust in the transition region.
298

LATE OLIGOCEF
1 EARLY MlOCEN

Fii la-d: Patwnian pl&c~ bualts and magmatic arc activity through the Ccnonoic.

Volcanism in the Eocene arc corresponds with a period of rapid subduction and major
volcanism along the Andean front and in the circum-Pacific region. The lack of arc volcanism
and the important plateau volcanism to the south is tentatively correlated with the Eocene
subduction of the Aluk-Farallon ridge in this region proposed by Cande and Leslie (1986).
Limited geochemistry on the basalts (Posadas Fm. and El Sombrerito basal& see Fig. 2) shows
no arc influence and is consistent with a depleted oceanic (ridge-like) source. The lack of a
residual arc geochemical signature is also consistent with a steep late Cretaceous subduction
zone below the Patagonian batholith that did not influence the lithosphere in the eastern
region.

Late Oligocene-Early Miocene

Late Oligocene-Early Miocene plateau volcanism occurred predominantly in the Somun-


cura plateau and central Chubut (40-45S) and was the most voluminous single magmatic
event of the Patagonian plateau. The initial stages of this volcanism (Somuncura Fm.) formed
the massive Somuncura plateau (40.5-43s) which is over 200 km long and up to 250 km wide
and from 10-100 m thick. Some of the plateau-forming flows, which are transitional tholeiitic
to alkaline basalts and hawaiites (52-5496 SiO2), were extremely fluid with single flows extend-
ing over 100 km. Their geochemistry hassimilarities to flood basalts and to Hawaiian tholeiites.
299

consistent with a major mantle melting event. These flows overlie and are underlain by tuffs
that bracket their ages between 34-25 Ma. An analyses of the available K-Ar whole rock ages
led Remesal(1988) to conclude that most of this volcanism took place near 26 Ma (i2).

In the eastern and southern parts of the Somuncura plateau, the Somuncura flows
(Ardolino 1981) are locally covered with the Quifielaf Formation which consists of less
voluminous, more alkaline mafic flows and differentiated alkaline rocks. The 25 Ma old tuffs
could be related to this event. The differentiated series follows the sequence - hawaiite,
mugearite. benmorite, trachyte. and comendite (Corbella 1984). Volcanism ends with the
eruption of the early Miocene alkalic basalts which are interlayered with acidic tuffs (Bajo
Hondo Fm.) and could temporally overlap the Quinelaf Formation. The geochemistry of these
later flows suggests a somewhat cooler mantle associated with lesser amounts of melting.

PATAGONIAN MAFlC ROCKS

o-0 20 40 60 80
Nb
Pi 1: Plot of Ba venue Nb (or Ta.16) ehoring regional and age ruiation# in relative amounts of meltins (increase
with nbundsnce) and influence of arc-like component (incnaring Ba/Nb ratio). Eocene (lined , Late Oligocsna-Early
Miocaw (unahded), Late Miocene - Plicanr (dotted), modem arc - SVZ (dark). Circled N - 18 40-44S), circled 9 -
(= 40-M - SO).

The cause of this voluminous volcanic event (= 1500 km3 f 50% in the Somuncura plateau)
is unclear. This important period of plateau volcanism appears to have been contemporaneous
with a period of relative quiescence in the arc as basaltic rocks to the west have extremely
subtle arc characteristics and are indicative of high degrees of melting or very depleted sources
(Rapela and Kay, 1989). Evidence of extension is minimal. There is no relation with a rifting
ocean basin or known hotspot. This volcanism coincides with a period of relative slow
convergence along the South American margin and plate adjustment in the southern Pacific,
but these events do not explain why voluminous activity occurred specifically in central
Patagonia. Plateau volcanism of this age appears to be essentially absent south of z 46S.

Perhaps, there is a relationship between the Somuncura volcanism and the locus of Eocene
300

subduction as the Somuncura plateau occurs to the east of the southerr. terminus of the Eocene
subduction zone. The Somuncura and Bajo Hondo flows have arc-like high alkaline earth/light
REE ratios (see Fig. 2), but intraplate-like Ta(or Nb)/light REE ratios. These ratios could
reflect complex lithospheric enrichment associated with a relatively shallow Eocene subduction
zone in this region andits later demise. The subducted plate would have been young and thin,
consistent with the transition to ridge subduction to the south. Removal of the subducting plate
from beneath an enriched. relatively hydrated lithosphere could have contributed to extensive
melting of the mantle.

Late Miocene-Plloceae

Late Miocene to Early Pliocene plateau volcanism is concentrated in southern Patagonia


(46-49S), although scattered basalts occur throughout the plateau. In the north, they are back-
arc to arc volcanism. Pliocene to Late Pleistocene activity occurs throughout the plateau and
north of 52S is generally back-arc to the active Andean volcanic arc.

The distribution and chemistry of Late Miocene-Pleistocene plateau basalts between 46-
493 (e.g.. Meseta de la Muerte, Fig. 2). as well as the distribution and chemistry of arc
volcanic rocks (Stern et al. 1984) and the youngest activity in the Patagonian fold and thrust
belt, appear to be spatially and temporally associated with the intersection of the Chile ridge
with the trench (Cande and Leslie 1986; Ramos 1989). Where the ridge is currently intersecting
the trench, arc volcanism is absent. To the south, where the ridge has already intersected the
trench, arc volcanism is occurring. These arc volcanic rocks have a larger slab component than
those north of the Chile ridge consistent with melting of a young slab in a preheated mantle.
To the south, where the ridge intersected the trench 6-16 Ma, compressional deformation
ended iust orior to the eruption of the Dlateau basalts in the Meseta de Muerte. These 8-4 Ma
plateau flows are dominantly alkali b&its which correlate in volume and incompatible element
abundances with the passage of a thermal anomaly (slab window) associated with the former
ridge beneath the region. They have no geochemical arc affinities consistent with the absence
of an arc to the west and no previous lithospheric enrichment associated with arc volcanism.

Late Pliocene-Pleistocene

This magmatic activity occurs throughout the plateau backarc to the Andean volcanic arc.
Activity in the Somuncura region (40-43s) is often clearly fault-related and has variable
amounts of a subdued arc signature. This signature is generally strongest in basalts overlying
the modern subducting slab. It also occurs in a few basalts in the northeastern part of the
region which may be related to the complex pattern of arc-like and intraplate basalts
recognized by Muf~oz and Stern (1988) to the north of 39% Extra-Andean basalts far to the
east generally represent smaller degrees of melt than those to the west.

References
Ardolina, A. lQ81. El vulcaniemo cenoroico de1 borda nwxirntaI de la meleta de Somun Curs. VIII Conp. Geol.
Art., Actv III: 7-26, Bucnoa Aires.

Cande, S.C. and Leak, R.B. 1988. Late Ccnonoic tectonica of the muthem Chila Trench. J. Geophye. Rn. 91: 471-
498.

Corballa, H. 1984. El Vulcaniww de la Akiplanicia de1 Somuncurn. IX * Con6. Ccol. Ar6.. Rclatorio:267-600.
Rap&, C.W. and Kay, SM. 1988. Late P&owic to ibcentmymatic rvolution of Northern Pata6enin. Epimda 11:
176-182.

Mulks, J. and Stem, C.R. 1988. The Quaternary of the louthem continental msr6iin of South America: tranwcne
structuraland pxhamical rui~tioru acrca the rcgment betmn SLI* 9 and 69.9. 1. South Am. Earth Sci. 1: 147-
162.

Rama, VA., 1989. Foothilh structure in Northern Magrllanw B-in. kuycntina. A.A.P.C. Bulletin 73(7): 887-QOS.

RcmcsaI, MB. 1Q88. Geolo6in y Petrolo6ta de lw Baa&on da la Meactn de Somuncura. Ph.D. Thaia. Uoivcnidd da
Bucnor Aira, 207~~.

Stem, C.R., Futa, K. and Muehlcnbache, K. 1984. Imtopa and trace clement data for erogenic and&tee from tha
Auntrsl Anda in Andean Mamnatiam: Chemical and isotonic conatraintq, Humon R.S. find Barreim (ad@)., Shiv*
Publishing Ltd., Cheshire, England, 61-48.
301

The Australandean Volcanic Zone (South Patagonia),

Rolf KILIAN

Mineralogical and Petrological Institut of the University of Tubingen.


Wilhelmstr. 56. 7400 Ttibinaen. W-Germu

Abstract

New geochemical data of volcanics of the Mt. Burney and Viedma volcano complete a
review on petrology of the six volcanic centers of the Australandean Volcanic Zone (AVZ:
49 - 55S). Volcanic products of recent activities of 1988 were observed on the Viedma ice-
field. The volcanics of the AVZ are acid andesites to dacites; basafts are missing; every
volcano of the AVZ shows small geochemic development. Volcanics of Cook island occupies
a special position, related to volcanics of rift zones. From Mt. Burney in 52S to Viedma
volcano in 49 S. the volcanics show increasing K. Rb. Ba contents as well as K/Rb and
%r/*Sr ratios. Volcanics of Mt. Burney belong to a low-K andesites and dacites. More to
the north, lavas and pumices of Lautaro, Viedma and Aguilera are med-K dacites. Crustal
contamination and magma mixing with fused crustal rocks are discussed for this SN
development.

Introduction

The volcanism in the south Patagonian Andes has remained almost unknown and only few
reports are available. Volcanic eruptions on the Patagonian Ice Cap were first reported at
the beginning of this century by QUENSEL (1910). Ash deposits in the Argentine pampas,
which originated in volcanic centers of the inaccessible Andes, were collected during a
Swedish expedition (1932-34) and examined by SALMI (1941) and more recently by STERN
(1990). Some rocks of the Lautaro. Aguilera, Mt. Burney and Cook island (fig. 1) have been
geochemically analysed by STERN et al. (1976, 1984) and FUTA & STERN (1988). The
volcano Reclus (Sr57S) has been described by South Chilean geologists (HARAMBOUR
1988 and MARTINIC 1988). In the area of the Patagonian Icefield, on the Viedma glacier,
a volcanic center and poducts of recent eruptions were described by KILIAN (1990).

Tectonic setting

The Austral Volcanic Zone (AVZ) extends between 49 and 55s. The Lautaro. Aguilera
and Viedma volcanoes are situated at the northern end, Reclus, Mt. Burney and Cook Island
are located more to the south (fig. 1). All these Cenozoic volcanoes are situated near the
convergent plate boundary between the oceanic Antarctic plate and the continental South
x.._____...-_ -- r

302

American plate. The convergence rate (3 cm/year. Minster et al. 1974). the tectonic and
. volcanic activity are lower than to the north of the plate triple junction, where the Nazca
plate is subducted at a rate of 5 to 10 cm/year. South of the plate triple junction. there is a
volcanic gap of 300 km at the continental margin.

Patagonlan Andes
wmlcmlozekvokUnou

area ot hvea~noll
cl
A cmozokvwkMo
---- Frwlur.zo~
__lA__ Ageofn-mowmk
cwslhmU.ye~
_._._.. Ff-

South American

Fig. 1: Kanoaoic volcanoes and plate tectonic structure of the Patagonian Andes.

Volcanic activity OQthe Viedma icefield of 1988

An extended area was observed on the Viedma icefield (Patagonia) showing intensely ice
melting, caused by the fall of hot pumices and large ash fractions after volcanic euptions.
Hot mud flows were formed, covering a large area of the glacier. By the hot mud flows a
valley net was formed in the ice surface with a depth of up to 35 m. The probable center
of eruption was located at 49225 and 7319W in an elevation of 1100 m. Pyroclastic
deposits of previous periodic eruptions were mapped; pumices and lavas were examined.
The eruptions must have taken place recently between Sept. and Nov. 1988.
. ._-I ._

303

Petrography

Minerals in volcanics of Viedma and Mt. Burney were determined by microscope and
electronmicroprobe messurements (65). Characteristic phenocrysts of volcanics of Lautaro.
Viedma, and Aguilera are plagioclase (1.5 - 4 mm), amphibole (0.5 - 2 mm) and a few
biotite (0.5 - 1.5 mm) and additionally a second phenocryst generation of idiomoph
hypersthene. Plagiocla& normally shows small anorthit xonation in the range between an 42
to an 60. A slightly inverse xonation often occurs. Large plagioclase phenocryst in many
cues show a strongly corroded middle xone with the highest anorthite contents (an 63) of
our microprobe messurements. Some xenolithes, consisting of orthoclase. plagioclase and
pyroxene, occur in lavas of Viedma and Aguilera and indicate contamination. Plagioclase,
biotite and amphibole phenocrysts are often rim corroded. Plagioclase phenocrysts of lavas
of Mt. Burner often show slightly inverse anorthite xonation in the range an 45 to 65.
Hypersthene is the most commun mafic mineral with higher Mg/Fe ratio (mg* 65) than in
K-dacites of Viedma.

Ceochemical and petrological aspects

The volcanics of the AVZ range from 59 to 68 wt.% SiO, and are acid andesites and dacites
in the K,O/SiO, diagramm (fig. 2). The analyses of the Viedma. Lautaro and Aguilera
volcanoes show the highest KsO levels of the AVZ. To the south, the K,O, Rb, Ba. and Zr
levels of rocks from the volcanoes Reclus and Mt. Burney decrease and correspond in the
latter case to a low-K calcalkaline series. High-Al basalts and basic andesites, typical rock
types of the SVZ (33 - 46 S), are missing in the AVZ. Apparently uniform plots of each
individual volcanic center of the AVZ in the K,O/SiO,