TECTONICS, VOL. 16, NO. 2, PAGES 239-258, APRIL 1997

Growthof the South Pyrenean orogenicwedge

AndrewJ. Meigs DivisionofGeological and Planetary Sciences, Calif•a Instituteof Technology,Pasadena

DouglasW. Burbank DepartmentofEarth Sciences, University.of Southern California, Los Angeles

Abstract. A six-stepreconstruction of the South Pyrenean DeCelles et al., 1993; DeCettes, 1994; DeCe!!es and Mitra, forelandfold-and-thrust belt in Spaindelineates the topographic 1995]. Othershave argued, in contrast,that critical-taper models slope,basal dtcollement angie, internal deformation, and thrust- have limited applicabilityto fold-and-thrustbelts becauseof front advance from the Early Eocene until the end of mechanicaland deformationalincompatibilities between the contractionaldeformation in the Late Oligocene. Style of thrust- geologicrecord and the model [Woodward, 1987; Price, 1988; frontadvance, dip of the basaldtcollement, slope of the upper Meigsand Burbank, 1993; Bombolakis, !994]. surface,and internal deformationare decoupledand not simply This paper is a retreat to geologic data from the South related. Internal deformation increased, decreased, and Pyreneanforeland fold-and-thrust belt in Spain. A uniquelyhigh- maintainedsurface slope angle at differentstages. From the onset resolutionrecord of deformation and topography,contained to the cessationof deformation, the basal dtcollement angle within sedimentsdeposited during thrusting, enablesdetailed decreasedoverall suggestingtranslation of the thrust belt onto reconstructionof the structuraland topographicdevelopment of strongercrust with time. Taper angle of the Pyreneanthrust thisorogen. An increasinglypopular approach is to interpretthe wedgewas fundamentally controlled by the flexuralrigidity of deformationaland depositional histories of thrustbelts in termsof thelower plate, the relative rate of creationof structuralrelief in critical-tapermaintenance [DeCelles and Mitra, 1995]. Our therear versus the front of the wedge,the extentof depositionof approach,in contrast,is to determine(1) how taper angle varies erodedmaterial within the deformingwedge, and the taperof the through time and (2) how that variation relates to spatial and pretectonicstratigraphic wedge. temporalpatterns of deformation,erosion, and deposition. We present a six-part sequential restoration that recovers the predeformationalgeometry of the thrust belt from its present, Introduction postdeformationalgeometry. Each sequentialrestoration is a time slice constrainedby abundant surface and subsurface One of the most significant, and controversial,theories for geologicdata and absoluteage data from syntectonicsediments. f01d-and-thrustbelt developmentduring the past 20 yearsis the The surface slope and basal d6collementvaried independently critical-taperwedge model [Chappie, 1978; Davis etal., 1983; andmused continuous taper-angle variation with time. Stochnal,1983; Davis and Engelder, 1985; Platt, 1986;Dahlen andSuppe, 1988; Jaumd and LilIie, 1988;Dahlen, !990; CoIletta GeologicalPredictions of the Critical-Taper eta!., 199!; Davis and Lillie, 1994]. Mechanicalproperties of Wedge Model rockswithin the thrustwedge, spatial and temporalpatterns of deformation,and topographic evolution are coupledaccording to In orderthat evolutionof the SpanishPyrenees be compared themodel. Thrust-belt growth revolves around maintenance of a with the critical-taperwedge model, explicit definition of the criticaltaper angle between the basal dtcollement and the surface model and of the geometric-kinematicpredictive framework it slope. Feedbackbetween internal deformation, which acts to provides is required. Six variables define the boundary buildtaper, and frontal accretion, tectonic thinning, and erosion, conditionsof the model. Two definethe taperangle itself: (1) the w!•ichact to decreasetaper, dictates the kinematichistory of a dip of basal dtcollement and (2) the surface slope; three criticallytapered fold-and-thrust belt. Only rarelyis datafrom determine the critical-taper value for a specific belt: (!) the thegeologic record of sufficientresolution to permitsimultaneous strengthof material within the wedge, (2) the strengthof the calibrationof material properties, deformational patterns, patterns material within which the basaldtcollement is localized, and (3) ofsyntectonic erosion and deposition, and topography [DeCelles thepore fluid pressmethroughout the wedge;the final variable, andMitra, 1995]. As a consequence,the modelis generally the backstop,provides the push on andgeometry of therear of the thoughtto be applicablebased on qualitativecompatibility wedge[Chapple, 1978; Davis etal., !983; Stockreal,1983; Dav/s betweenspatial and temporalpatterns of thrustingin some and Engelder, 1985; Platt, 1986; Byrne and Hibbard, 1987; 0rogensand the distributionof deformationpredicted by critical- Mulugetaand Koyi, 1987;Datden and Suppe,1988; Jaumd and tapermodels [Boyer and Geiser,!987; Geiserand Boyer,1987; LiIlie, 1988;Mulugeta, 1988; Dahlen, 1990;Colletta etal., 1991; Morley,1988; Lucas, 1989; Boyer, 1992; Burbank eta/., 1992b; Byrne eta/., 1993; Davis and Lillie, 1994]. If thrust-belt developmentfollows critical-taper wedge theory,a critical angle Copyright1997 by the American Geophysical Union. betweenthe topographicslope and the basaldtcollement must be maintained[Davis etal., 1983; Stockmai• 1983; Davis and Papernumber 96TC03641. Engelder,1985; Platt, 1986;Dahlen and Suppe, 1988; Jaumd and 0278-7407/97/96TC-03641$12.00 Litlie, 1988; Dahlen, 1990; Davis and Lillie, 1994]. Patternsof

239 240 MEIGS AND BURBANK: GROWTH OF THE SOUTH PYRENEAN OROGENICWEDGE

A. Stables•ng

rrater•l to be no/nrernal (2) •ccret•to toe

..... •. -.. •?.'-.':.•-?...-...... -...... ,?•

(3) i•tema•restoreuer•r•t/on taper to ...... ?-•-,.,.-,..'.,...,. -..-.. ... - ...... •...

Figure 1. Schematicdiagram contrasting modes of thrust-frontadvance. (a) A stable-slidingadvance (position 2) from an initial position(position 1) occursconcurrently with aggradationof materialin frontof advancingwedge (open dot pattern). In the absenceof erosion,accretion, or extension,the wedge maintains its taper. Co)Thrust-front advance by accretion(position 3) occursafter aggradationof materialin front (opendot pattern)and on top (vertical line pattern)of wedge(,position 2). Additionof materialto frontof wedgedecreases taper angle. internal deformation are modulated by changesin the surface to thecritical angle. Alternatively, if thethrust front advances by slope and in material propertieswithin and at the base of the stablesliding between t 1 and t2 (Figure 2b), the wedgeremains at wedge. Stresstransmission to and accretionof material at the toe critical taper, internal deformation ceases,and the thrust front occursonly when the wedgeis critically tapered. advancesgradually toward the foreland. Although theseend- Subcritical,critical, and supercriticalare the possiblewedge- members imply a simple relationship between internal taper states given by the model [Davis et al., 1983]. A deformation and thrust-front advance, they do not accountfor subcriticallytapered wedge has a taper angle below that required processessuch as erosion,isostatic depression from loading,or for slip transmissionon the basewhereas a supercriticallytapered tectonicthinning of the hinterlandthat alsoimpact taper with time wedge'staper angle is greaterthan the critical value [Davis et al., [Davis eta/., 1983; Platt, 1986; Woodward, 1987; DeCelles and 1983; Woodward,1987; Dahlen, 1990;Dahlen and Suppe,1988; Mitra, 1995]. DeCellesand Mitra, 1995]. Once a wedge has attainedcritical A particularset of geologicobservations are required,based on taper, two contrasting modes of thrust-front advance, that is, this framework,to comparea foreland fold-and-thrustbelt with repositioningof the deformation front toward the foreland, are thecritical-taper wedge model. Explicitly,the necessary geologic implied by themodel [Davis et al., 1983] (Figure 1). Translation data that must be determinedsimultaneously throughout an of a critically tapered wedge along its base without additionor orogenyare (1) the dip of basald6collement; (2) the dip of the removal of material to either its toe or baserequires no internal topographicsurface; (3) the spatial and temporaldistribution of deformation because the taper angle is maintained (herein deformation;(4) the contrastingstrength of materialswithin and referred to as stable sliding; Figure la). Alternatively, a at the baseof the wedge (includingdown dip changesin the critically tapered wedge whose thrust-front advances by the strengthof thebasal d6collemen0; and (5) thepore fluid pressure. accretion of material to its toe must experience concurrent Whereas the first three variables can be constrained with internal deformationin order to maintain the critical taper angle confidenc6in thePyrenees and the comparative strengths of rocks (hereinreferred to as accretion,Figure lb). withinthe wedge and at its basecan be qualitativelyassessed, the If periods of thrust-front advance by accretion can be pore fluid conditionscan not be estimated. Below we document differentiatedfrom periodsof stablesliding, a contemporaneous the changes(or lack thereof)in thesevariables between six well- and appropriate internal deformational responseis expected constrainedtemporal and geometricalreconstructions of the within the wedge. During the courseof a hypotheticalorogeny, SpanishPyrenean foreland fold-and-thrust belt. the couplingbetween thrust front advance,internal deformation, and taper angle are illustrated in Figure 2. At time to the thrust front relocates toward the foreland at the base of a tapered, GeneralCharacteristics of the PyreneanWedge stratigraphicwedge. Such an accretionevent is comparableto Collisionof the Iberianplate with Europeanplate created a that of initiation of the Salt Range thrust 1130km toward the compact,two-sided orogen (Figure 3a) [Mutloz, 1992]. Paired forelandof the previousthrust front in the PakistanHimalaya foreland fold-and-thrust belts and foreland basins were formed to [Davisand Engelder,1985; Burbank. and Beck,1989; Davis and the north and south of the axial zone, an imbricate stack of Lillie, 1994]. Between to and t 1, the thrustfront remainsfixed, crystallinethrust sheets. This study focused on the titrustsheets internal deformationaccumulates, and the taper angle builds to a of the centralsegment of the SpanishPyrenean foreland thrust critical angle (Figure 2). The history of the wedge after t 1 bell the South-CentralUnit [Choukrouneand Seguret,1973; dependson whether the next thrust-front advance occurs by Williamsand Fischer, 1984; Williams,1985; Murloz, 1992]. The accretionor stable sliding. If the advance at t 1 is by accretion South-CentralUnit consistsof two thrustsheets on the west(the (Figure 2a), the wedge lengthensand the taper angle decreases. Montsecand Sierras Marginales) and three thrust sheets on the Like the to-t 1 period, the thrustfront subsequentlyremains fixed east(the B6ixols, Montsec, and Sierras Marginales) (Figure 3b). duringthe tl-t 2 intervaland internaldeformation rebuilds taper Our analysisof a corridorparallel to theRibagorzana River on MEIGS AND BURBANK: GROWTH OF THE SOUTH PYRENEAN OROGENIC WEDGE 241

A. accretion, accretion B. accretion,stable sliding taper taper supercritical critical supercritical

subcritical subcriticalcritical

position position foreland foreland

internal defc .tioi1 i "i ./•i• o internaldeformation I • i (stable sliding) •- internaldeformation [ i 1" hinterland hinterland internaldeformation} i

tO tl t2 tl t2 time time • thrust-front advance F1 trailing-edge advanceOran) O taper(ct + 15,degrees) Figure2. Plotof wedge taper and position ofttmast front and trailing edge of thrust belt for accretion followed by(a) accretionand accretionfollowed by (b) stable sliding kinematic paths. In both casest o depicts the case of instantaneousaccretion such as that whichoccurs when the basal detachment suddenly propagates into the forelandin a stepwisefashion. Between to andtl, internaldeformation builds taper to a criticalangle. Note that in thecase of accretionfollowed by accretion,the total advanceof the trailingedge of the thrustbelt is due to internaldeformation (Figure 2a). The trailing-edge-thrust-frontdistance increases, decreases, increases, and decreaseswith time in this scenario(compare the thrust-frontand trailing-edge advance curves, Figure 2a). In contrast,the trailing edge advances becauseof internaldeformation and stablesliding in the caseof accretionfollowed by stablesliding, respectively (Figure2b). At t 1, the thrust-front-trailing-edgedistance remains fixed (comparethe thrust-frontand trailing-edge advancecurves, Figure 2b). Taperis thesum of B thedip of basald&ollement and a, thedip of thetopographic slope. thewestern side of the South-CentralUnit provides an excellent zone on the north (Figures 3, 4, and 6). Several general areato studythe evolutionof an entireforeland thrust-belt system observationsare obviouson initial inspectionof the section. The forseveral reasons (Figure 4). Major structuresare exceptionally Barbastro anticline is the southernmostsurface expressionof wellexposed from the forelandacross strike to the axial zone and deformation and is related to folding at the southerntip of the locallyhave vertical exposure up to 1 km, age relationships basal d&ollement [Martinez Pega anzl Pocovœ,1988; Losantoset betweenmajor and minor structuresare clear [Meigs, 1997], and M., 1989]. Two structuraldomains characterize the hangingwall the stratigraphicframework is well established (Figure 5) structuresof the SierrasMarginales thrust sheet [Meigs, 1997]. [Pocovœ,1978a, b; Siredand Puigdefctbresas,1985; Williams, On the south, a dense array of imbricate thrusts substantially 1985;Senz and Zamorano,1992; Meigs, 1997]. In addition,the modified the leading edge of the sheet (Figures 4 and 6). In neighboringEtude Continentaleet Oceaniquepar Reflexionet contrast,broad, open folds characterizestructures in the hanging RefractionSismique (ECORS) line, a deep seismic profile wall to the north of the Montargull thrust. No internal (Figure3), other structuralanalyses in the region,and a dense deformationis seenwithin the Montsec thrust sheetexcept on its array of subsurface information from well and seismic data southernedge where a disruptedanticline is preserved[Misch, providefight constraints on thedip of the basald6collement, the 1934; Williams, 1985]. A wedgeof undifferentiatedsedimentary subsurfacelocations of importantcutoffs, and ages of unitsbelow and metamorphicrocks of the axial zone mark the northernedge thebasal d•collement [Misch, 1934;Pocovt; 1978a; Williams and of the section. This axial zone wedge is seen along the strike Fischer,1984; Williams, 1985; MartinezPetla and Pocovœ,1988; length of the Pyrenees and is interpreted to represent the Choukrouneand ECORS Team, 1989; Losantos eta/., 1989; stmcturallyhighest horse in theantiformal stack duplex [Williams Roureet al., 1989;Sdez eta/., 1991;Mu•oz, 1992;Vergffs et al., and Fischer, 1984; Williams, 1985; Martœnezet al., 1988; 1992,1996; Verggs, 1993; Senz and Zamorano, 1992; Meigs et 1992;Vergds, 1993; Verggs et al., 1996]. In this sense,the axial a/.,1996; Meigs, 1997] zone acted as the backstop[Davis et aL, 1983; Byrne and Hibbard, 1987; Davis and Lillie, 1994] in the eraplacementof the A GeologicCross Section of thePyrenean foreland fold-and-thrust belt. Included in this undifferentiated ForelandFold-and-Thrust Belt backstopis the Nougatszone, a second,smaller duplex formed in A present-daycross-section of the foreland fold-and-thrust belt Triassic and basementrocks above the uppermosthome in the extendsfrom the uncleformed foreland on the south to the -ial axial zone duplex [Williams, 1985; Mu•oz, 1992]. Two 242 MEIGS AND BURBANK: GROWTH OF THE SOUTH PYRENEAN OROGENIC WEDGE

0o0 , 0ø3(7E 1ø0' E 1o30'E E•aof Blsca• FranceI I •/////7/• ' / / ///////,-,,,

42o20' N •ENE•$B,m--'l'.. ' "'S ß • •r ' Axia/Zone• 42o20'N

42o0' N 42øo N

. •]' '

+ + F•.; m B o •uma-1 E•lus-1 •a]agucr 41ø40'N 4 o40' N

0o0 , 0ø3•E I•'E lO30,E •Trk K •P-• •B•b ••-O anticl,•• ncl•e+well ••t fa•t Figure3. Locationgeologic maps ofthe Pyrenees andstudy area. (a) Regional tectonic map showing themajor geologicfeatures ofthe Pyrenean orogen. SCU identifies thesouth central sector ofthe Spanish Pyrenees (modified fromVerggs [1993]). Note the position of the Ribagorzana crosssection (Figure 6) withrespect tothe Etude ContinentaleetOceanique parReflexion etRefraction Sismique (ECORS) deep crustal seismic profile. The axial zone isindicated bythe cross hatch pattern beneath theword Pyrenees. (b)Geologic map of the South-Central Unitof the SpanishPyrenees. TheBdixols thrust, which has considerable displacement tothe east, loses displacement tothe west andis not present; only the Montsec andSierras Marginales thrustsheets areseen (base after Teixell [1992]). Note thatthe frontal portion ofthe Sierras Marginales thrustsheet ispartially obscured byOligocene syntectonic strata.Key featurestonote are the Barbastro anticline inthe foreland, interpreted torepresent theleading edge of Pyrenean deformation,andthe Montargull thrust 0VIT) within the Sierras Marginales thi'ustsheet. Stratigraphic sections usedto constructthestratigraphic panelof figure 5 are indicated (sections A-E). SFZ denotes themostly buried Segre Fault Zone;abbreviations areas follows: AB, Ager basin; TGB, Tremp-Graus basin;Az, undifferentiated rocksof the axial zone;Trk, Triassic evaporites; UK,Upper Cretaceous; P-MEo,Paleocene toMiddle Eocene; B,Barbastro evaporites (baseofthe Ebro foreland basin sequence); andUEo-Oligocene, UpperEocene toOligocene syntectonic strata. Pretectonicrocks are the Triassic through Paleocene. Thelocations ofFigures 4 and 6 areshown. extensivelystudied Early to Middle Eocene piggyback basins, the (Figures3 and 4). Extensivemapping inthe southern portion of Ager and Tremp-Grausbasins, sit above the northernSierras thestudy area revealed a number of keyobservations that define Marginalesand Montsec thrust sheets, respectively (Figures 3, 4, thestructural sequence [Meigs , 1997]. Thismapping was and6) [Muttiet al., 1985, 1988;Losantos et al., 1989;Muhoz, supplementedby previous structural analyses on theSierras 1992]. Marginalesthrust sheet and detailed studies ofthe leading edge of theMontsec thrust sheet [Misch, 1934; Pocovt; 1978a; Mutti et a/., 1985,1988; Losantos etaL, 1989].Data for the section from Structural Data theTremp-Graus basin to the north was derived from the geologic Surficialstructural data for the sectioncame from two sources. mapCatalunya [Losantos et al., 1989],a previouslypublished crustalsection [Williams, 1985], and a detailedregional study of Detailedmapping from foreland to thesouthern margin of the thecrustal structure of the eastern part of theSouth-Central Unit Tremp-Grausbasin was conducted todelineate the geometries of [Verggs,1993]. The northernpart of the sectiongreatly majorstructures and to determine the structural and stratigraphic simplifies structure north of thetrailing edge of theMontsec relationshipsbetween structure and synorogenicsediments thrustsheet (Figure 6). MEIGS AND BURBANK: GROWTH OF THE SOUTH PYRENEAN OROGENIC WEDGE 243

Legend $tratieralghic Unit,

UO Unit 3 UO Unit 2 - LO Unit 1 } Jr + Protectonic UEo- LOb !•LEo +

x•Xsyndine • anticline %thmst fault•contact

strike+ dip. (-•__.•)growth strata ßx,'• horizontalinclined k vertical

o tOWil • RibagomnaRiver

X elevation

2km N •' " •o

41ø57' N

meters

'4't.' ßß . o ,. -•, "*..- . . . ::' "'".:.., '..'..•. ,,o .-'•b.. ," "..... ' ...... ',•.. a•. sa '-.., •3 '

4x ' .o a- •'

ß.....•0 meters• 41•2'Nl'&a,a'a,'••'•.'&.-.'•a,a'a,•'.-.'&.•• •I• 2•. .o"• o ' •- •.•.•:••.:•.•.'•'. 41•N

i

F•ure g. G•logic mapof •e study•em Compiledfrom ex•nsive orig• fieldmapping, modfficafions of •e m• of Pocov[[ 1978a],• •e geologicmap of Cat•unya [•santos et M., 1989]. •e elevationsof •e highest•h• on ß e •p of •e youngestnomic syntectofficunit • denot• by crosses•d •e elevation• me•rs a•ve s• level. SMT is •e Sie•as M•ginales t•ust and MT is the Mont•guH •ust. S•atigraphicunim •e subdivid• in• pre•nic andsynmc•nic groups. •e•ctonic u• •e Trk, TriassicKeu•r •d M•che•a• facies;Jr + • J•sic •d C••us c•bona•s undffferenfiated;and Pg, PMeo•ne G•mnian facies.Syn•c•nic •i• •e L• + MEo, •wer •d Middle Eocene l•estones undifferentiated;UEo - LOb, B•bas•o Formation; LO Unit 1, unit I in nom•me s•ata; UO Unit 2, unit 2; and UO Unit 3, unit 3. Bre• •tween •o + MEo and UEo - LOb m•ks •e changefrom m•ne to no••ne de•sifion in •e s•t•maic suckssion.•e sou•em•ion of •, •st• section (Fig•e 6) is hdicated. 244 MEIGS AND BURBA•: GROWTH OF THE SOUTH PYRENEAN OROGENIC WEDGE

A B C D E 2w•erOligocene U Oligocene - < 26 Ma) Middle Eocene (30 - (51 - 47 Ma) UpperEocene- Lower Eocene LowerOligocene (BarbastroFm.) (36.5- 31 Ma) Paleocene

Cretaceous

10xVE kilometers

Jurassic

Syntectoni9Strata. Trimsic :• Unit3(units 3and 2combined, 28- <26 Ma) :'::..:'"•Unit1 (30- 28 Ma)

,,• Undifferentiated(includesBarbastro foreland-basinFm.; 36.5 - deposits31 Ma)

[--• Lower(55and- Middle51 Ma andEocene51 - limestones47Ma, respectively) undifferenfiated Figure5. Stratigraphiccolumns forthe study area. A northwardthickening wedge ofpretectonic strata (Triassic- Paleocene)isoverlain by syntectonic strata (Eoeene-Oligocene). Marineand terrestrial deposition characterize both pretectonicandsyntectonic strata. Column A incorporates datafrom Meigs [1997] and Senz and Zamorano [1992]. ColumnsB and C weremeasured byPocovœ [1978a, b]. TheMontsec sections arefrom Simd and Puigdefabregas [1985](columns D and E). Patemsare the same as Figure 4. SeeFigure 3 for localities A-E.

Stratigraphic Data (Figures5 and 7). Marinedeposition on a northfacing passive margincharacterizes the Triassicthrough Cretaceous [Pocovi Stratigraphicinformation comes from a varietyof sources. 1978a;Sire6 and Puigdefabregas, 1985]. Lacustrineand fluvial Measuredsections and detailedstratigraphic analyses are rocksare preservedwidely acrossthe South-CentralUnit and abundantin the region. A studyof an exposedpart of the marka short-livedinterval of continentaldeposition in the forelandbasin succession to the westof thestudy area which Paleocene[Pocovœ, 1978a; Williams, 1985]. Whereas these rocks integratedinformation from well logs and measuredsections are syntectonicwith respectto the B6ixols thrustin the east [Senz and Zamorano, 1992] and a measuredsection across a (Figure3), theyare pretectonic with respect to theMontsee and piggyback-basinsuccession within the studyarea were used to SierrasMarginales thrust on the west [Pocovœ, 1978a; Williams, constructthe stratigraphic column for theforeland (Figure 5A) 1985,Muiioz, 1992; Puigdefc•bregas etal., 1992;Vergds, 1993]. [Meigs, 1997]. Thickness variations within the Sierras Becausethey are relatively thin, display only a gradualnorthward Marginalesthrust sheet are known from a seriesof 14 measured thickening,and axe well exposedthroughout the region, the sections(Figure 5, columnsB andC) [Pocovi,1978a, b). The Paleocenestrata provide an ideal horizontalreference for greatestchange in stratigraphicthickness within the Montsec stratigraphicand structuralreconstructions [Williams, 1985; thrustsheet is exhibitedby Cretaceous rocks studied in detailby Vergds,1993]. Overall, the pretectonic section thickens gradually Sirndand Puigdefabregas [1985] (Figure 5, columnsD andE). from < 300 to 400 m on the southto > 5000 m on the north Thesestratigraphic data were combined with surficialstructural (Figure5). informationto constrainthickness on the crosssections. A returnto marinesedimentation is coincidentwith theonset A numberof stratigraphicobservations provide critical ofcontractional deformation inthe study area (Figure 7) [Pocov• constraintson thespatial, temporal, and geometrical evolution of 1978a;Meigs and Burbank, 1993; Meigs, 1997]. Syntectonie thethrust belt. The stratigraphy consists of rocksdeposited prior stratigraphicgeometries observed in Lower Eocenemarine to theonset of contractionaldeformation, pretectonic strata, and limestonesmark earliest-formedstructures. Middle Eocene syntectonicrocks deposited toerally with folding and thrusting syntectonicstrata consistof marine rocks on the south and MEIGS AND BURBANK: GROWTH OF THE SOUTH PYRENEAN OROGENIC WEDGE 245 246 MEIGS AND BURBANK: GROWTH OF THE SOUTH PYRENEAN OROGENIC WEDGE

.• • Tectonicevent , ß,- •'• ß , •, Source • ••i•'• <28(Fig.Ma6) MontargullFoldinginthethrust. foreland. 1.2 Meigs[1997]etal. [1996] BatbastroauG•.l• • Montaegullthm•t •.•,.'_.•,.'.- • _ • ' • Tre.mp-Cna• Az/aZZo•.

Meigset aL [1996] .= (Fig. 8) hangingwallof Sierras 3.5 Meigs [1997] '-, Marginalesthrust sheeL ,--• 30-28Ma Folding inand thethrusting foreland.in B•b,•tro•ticli• inactiveSierras Mar•inales thrust __ ß- • Mort uEdtru•t Tmmp-Orau,

......

-o o Meigset at [1996] •' '• (Fig.9) SierrasMarginales thrust 29 • oe sheetover foreland. -9ø••' !• 96-30Ma FoldingFinal translationinthe foreland. ofthe ß-ticline SierrasMa•Saakn thrust •_ Mo.tazr_tullthran Montes: thm•t Tt•mp-.Onu• Ax••.

(Fig.10) M•ginales•mst s•et 20 Meigset aL [19•] 51-36Ma overInitialforeland.•slation of Sie•as BuickPocov[,[1978a,etaL [1 b]•a] Montsecthrust. Verges,[1993] Vergeset al. [1•2] •• • Mo•tu••

55-51Ma Foldinga• •r thrusfi• Pocov[[1978] (Fig.11) acrosst• •dth of the Meos[199• thrust-belt. •.8 MMtietal. [1985] Initialfo•ation ofa• zone. M• etal. [1988] E•lace•nt ofPont • Mu•oz[l•] Suedet•ust sh•t. Verges[1•3]

> 55Ma pretectonicdeposition Pocovœ[1978] (Fig.12) Puigdef•bregasetal. [1992] verges[1993] 14t711iams[1985]

Figure7. Summarydiagram of syntectonicstratigraphic units, ages, tectonic events occurring contemporaneously withdeposition ofeach distinct unit, and data source. Note that the figure of the corresponding structural cross section isindicated inparentheses below the age. Dashed lines represent inferred orknown conformable contacts. Wavy lines representmapped unconformities (seeFigure 4). Totalshortening measured fora giveninterval is given in kilometers. Seetext for complete descriptions ofthe tectonic events. Structural development sketches areapproximately scaled horizontallyand have no implied vertical scale. Coeval deposition isschematically depicted in the foreland but not in piggybackbasins developed on the thrustbelt.

transitionalto nonmarinerocks on the north [Pocovœ,1978a; 1985;Mutti et al., 1988],they constrain the position of thethrust Mutti et al., 1985,Puigdefttbregas et al., 1992;Vergds et al., belt'supper surface with respectto sealevel duringthe early 1992; Vergds,1993]. BecauseLower and Middle Eocenerocks stagesof development.Finally, Upper Eocene through Oligocene arecharacterized by shallowmarine andshallow interfingering nonmarinesediments in theEbro foreland basin and in piggyback with nonmarinefacies, respectively [Pocovœ, 1978a; Mutti et aL, basinson topof theevolving thrust belt constrain depth of fill, MEIGS AND BURBANK: GROWTH OF THE SOUTH PYREI•AN OROGENIC WEDGE 247 extentof burial of structures,paleorelief on basin-margin Age Data structures,and synorogenic topography during later stages of Absoluteages of deformationalperiods come exclusively from c0n•actionaldeformation [Burbank eta!., 1992b;Vergis, 1993; Lower Eocene to O!igocenesyntectonic sediments dated Burbankand Vergis, 1994; Meigs et al., 1996;Meigs, 1997]. biostratigraphica!lyandmagnetostratigraphically (Figure 7). No Presentstratigraphic thickness were used and no attemptwas thermochronologicagedata was available for thearea. Upper madeto decompact the syntectonic strata. Eoceneand Oligocenenonmarine deposits have beendated palcomagnetically[Meigs et al., 1996;Meigs, 1997]. In addition to thedeformational chronology in thestudy area [Meigs, 1997], SubsurfaceData a chronologydetermined along strike to the eastat Artesade Althoughno subsurfacedata were availablewithin the study Segre(Figure 3) is incorporatedbecause it containsa similar area,a wide range of subsurfaceinformation has been published recordof deformationto thatof thestudy area and clear evidence throughoutthe region. Severalmajor questions regarding the of olderdeformation in addition[Meigs et al., 1996j. Lowerand deepcrustal structure of the Pyreneeswere resolved after the MiddleEocene marine rocks have been dated biostratigraphically acquisitionandinterpretation of the ECORS deep seismic profile [Pocovt,1978a, b; Williamsand Fischer, 1984; Mutti et at, 1985, (Figure3) [Choukrouneand ECORS Team, 1989; Mu5oz, 1992; 1988;Farrell et al., 1987;Mu•oz, 1992;Puigclefdbregas et at, Roureet at, 1989]. This seismicline clearly imagedthe basal 1992]. Six distinctstages in thedevelopment of thePyrenean d&ollementof thePyrenees and constrained its depthand nearly foreland fold-and-thrust belt can be differentiated with confidence constantdip northwardto beneaththe axial zone[Mur•z, 1992]. basedon syntectonicsediments and structuralrelationships. Theexact dip anddepth are somewhatvariable along strike (3.5 ø Sequentialrestoration of the presentcross section (Figure 6) -4'; [Mufwz,1992; Vergds, 1993]). Thereforestratigraphic data revealsthe thrustbelt geometryas the effects of deformationafter withinthe study area, a crustal section to the west [Senz and 28.0 Ma (Figure8), between28 and30 Ma (Figure9), between Zamorano,1992], and a well-constrained grid of derailed 30 and36 Ma (Figure10), between36 and51 Ma (Figure11), balancedcross sections constructed to the east [Verggs, 1993] andbetween 51 and55 Ma (Figure12) arestripped away. wereused to constrainthe present dip as -4 ø +/- 0.5ø at a depthof The mostrecent deformation aœfected the youngestexposed -1000 m on the south beneath the foreland to -4500 m below the syntectonicsediments on a few structuresin the studyarea, but tremp•rausbasin on thenorth (Figure 6). thesestrata are otherwise undeformed and unconformably overlie A numberof otherimportant subsurface structures, which have nearly every structure(Figures 4 and 6). These nonmarine noobvious surface expressionin the study area but that are alluvial and fluvial rocksare informallyreferred to as unit 3, assumedto exist at depthand are included in the geologicsection, whichincludes, for thepurposes of thisstudy, a distinctlocally wererevealed by the ECORS project and subsequentstudies exposedand preserved conglomeratic unit (unit 2 [Meigs,1997]). (Figure6). A footwall anticlinebeneath the Montsecthrust fault Unit 3 rangesin agefrom 27.9 to < 25.8 Ma. A lithologically waspenetrated by the Comiols well (Figure 3), is seenon the similar deposit occupying an analogous structural and ECORSsection, and is inferredto existalong the strikelength of stratigraphicposition at Artesade Segreranges in agefrom 27.8 thefault [lduaoz, 1992; Vergds, 1993]. Althoughthe Sierras to < 24.6 Ma [Meigset al., 1996]. Deformationaffecting unit 3 Marginalesthrust was previouslysuspected to be significantly occurred after 28.0 Ma. displacedwith respectto its footwall cutoff [MartinezPeFm and Deformationbetween 30 and28 Ma is markedby foldingand Pocovi,1988], the position of thecutoff was unconstrained prior thrustingof unit 1 (Figures4, 5, and7). Unit 1 unconformably tothe ECORS profile. Giventhe ECORS image [Mugoz, 1992] overliesa varietyof olderrocks on boththe SierrasMarginales andsubsequent structural, stratigraphic, and paleogeographic thrust sheet and in the foreland [Meigs et at., 1996]. It is reconstructions[Vergds, 1993], the displacementof the Sierras unconformablyoverlain by unit 3. In contrastto unit 3, unit 1 is Marginalesthrust sheet is well constrainedregionally as _45 m. deformedeverywhere it is exposed(Figure 4). Unit 1 rangesin Acomparable amount of displacementis inferred in thisstudy. age from 30.1 to 28.0 Ma [Mei$s, 1997]. Unit 1 also has a A final debate,spawned in part by the ECORS results, stmcturally and stratigraphicallycorrelative unit at Artesa de concernsthe nature of the rocks sandwiched between the base of Segrethat rangesin age from 29.5 to 28.5 Ma [Meigs et al., theSierras Marginales thrust sheet and the basald/collement. 1996]. Growth stratal geometriesin unit 1 indicate that the Mu•z [1992]argued on the basis of structuralrelationships near deformationoccurred coevally with deposition. thesouthern edge of the sheetthat this subsurfacespace was The third deformationalepisode (36 to 30 Ma) is tightly filledby a "double"Sierras Marginales, a completethrust constrainedby relationshipsat Artesade Segre[Verges, 1993; duplicationof the sheet. On the basis of structural,stratigraphic, Ver$is et al., 1992; Meigs et al., 1996]. Map relationships andsubsurface data, in contrast,subsequent authors argued that suggestthat the SierrasMarginales thrust sheet was eraplaced theSierras Marginales thrust sheet had translatedover a acrosspreviously folded forelandbasin strata. Both the Sierras significantportion of the foreland[Senz and Zamorano,1992; Marginales thrust and folded foreland sequence are Vergdset at, 1992;Vergds, 1993;Meigs eta!., 1996].Because unconformablyoverlain by unit 1. The exposedportion of the substantialduplication of theSierras Marginales thrust sheet is foreland-basinsuccession at Artesaranges in age from 36.5 to 0nlylocally observed in thestudy area [Meigs, 1997], the Sierras 30.7 Ma. This unconformablerelationship indicates that the Marginalesthrust sheet is interpreted tohave been displaced over SierrasMarginales thrust sheet was in its presentposition by -30 theforeland (Figure 6). Theposition of theramp across the Ma. A combination of geometrical observations,subsurface forelandisinferred to sit below the south flank of theAger basin, relationships,and paleogeographicinterpretations support the aninterpretation supported by surfacedip dataand consistent inference that the Sierras Marginales thrust sheet began to withother analyses [$enz and Zamorano, 1992; Vergds, 1993]. override the foreland succession at-36.5 Ma. 248 MEIGS AND BURBANK: GROWTH OF THE SOUTH PYRENEAN OROGENIC WEDGE

A poorlyconstrained period of shorteningbetween 51 and 36 Ma is inferredfor the study area based on deformational evenh elsewherein thePyrenees during the same period. Within the study area, Middle Eocene (Lutetian) limestonesthat unconformablyoverlie tilted older strata are themselvesfolded andunconformably overlain by unit 1 [Pocovœ,!978a, b; Meigs, 1997]. MiddleEocene strata are present in thefootwall of the SierrasMarginales thrust sheet (beneath the Ager basinand Montsecthrust, Figure 6). Well-constrainedMiddle Eocene shorteningon the structurallycorrelative thrust sheet of the SierrasMarginales thrust sheet in theeastern Pyrenees, the lower Pedraforcathrust sheet [Martœnez etal., 1988;Burbank et 1992a;Vergds etal., 1992, 1995;Vergds, 1993], implies contemporaneousshortening in the South-CentralUnit. On this basis,we differentiatea discretedeformational period between -51 Ma and 36 Ma. Initialshortening in thestudy area is markedby growthstrat• geometries[Riba, 1976; Anaddn et al., 1986;Vergds etal., 1996] in Lower Eocene marine limestones[Meigs, 1997]. Well- constrainedbiostratigraphically determined ages of these limestonesplace the initial deformation in theEarly Ypresian (55 to 51 Ma [Pocovœ,1978a; Mutti eta/., 1985, 1988; Candeand Kent, 1992; Mu•oz, 1992; Vergds,1993]). Restorationof this deformationdefines the pre-55-Ma geometry of thepretectonic s•atigraphicwedge. When combined with stratal thickness and structural I.l.I O geometries,these constraints allow sequentialrestoration of the dip of the basaldtcollement, the topographicsurface, and the internalstructural geometry of the Pyreneanthrust belt fromits present,postdeformational state backwardthrough time to its predeformationalstate in fivedistinct temporal steps.

SixSteps in theEvolution of thePyrenean Wedge A deformedstate cross section is an aggregateof superposed deformations[Geiser, 1988; Meigs, 1997]. Sequentialrestoration is a processof tectonic"ackstripping"which, through the restorationof structuresresulting from an individualperiod deformation, reveals the consequencesof successive deformations. The Pyreneanforeland fold-and-thrustbelt is steppedbacl•ward through time from its present cross-sectional geometryto its predeformationalgeometry in four intermediate steps(Figures 6 and8-12). Eachrestoration involves the removal of shorteningaccommodated by displacement on faultsand/or by foldingas recorded by therelationship between folds and faults andsyntectonic strata [DeCelles eta/., 1991;Vergds et al., 1996; Meigs,1997]. Details of the restorationprocedure and geometricaljustification for structures on the southern edge of the SierrasMarginales thrust sheet are given by Meigs[1997]. In

Figure 8. 28.0-Ma reconstruction.(a) Total recovered shorteningis 1.2 km from reconstructionof a fold in the former positionof theMontargull thrust (Figure 6). Topographicslope is 0ø and dip of thebasal dtcollement is4 ø. (b) Plotof basement subsidencerelative to 55 Ma basement.Note that the position of thebasement is inferred to haveremained approximately fixed after30 Ma. Seetext for discussion. (c) Shorteningrecovered from restorationof structuresactive after 28 Ma. Patternsand stratigraphicunits are the same as Figure 6. MEIGS AND BURBANK: GROWTH OF THE SOUTH PYRENEAN OROGENIC WEDGE 249

addition,the tectonic subsidence of the basement isrecovered by "hanging"partial restorations of the deformedstate cross section constrainedstratigraphically and structurally. Linelength and area are conserved between each step for the pretectonic strata and Triassic evaporites, respectively. Shorteningis measuredby thechange in positionof a pin line locatedin the axial zone with respectto a pin line in the undeformed foreland on the south of the section. The hinterland pin line has no geologicalsignificance, it merelyserves as a referencewith which the forelandward displacement of the rear of thedeforming wedge can be measured. Because of uncertaintyin faultdisplacements where hanging wall cutoffsare eroded and shorteningrelated to undocumentedfabrics, such as cleavage, formedby layer-parallelshortening, intermediate and total shorteningestimates are minimum values. Suchreconstructions illustratethe evolution of internaldeformation and emplacement of thethrust belt. Structuralrelief and stratigraphic relationships allowapproximation of topographyfor eachstep. In particular, thepreserved thickness of syntectonicstrata are usedto constrain theupper surface of thewedge at eachstep.

Present (Post-28-Ma) Cross Section Thepresent-day cross section (Figure 6) reflectsthe geometry of thethrust belt after 28 Ma, afterthe final deformationalpulse [Meigs,1997]. Nearlyevery structure is buriedby theyoungest nonmarineelastic deposit (unit 3) whoseage ranges from 27.9 to < 25.8 Ma (Figures 4 and 6). Unit I is deformedby the Montargullthrust and related imbricates and probablyby late modificationof theBarbastro anticline in theforeland (Figure 7). A cross-sectionaltaper of 5 ø is given by the 4ø-dip basal detachmentand lø-slopeof the top of unit 3 (Figure6). This surfaceis constrainedfrom the hangingwall of the Montsec thrustto the forelandby projectionof the highestexposures of unit 3 alongstrike onto the line of the section(Figure 4). This surfaceis only a "bestguess" of topographybecause the ageof the highestunit at any spotis likely to be diachronous.Tight limits on possibleuncertainty in the angleof the surfaceslope (+/- << 0.5ø) are indicated to the southof the Montsec because unit 3 barely overtopsstructural highs that otherwisewould define the surfaceslope. The projection to the north of the Montsecis an extrapolationof the averageslope defined to the south (dashed line, Figure 6). Its elevation and thickness are consistentwith nonmarinealluvial and fluvial deposits(Collegats Formation) that discontinuouslyoverlie the contactbetween the axial zone and the rear of the foreland fold-and-thrust belt [Mellere, 1992, 1993; Vergds,1993]. Note that the depth-to- basementand angle of the basementare the same after 30 Ma

, Figure 9. 30.0-Mareconstruction. (a) Restorationof nearlyall thrustfaults within the SierrasMarginales thrust sheetaccounts for 3.5 km of shortening,and the Montsec thrust and related imbricatesare associatedwith 5 km of shortening.Topographic slopeis 0.5ø anddip of the basaldtcollement is 4ø. (b) Plot of basement subsidence relative to 55 ø Ma basement. Note that significant basement subsidence is inferred to have occurred between36 and30 Ma. Seetext for discussion.(c) Shortening recovered from restoration of structures active after 28 Ma and between28 and30 Ma. Patternsand stratigraphicunits are the sameas Figure 6. 250 MEIGSAND BURBA•: GROWTHOF THE SOUTHPYRENEAN OROGENIC WEDGE

(Figures6, 8, and9). Although0.5-1 km of materialhave been removedfrom above the Tremp-Graus basin since 28 Ma (Figure 6), the taper is basedon structuraland rock thicknesses. Erosionalremoval of thismaterial and consequent rebound would changeboth the basal dtcollement and surface slope angles, b thetotal taper would remain the same.

28-Ma Cross Section

Removalof unit 3 andrestoration of structuresthat deformit definesthe 28 Ma cross section (Figure 8). Structuresremaining in the sectionreflect shortening prior to 28 Ma. Unit 3 hasbeen strippedoff the section,a fold has been reconstructedin the former site of the Montargull thrust, and a minor amountof shorteningis removedfrom the Barbastro anticline [Meigs et al, 1996; Meigs, 1997], otherwisethe sectionis the sameas in the previouscross section. Restorationof thesestructures recovers 1.2 km of shortening.Note thatthe dip of the basaldetachment, 4 ø, would have changedonly if significantthickening in the TrempGraus basin or shorteningto the southhad occurred.N0 evidenceexists for thisdeformation. The 0ø topographicslope is approximatedby connectinga line from the higheststructural reliefon thenorth end of thesection with the foreland.A wedge taperof 4ø is indicated.This representsa minimumangle given uncertaintyin thesynorogenic relief on theMontsec hanging wall or the Milllt and Canelles anticlines. Whereas the crests of the remaininganticlines project both above and below this line, the top of unit 1 lies entirelybelow. We assume,conservatively, that the eroded crest of these structures reflects crestal erosion that occurredduring formation of the fold. It is inferredthat therat• of erosionwas less than the rate of uplift becauseresistant strata are exposedon the flanks at the surfaceand becauseof their presentstructural relief [DeCelleset al., 1991; Hogan, 1993; Burbankand Vergds,1994; DeCelles and Mitra, 1995].

30-Ma Cross Section

Foldingand faulting of unit 1 resultedfrom a modestmount of shorteningon the southernedge, in the hanging wall, of the SierrasMarginales thrust sheet between 30 and 28 Ma [Meigs, 1997]; the effects of this deformation must be removed to constructthe 30.0 Ma crosssection (Figure 9). Also recoveredis a minor amountof shorteningrelated to. the Barbastro,Candles, and Millfi anticlines. Unit 1 unconformablyoverlies the southern flankof theCanelles anticline and is folded(Figure 4). TheMillti anticlineis interpretedto have been slightly modified duringits

Figure 10. 36.0-Ma reconstruction.(a) Restorationof translation of the SierrasMarginales thrust sheet and minor foldingin the foreland.and the geometryof the thrustbelt just afteractivation of the detachment at the base and accretion of the foreland-basin successionis depicted.Topographic slope is 0.5ø . Averagedip of the basal dtcollement is 3 ø but consists,in detail, of a 4.5ø- reach beneath the deformed thrust belt and a lø-reach beneath the undeformedforeland. (b) Plot of basementsubsidence relative to 55 Ma basement.Note thatonly a moderateamount of basement subsidence is inferred to have occurred between 51 and 36 Ma. See text for discussion. (c) Shorteningrecovered from restorationof structuresactive after 28 Ma, between28 and30 Ma, andbetween 30 and36 Ma Patternsand stratigraphic units arethe same as Figure 6. MEIGS AND BURBANK: GROWTH OF THE SOUTH PYRENEAN OROGENIC WEDGE 251

passageover the footwall ramp cutting the foreland-basin sequencein the subsurface.A total of 3.5 km of shorteningis relatedto structuresthat deformunit 1. The dip of the basal detachment remained constant at 4 ø. Because structural relief createdby folds and faults at the leading edge of the Siexxas Marginales thrust sheet was less than the differential relief producedby displacementof therear of the wedgeup the basal dtcollementbetween 30 and 28 Ma, a positiveslope to the topographicsurface (0.5 ø ) is reconstructed(Figure 9).

36-Ma Cross Section Most of the shorteningthat remainswithin the partially restored section at 30 Ma accumulated between 51 and 30 Ma as a consequenceof emplacementof the SierrasMarginales thrust sheet(Figures 9, 10, and 11) [MartœnezPeF•a and Pocovœ,1988; Muitoz, 1992; Vergds,1993; Meigset al., 1996]. Translationof theSierras Marginales thrust sheet occurred in twodiscrete stages (Figure7): (1) an initial 14-kin translationfrom its footwallcutoff to approximatelythe base of a footwallramp across the foreland-' basin sequencebetween 51 and 36 Ma (Figures 10 and 11) [Murioz,1992; Vergds,1993; Vergeset al., 1992] and(2) a 29-kin translation to its present position across the foreland-basin sequencebetween 36 and30 Ma (Figures9 and 10) [Meigset al., 1996]. Map, stratigraphic,and geometricalrelationships suggest that translation of the Sierras Marginales thrust sheet (and, by extension,piggy backtranslation of the entire thrustbe10 over the foreland-basinsuccession occurred after 36 Ma andwas complete by 29.5 Ma [Meigs et al., 1996]. Early folding of the foreland- basin succession is also recovered in the 36 Ma restoration (Figure 10). Initiation of slip on the segment of the basal dtcollement at the base of the foreland-basin succession occurred at ~ 36.5 Ma over a broad region of the thrust belt (Oliana [Burbank et aL, 1992b]; Artesa de Segre [Meigs et al., 1996], Figure 3) and was coincident with initial displacementof the Sierras Marginales across the foreland [Meigs et al., 1996]. Given the geometrical resolution of the stratigraphicdata, no internaldeformation within the thrustbelt is recognizedbetween 36 and 30 M a. Topographydeveloped between 51 and 36 Ma was influenced primarily by displacementon the Montsec thrustand translation of the thick stratigraphicsection at the rear of wedgeup the basal dtcollement (Figures 10 and 11). A surface slope of 0.5 ø is indicatedfor this restoration. An average surfaceslope lower

Figure 11. 51.0-Ma reconstruction.(a) The stratigraphicwedge is foreshortened2.8 km by a seriesof folds whoseamplitude and spacingdecrease to the south. Note 18.0 km of shortening between the axial zone and Pont de Suerte thrust sheet and the

oo thrust belt. Note also that the top of the wedge is at sea level. Topographicslope is 0 ø anddip of the basaldtcollement is 4.5ø. (b) Plot of basementsubsidence relative to 55 Ma basement. Note that becausethe topographicslope is 0ø, thewedge taper is a reflection of basementsubsidence driven by tectonic loading between51 and55 Ma. See text for discussion.(c) Shortening recovered from restoration of structures active after 28 Ma, between 28 and 30 Ma, between 30 and 36 Ma, and between 36 and 51 Ma. Patterns and stratigraphicunits are the same as Figure6. 252 MEIGSAND BURBANK:GROWTH OF THE SOUTHPYRENEAN OROGENIC WEDGE

o than0.5 ø is unlikelygiven that the basin-boundingstructures of II E II theAger and Tremp piggyback basins were probably emergent abovethe Middle Eocene depositional surface (Figures 4 and10). Middle Eocenestrata onlap the flanksand are structurallyand topographicallylower than the crests of adjacentbasin-bounding slxucmres.The basald6collement dip at 36 Ma (3.0 ø) is defined by a line connectingthe baseof the rear of the wedgewith the thrusttip (Figure10). Clearly,this is an averagedip anddoes not reflect the fact that the geometryof the basald6collement at the instant of accretion consistsof two domains: (1) a 4.5" domain beneath the deformed wedge developed after deformation between55 and36 Ma and(2) a 1o domainbeneath the accreted, but asyet undeformed,foreland-basin material.

51-Ma Cross Section

The 51 Ma crosssection reflects restoration of shorteningdue to approximately6 kms of displacementon the Montsecthrust and initial displacementof the SierrasMarginales thrustsheet from its footwall cutoff to its 36 Ma position at the baseof the ramp acrossthe foreland-basinsuccession (Figures 10 and11). Folding and thrust displacementassociated with the Montsec thrust and related imbricatesbegan after 55 Ma and continued until 33.5 Ma [Williams and Fischer, 1984; Williams, 1985; Burbank et al., 1992b; Mu•oz, 1992; Vergds, 1993]. Because Ypresian (55-51 Ma [Cande and Kent, 1992]) growth strata depositedon the limbs of a fold now cut by the Montsecthrust (Figures4 and 10) [Misch, 1934; Mutti eta/., 1985, 1988; Martinez Petraand Pocovi, 1988], thrustdisplacement is inferred to have occurredafter initial folding between55 and 51 Ma but before 36 Ma. Palinspasticrestoration of the structurally correlativethrust sheetin the easternPyrenees, constrained by syntectonicstrata, implies a _12 km translationof the Sierras Marginalesthrust sheet between 51 and 36 Ma [Burbanket al., 1992b; Vergdseta/., 1992; Vergds,1993] Fourteenkilometers are estimatedfor this line of section(Figure 11). Distributed deformation across the width of the thrust belt from 55 to 51 Ma is indicatedby growth-stratalgeometries in Lower Eocenemarine stratadeposited in a seriesof piggy back basinspreserved within the thrustbelt [Pocovœ,1978a, b; Mutti et al., 1985, 1988; Meigs, 1997]. These strata demonstratethat folds in the future site of the Montsec thrust and related imbricates, the Milllt and Canelles anticlines, and two smaller foldsnear the leadingedge of the SierrasMarginales thrust sheet developedat this time. Both the Tremp-Grausbasin and therear of the wedgeare treatedas fixed by 51 Ma and carriedforward in time unmodified.This is significantbecause the thicknessin the center of the Tremp-Graus basin and the structural relief represented by its north flank influence significantly

Figure 12. Pre-55.0-Ma reconstruction,a total restorationof figure 6. A) A total of 65.8 km of forelandwardtranslation of slip is recoveredby restoringall major structures.Displacement of the axial zonewedge with respectto the rear of the thrustbelt includes-7 km of backthrust(to the north)displacement. The stratigraphictaper is 4ø. (b) Shortening recovered from restorationof structuresactive after 28 Ma, between 28 and 30 Ma, between 30 and 36 Ma, between 36 and 51 Ma, and between 51 and55 Ma. Patternsand stratigraphic units are the sameas Figure6. MEIGS AND BURBANK: GROWTH OF THE SOUTH PYRENEAN OROGENIC WEDGE 253

reconslxuctionof the surface slope in subsequentrestorations. A geologiccross section (Figure 12). At - 55 Ma, the thrustfront minormount of deformationon thenorth flank of thebasin that propagated --100 km to the south, and internal deformation causedslight modification of therear of thewedge until the Early initiatedby foldingacross the width of thethrust belt (Figures 11 Oligocene[Mellere, 1993] is notincorporated into successively and 13). Becausethe surface of the thrust belt was at or near sea youngersections (Figures 10, 9, 8, and6). Forthe 51 Ma levelbetween 55 and51 Ma (Figures13 and14) weuse the top restoration(Figure 11), a combinationof internal structural of marinestrata deposited at thistime as a proxyfor sea level to thickeningand using the top of Lower Eocene rocks to infera 0ø surfaceslope for thewedge at thisstage. The taper- approximatepaleo sea level, (an inferencesupported by the anglechange from 4ø at 55 Ma to 4.5ø at 51 Ma thereforeresulted observationthat the Lower Eocene is dominated by shallow from tectonicsubsidence of the basementas a consequence, marinefacies [Pocovœ, 1978a, b; Mutti et al., 1985, 1988] allowsa primarily, of initial formation of the axial zone in the hinterland 00topographic slope and 4.5 ø basal dtcollement angle at 51 Ma and,secondarily, from internaldeformation (Figure lib). A to be inferred.

TotallyRestored (Pre-55-Ma) CrossSection A. taper 60 50 40 30 20 Thepre-55-Ma crosssection is a full restorationof the post- 6 28.0-Ma cross section (Figures 6 and 12). Paleocenerocks, comprisingcontinental alluvial and lacustrinestrata, serve as a •5 horizontaldatum for restorationof the stratigraphicunits because of their widespread preservation over the South-Central Unit (Figure4) [Williams,1985; Puigde•abregas et al., 1992;Vergds, 1993].Approximately 20.8 km of shorteningremained in the 51- Ma section(Figure 11): 2.8 km are due to folding within the forelandfold-and-thrust belt and 18.0km are dueto displacement of theaxial zone wedge beneath the rear of the thrust-belt(Figure B. position 12). The restoredposition of the Pont De Suertethrust sheet is dictatedby therestored area of Triassicstrata. A 4ø taperof pre- Paleocene rocks results from this restoration. The total restored 4o4 shorteningof the internal structuresis 75.5 km, whereasthe total forelandwarddisplacement of the trailing pin line is 68.8 km. 120 --i...... t..... i ...... Thissuggests that backthrustdisplacement, toward the hinterland, acrossthe fault separating the foreland thrust-belt from the crystallinecore is _6.7 km. Structuralgeometries are highly simplifiedon the north end of the section, so thesevalues should 60--.!!.... !.__t _1 i be consideredonly a rough estimate of the actual value of i displacement.A limited amountof backthrustdisplacement 40---,...... acrossthis fault is moreconsistent with crustalinterpretations in whichthe total shortening in the axialzone is nearlybalanced by thatin theforeland fold-and-thrust belt [Verggs,1993; Verggset al., 1995]than interpretations in whichshortening in the axial ...... -! !...... i..... I 1...... zoneis greaterthan in the thrustbelt [Mugoz, 1992]. 60 50 40 30 20 Theage of formationof the axial zoneis not well constrained. millions of years Fromstructural considerations alone, restoration of the rear of the •, thrust-frontposition •edgein thisinterval is reasonableif shorteningin theaxial zone (A = accretionary;SS = stablesliding; F = fixed) antiformal-stackduplex is coupledto shorteningin the foreland becausethe structurally highest and oldest horse of theduplex, El trailing-edgeadvance (km) theNougats thrust sheet, lies directly beneath the trailingedge of O taper(et + I•, degrees) theforeland fold-and-thrust belt [Williams and Fischer, 1984; Williams,1985; Mu•oz, 1992;Verggs, 1993; Verggs et al., 1996]. Figure 13. Summaryplot of (a) taperangle (a is the topographic Conglomeratesdeposited over the fault betweenthe axial zone slopeand b is the dip of the basaldetachmen0 and ([•) thrust andtrailing edge of thethrust belt are Upper Eocene to lowermost front and trailing-edgeadvance (in kilometers(kin)). The taper Oligocene,suggesting emplacement of the highesthorses of the history Between 36 and 30 Ma dependson whetherthe accreted duplexbefore -34 Ma [Mellere,1992, 1993; Sudre eta/., 1992]. materialis consideredpart of the wedgeinstantaneously (shaded, solid line) or progressively(bold, dashed line). See text for Thrust-FrontAdvance, Internal Deformation,and discussion. Note that after 30.0 Ma taper decreasesuntil 28 Ma TaperEvolution of the Pyrenees because of internal deformation. Taper rebuilds after 28 Ma becauseof aggradationof material on to the top of the thrust Comparisonof the amount and style of thrust-frontadvance, wedge. Trailing-edgeadvance occurs continuously, although at internaldeformation, and taper of thethrust belt through time is variable rates, throughoutthe deformation. Between 51 Ma and illustrative(Figure 13). Beforethe onset of contraction,a 4 ø 36 Ma, the slopeof the trailing curve is greater than the thrust- stratigraphictaper is givenby the completerestoration of the front advance.curve reflecting concurrent internal deformation. 254 MEIGS AND BURBANK: GROWTH OF THE SOUTH-PYRE•AN OROGENIC WEDGE

60 50 40 30 20 with the thrust front defines an average dip, 3ø , which overestimatesthe dip of thed6collement beneath the foreland and underestimatesthe dip beneaththe northernthrust bell If this projectionmethod is chosen,the taperinstantaneously drops to 3.5ø at 36 Ma (Figure 13). Alternatively,if the dip of d6collementbeneath the deformed thrust belt is used,the taper remainsunchanged at 36 Ma. Explicitdistinction is not purely semantic as it determines how internal deformation affects subsequentchanges in wedge geometry. Although the latter Figure 14. Plot of the surfaceslope (oc,bold dashedline and interpretationimplies that the taper angle smoothlydecreased solidboxes) versus dip of the basald6collement ([3, shaded solid with time, we define the wedge as the averageenvelope whose line and shadedcircles) during each developmentalstage of the toe is the thrust front. Wedge taper drops from 5 ø to 3.5ø after thrustbell SeeFigure 13A for a plot of o; + • with time. Note accretionat 36 Ma and subsequentlyrebuilds (Figure 13). that surface slope (or) varies between 0 ø and 1ø and basal Significant basementsubsidence and a decreasein mean d6collementangle (•) variesbetween 3.5 ø and 4.5ø. Because d6collementangle accompanied deformation after 36 Ma (Figures variationsin surfaceslope and basal dip are not coordinated,it is 10b,9b, and 14). The singlegreatest shortening event during concludedthat they are decoupledand vary independently. thrust-belt construction, the 29-kin translation of the Sierras Marginalesthrust sheet over the foreland,occurred during this interval (Figures 7 and 9). Emplacementof the thrustsheet positive, foreland-sloping topographic surface was first occurred on the hinterland side of the thrust front. Near the thrust establishedafter 51 Ma (Figure 14). Significantstructural relief front, a minor amountof shorteningwas absorbedby foldingin createdin the rear of the wedge becauseof the formationof the the accretedforeland. Several importantimplications arise from Montsec thrust and displacementof the rear of the wedge up the comparisonof the 36 and 30 Ma reconstructions(Figures 9 and basal d6collement created a 0.5 ø surface slope (Figure 10). 10, respectively). Although the dip of the basal d6collement remained constant Surfaceslope remains unchanged between 36 and30 Ma (0.5ø, from 51 until 36 Ma, initial translation of the thrust belt on its Figure 14), which implies that structuralthickening near thefront basal detachment by displacement of the Sierras Marginales of the wedgerelated to eraplacementof the SierrasMarginales thrust southwardfrom its footwall cutoff likely drove continued thrustsheet was balancedby raising of the surfaceof the rearof basement subsidence. At any position in the underthrust the wedgedue to translationalong the d6collement.This allowed basement,the overlying wedge thickened by ~1 km (due to the differentialrelief and the slope betweenfront and rear of the translation) and resulted in _800 m of basement subsidence wedgeto be sustainedbetween 36 and 30 Ma. The existenceof (Figuresl lb and lob). internal deformation on the hinterland side of the thrust front, A steadyincrease in taperangle from 4.5ø at 51 Ma to 5 ø at 36 becauseit is a means building surface slope and recovering Ma (+/- 0.5*) is inferred (Figure 13). The relationshipbetween critical taper, is often cited as circumstantial evidence for the thrust-frontadvance, internal deformation, and taper until 36 Ma applicability of critical-taper wedge models (assumingthe basal can be summarizedas follows (Figures 13 and 14): (1) A thrust- d•collement angle remained unchanged) [Davis et al., 1983; front advanceby accretionat 55 Ma was followed by internal Boyer and Geiser, 1987; Geiser and Boyer, 1987; Woodward, deformationand taper building until 51 Ma. (2) Stable-sliding 1987;Morley, 1988; Lucas, 1989; Dahlen, 1990; Boyer, 1992; thrust-frontadvance until 36 Ma was accompaniedby additional Burbank et al., 1992b; DeCelles et al., 1993; DeCelles, 1994; internaldeformation and developmentof a foreland-slopingupper De Celles and Mitra, 1995]. In this case, taper angle was surfacethat increasedthe taper to 5* by 36 Ma. Note that the 55- recovered largely becauseof a change in the average basal 51 Ma sequenceof events is qualitatively compatible with a d•collement angle from 4.5 ø before 36 Ma to 4 ø at 30 Ma; critical-taperwedge-like evolution (compareFigures 2 and 13), internal deformation served to sustain the 0.5 ø surface slope. but explanationof the taper increaseaccompanying stable sliding Finally, althougha clear tectonicsubsidence of the basementcan during the 51-36 Ma interval requires either a decreasein pore be seen (Figure 9b), the decreased basal d•collement angle fluid pressureor strengthcontrast between material of the wedge suggeststhat displacementof the wedge onto crust with greater and of the basal d6collement. flexural rigidity. This inference is supportedby the fact that The final thrust-frontadvance occurred by accretionat _36 Ma thinningof the lower crustnorthward of the Montsec thrustwas when the detachmentat the base of the foreland-basinsequence observedon the ECORS profile [Muhoz, 1992]. was activated_40 km southof its pr•-36-Ma position(Figure 13) Hangingwall deformationlocalized at the southernedge of the [Meigs et aI., 1996]. Becausethe most forelandwardstructure SierrasMarginales thrust sheet corresponds with a decreasein beganforming at this time, all subsequentshortening occurred on taper anglebetween 30 and 28 Ma (Figures8, 9, and 13). A the hinterland side of the thrust front. Accretion of the foreland moderate amount of shortening was accommodatedby posesa dilemmafor deemingthe 36-Ma,postaccretion wedge reactivationof preexistingthrusts and folds (Figure 8). Like the taper. Whereasa constant-slopeupper surface is definedas the 36-30 Ma window,shortening was concentxatednear the leading line thatconnects the highpoint at the rear of the wedgewith the edge of the thrustbelt. Structuralrelief createdat this time by thrustfront, the basald6collement has strong curvature separating furthergrowth of the Canellesanticline and thrustfauifing at the a 4.5" from a 1" reach beneath the thrust belt to the north and the toeof theSierras Marginales thrust sheet, despite the factthat the uncleformedforeland to the south, respectively (Figure 10). thrustsheets were relatively thin (in somecases less than 300 m Projectionof a line connectingthe baseof the rear of the wedge thick,compare Figures 8 and9), was sufficientto decreasethe MEIGS AND BURBANK: GROWTH OF THE SOUTH PYRENEAN OROGENIC WEDGE 255 differentialrelief between the front and rear of thewedge and Sequentialrestoration and reconstructionof taper angle is resultedina decrease ofthe surface slope to 0 ø. Weinfer that the ambitious. Uncertainties, and potential sources of error, basald•ollement remained constant after 30 Ma (Figures6,8, 9, consequently, in taper reconstruction arise from (1) a and14) recognizing that the amountof structuralthickening discontinuousstructural and stratigraphicrecord, (2) variable alongthe profile for each step varied slightly in detail,causing resolutionof agedata, (3) uncertaintyin absolutevalues of basal minordifferential tectonic subsidence and modification of the dicollementdip and surfaceslope with time, (4) criteria for d•collementangle. Sediments deposited between 30 and28 Ma determiningsurface slope, (5) changesin structuralrelief due to (unit1,Figure 5) arecharacterized bylacustrine facies and rapid the interplaybetween erosional lowering and structuraluplift of sediment-accumulationrates[Meigs eta/., 1996;Meigs, 1997]. anticlinalcrests, (6) isostaticadjustments due to tectonicand Inaddition to structuraldamming by theuplift of basin-margin sedimentaryloading and erosional unloading, and (7) variations structures[Ori andFriend, 1984; Burbank and Beck, 1991a, b; in flexural rigidity of the underthrustingplate. Becauseboth l)eCelleset al.; 1991; Coogan,1992; Tolling et al., 1995; stratigraphicthicknesses and the structuralevolution of major Burbanketa/., 1996], anoverall regional lowering of theaverage structuresare well constrained[Pocovi, 1978a, b; Meigs et al., surfaceslope may have played an importantrole in the 1996;Meigs, 1997], we areconfident that thereconstructed taper developmentof rapidly accumulating lakes in thepiggy back anglesare representative of relativechanges in taperwith time. basins. In contrast,we havevariable confidence in the reliabilityof the A foreland-slopingtopographic surface (1 o) wasreestablished absolutevalues for basald6collement and surfaceslope angles. after28 Ma, whereasthe basal d•collement angle remained For example, in time windows where palcohorizontal unchanged(4ø , Figure14). Minorinternal deformation after 28 stratigraphicdata such as the Lower Eocene marine strata or Ma is reflected by further folding in the foreland and Paleocenestrata used to reconstructthe 51-Ma and pre-55-Ma displacementon the Montargull thrust (Figure 6). Regionally surf•s, respectively,the restorations represent reliably the likely distributedaõgradation and backfilling by the youngestalluvial, surfaceand basalslopes (Figures 11 and 12). In contrast,where fluvial,and lacustrine unit (unit 3; Figure4) is observedon top of the surfaceslope was generalizedas a line connectingthe rear of thethrust belt after 28 Ma (Figure 6) [Mellere, 1992, 1993; thewedge with thethrust front in the spiritof Davis eta/. [ 1983], l&r•ts,1993; Melts et al., 1996;Melts, 1997]. Thisperiod of this envelopedescribes the meandifferential relief but is not an a[õradationreflects trapping of sedimentderived from the envelopeof maximum,minimum, or mean elevation. Differential forelandfold-and-thrust belt and from the axial zone. The structural-reliefapproximations of the surfaceslope, such as the negligiblesurface slope at 28 Ma mayhave facilitated, in part, one we used, demonstratethe coupling between structural widespreaddeposition on top of, ratherthan bypassing of, the evolution and surfaceslope and are an adequateapproximation thrustbelt. Erosionin the axial zoneand development of graded for our pro'poses.Attempts to investigatethe role of topography alepositionalsystems across the fold-and-thrustbelt were in the smacturaldevelopment of thrustbelts need to reconstruct thereforeresponsible for constructionof the 1o surfaceslope after "rue"topographicattributes, that is, mean depositionalslope, 28Ma. Uplift of strata,erosion, and depositionin actively however[Beaumont et aI., 1996]. deformingthrust belts are complexprocesses whose effects vary How doesthe coupling between advance of the thrustfront and accordingto the organization of drainagesystems, the erodability internal deformation within the Pyrerican orogenic wedge ofuplifting strata, and the a•ailability of accommodationspace compareto the couplingpredicted for an •ideal"critical-taper [B•rba•Ica•d Beck, 1991a, b; DeCelleset al., 1991, 1993; wedge? In contrastto an ideal stable-slidingadvance (time Coosan,1992; Barbank and Vcr$•s 1994; D½C½11½•,1994; La•to• et al., 1994; DeCellesand Mitra, 1995; Talti•$ ½tal., intervaltl-t 2, Figure2b), internaldeformation accompanied the 1995;B•rbank er al., 1996]. one stable-slidingadvance between 51 and 36 Ma, causingthe Wecan make a fu'st-orderapproximation of thetopography for surfaceslope to increaseby 0.5ø (Figure 13). Thrust-front advancesby accretionat 55 and36 Ma were eachfollowed by eachof ourtime steps but areunable to characterizethe structural reliefexactly, at specific points in time, on any individual internaldeformation (time intervalt0-tl, Figure 2), althoughthe structure.Preservation of anticlinal crests indicates that the rate styleof internaldeformation and resultant taper angle associated ofuplift was greater than the rate of erosionfor eachindividual with eachdeformational pulse was different (Figures 13 and 14). structure,but a uniquecrestal uplift ratefor eachfold cannot be After 55 Ma, foldingacross the width of the thrustbelt createda defined.Estimating the error introducedby assuminga mean wedgewith a 0ø surfaceslope and 4.5 ø basald•collement. After slope,particularly if local topographyvaried substantially with 36 Ma, wedgetaper shows considerable instability until the end time,is problematic. Topographic slopes estimated for eachstep of deformationafter 28 Ma (Figure14). Translationof a single area minimumand potentially slightly underestimate the true thrustsheet (-29 kin) between36 and 30 Ma createda 4.5ø- tapergiven uncertainty in incrementalstructural relief. It seems taperedwedge. Interestingly,taper recoverednot from an likelythat the durability,or relative resistanceto erosion increasein surfaceslope driven by internaldeformation (surface [DeCetlesand Mitra, 1995],of theupper surface changed only in slopewas 0.5 ø at36 and30 Ma,Figures 10 and9) butbecause of theregions of theCanelles anticline and Montargull thrust where a decreasein thebasal d•ollement anglefrom 4.5ø to 4ø (Figure Triassicevaporites are exposed (Figure 6). Overthe rest of the 14). Translationof an existingtectonic load to the southonto thrustbelt, Mesozoiccarbonates are exposedin the coresof strongercrust may account for thedecreased angle. Surface slope anticlines(Figures 4 and6). It canbe inferred therefore that the dropsto 0ø andtaper is reducedto 4ø at 28 Ma asa consequence rateof erosion across the width of the foreland fold-and-thrust of continuedthickening near the toe of the wedge. Erosionof the beltdid not vary significantly as a consequenceof progressive hinterland probably lowered the surfaceslope of the hinterland unroofmg. after 28 Ma but deliveredmaterial to the thrustbelt. Aggradation 256 MEIGS AND BURBANK: GROWTH OF THE SOUTH PYRE•AN OKOGENIC WEDGE

of hinterland-derived detritus within the thrust belt was sufficient Ma (Figure14), the surface slope increased. When the rate was to createa 1øforelandward sloping upper surface (Figure 6). greaterat the toe, slope angle decreased, for example, between 30 Inability to constrainpore fluid pressuresand variationsin and28 Ma. Whenthe rate that structural relief is createdatthe mechanicalproperties limits the ability to comparethe Pyrenees, toe balancedthat at the rear, the surfaceslope remained or any orogen,with the critical taper-wedgemodel. Elevated unchanged.Overthrusting ofthe foreland between 36 and 30 porefluid pressurelowers the slopeof the criticaltaper envelope. internaldeformation that accounts for nearly40% of thetotal It seemscertain that pore fluids played a rolein the development shorteningand coinciding with translation of thethrust belt onto of this thrust belt as it evolved from a submaxine to a subaerial lower-angleportion of thebasal d6collement, caused a decrease belt with time. More explicit definitionof therole is difficult in in wedgetaper because the topographic slope remained constant as much as very few pore fluid pressuredata are availablefor (Figures9 and10). active subaerialthrust belts [Dav/s et al., 1983], and it is difficult Considerationof the 51-Ma reconstructiondemonstrates the to estimate fluid pressureconditions in ancient thrust belts role of stratigraphyin the spatialpartitioning of deformation [DeCellesand Mitra, 1995]. Qualitatively,a rapid thrust-front duringinitial deformation(Figure 11). A thrust-belt-widefold advancemay elevate pore fluid pressuresand create a self- train, a seriesof folds whosewavelength and amplitudedecrease generatingfeedback loop whereby fault propagationis facilitated systematicallytoward the foreland,characterize the first-formed by and causescontinued generation of pore fluid pressure(D. structure. A systematicdecrease in fold wavelengthand Davis, personalcommunication, 1996). Becausegeneration of amplitudecorresponds with an overallsouthward thinning of the pore fluid pressure in such a scenario is both nonlinear and pretectonicstratigraphic succession (Figure 5). Subsequent nonsteadystate, exact, quantitativecomparison between taper deformationwas localizedat thesesites of initial folding. evolution and structural development of thrust belts is Experimentalfolding of materialswith variousstrength contrasts problematicat best and shouldbe treatedwith skepticism.In confn'mthe importanceof the initial stratigraphicsuccession in contrast,reasonable qualitative inferences of the strengthof the the internalgeometry, partitioning of deformation,and creation of wedge and the basal layer are possible [DeCelIes and Mitra, structuralrelief [Currie et d., 1962; Dixon and Tirrul, 1991;L/u 1995]. and Dixon, 1991;Dixon and Liu, 1992]. Clearly,the pretectonic It seemslikely that the strengthof both the wedge and the stratigraphicsequence acted as a first-ordercontrol on the sitesof basal layer showed little variability with time becauseof the subsequentdeformation. relatively shallow depth of emp!acement, lack of fabric Observationsand their interpretationare aidedby models,such development,and the homogenouscharacter of the material as the critical-taperwedge model, becausethey providea within the wedge (dominated by massive carbonates). A theoreticalframework within which interrelationshipsmay be relatively high strength contrast can be inferred between the explored[Beaumont eta/., 1996]. Our goal has not beento carbonateswithin the wedgeand the evaporiteswithin whichthe determinewhether wedge models are fight or wrongbut rather to basald6collement localized. It has been arguedthat the critical- explore the interplay between internal structural evolution, taperenvelope for thrustbelts aboveevaporite detachments, like surfaceslope evolution,and basal d6collementangle as the the Pyrenees,have extremely low slopes(or + [• =1ø-2ø [Daviset Pyreneanforeland fold-and-thrust belt waseraplaced. In some al., 1983; Davis and EngeIder, 1985; Jaumdand LilIie, 1988; ways, the Pyreneesappear to behavelike a critically tapered Davis and Lillie, 1994]). Clearly the observedmagnitude of the wedge. For example,accretion lowers taper and is followedby a internal deformationwould not have been predictedfrom either period of taper reconstruction.In other ways, the coupling the initial stratigraphictaper or the deformed-statetaper of 4ø and betweeninternal deformation, surface slope development, and 4ø-5ø, respectively[Davis and Engelder, 1985;Jautnd and Lillie, basal d6collement angle are not simply related. Internal 1988; Iazwtonet al., 1994; Davis and LiIlie, 1994]. Furthermore, deformationcaused basementsubsidence without building comparablemounts of internaldeformation are seen in thisstudy surfaceslope. Taper anglewas maintainedand even lowered and on the ECORS profile [Mu•oz, 1992; ¾ergds,1993]. This, becauseinternal deformation was insufficient to buildslope with too, is an unexpectedresult becausestratigraphic wedges with contemporaneoustranslation on to strongercrust. Given thatthe greater initial taper, such as along the ECORS transect(5 ø, mechanical characteristics of foreland fold-and-thrust belts [Mu•oz, 1992]), should have systematically less internal divergefrom thoseassumed in the critical-taperwedge model deformation because initially they are close to critical taper [Bombolakis,1994], additionalhigh-resolution studies of the [lawton eta/., 1994]. spatial,temporal, and geometricalevolution of thrustbelts are The internaland externalevolution of the Pyreneesbetween 55 neededto understandthe couplingbetween their internaland and 28 Ma were coupled,but the topographicslope and the basal externaldevelopment. Modeling strategies that considerlarger, d6collementangles were decoupled (Figure 14). Rather than plate boundary scale interactions of denudation, mantle revolvingaround maintenance of a fundamentaltaper angle, taper subduction,subduction load, coupling between crust and mantle, angle was primarily controlledby relative ratesof creationof crustaldelamination, and shear zone strength are likely to provide structuralrelief, the flexural rigidity o•' the lower plate, and the greaterinsight into orogenic processes than studies that focus on taperof thepretectonic stratigraphic wedge. The relativerate of oneelement (the thrust belt) of a larger,coupled system [Royden changein structuralrelief at the toe versusthe rear of the thrust andBurchfiel, 1989; Willett, 1992; Vergds, 1993; Witlett et al., belt determinedthe averagesurface slope, except after 28 Ma 1993;Beaumont et al., 1996]. when aggradationcaused a 1ø increasein surfaceslope. When therate of relief createdby translationof therear of thewedge up Acknowled!onents.Supported by NSFgrant EAR-9304863 and grants the d6col!ement exceeded the rate of relief created by fromthe Donorsof the PetroleumResearch Fund (ACS-28270-AC) to deformationat the toe of the wedge,such as between51 and 36 D.W.B.A.J.M. wassupported by grantsfrom A.A.P.G., G.S.A., and IVIEIGS AND BUKBANK: GRO•H OF THE SOUTH PYRENEAN OROGENIC WEDGE 257

SigmaXiand a ShellFoundation Fellowship. Thestereonet program was andfield and interpretation assistance. Dan Davis and Franqoise Rotate providedbyR. Allmendinger. Janme Vergts, Josep Anton Mu•oz, and are thankedfor their positiveand constructivereviews on behalf of KenFowler are thanked for skepticismand encouragement, discussions, tectonics.

References ' Aaadtn,P., L. Calxera,F. Colombo,M. Byrne,T., and$. Hibbard,Landward vergence R. B. Cole,P. Srivastava,N. Pequera,and Marzo, and O. Riba, •Syntectonic in accretionaryprisms: The role of the D. Pivnik,D. A., Kinematichistory of a intraformationalunconformities in alluvial backstopand thermal history, Geology, 15, forelanduplift from Paleocenesynorogenie fandeposits, eastern Ebro Basin margins 1163-1167, 1987. conglomerate,Beartooth Range, Wyoming (NESpain), in ForelandBasins, edited by Cande, S.C., and D. V. Kent, A new and Montana. Geol. Soc. Am. Bull., 103, P.A. Allenand P. Homewood,pp. 259-271, geomagneticpolarity time scalefor the Late 1458-1475, 1991. BlackwellScientific, Oxford, 1986. Cretaceousand Cenozoic, J. Geophys.Res., DeCdies,P. G., H. T. Pile,and J. C. Coogan, Beaumoat,C., P. J. J. Kamp,J. Hamilton,and 97, 13917-13951,1992. Kinematichistory of the Meadethrust based ?. Fullsack,The continentalcollision zone, Chappie,W. M., Mechanicsof thin-skinned on provenaneeof the Bechlerconglomerate SouthIsland, New Zealand:Comparison of fold-and-thrustbelts, Geol.Soc. Am. Bull., at Red Mountain,Idaho, Sevierthrust belt, geodynamicalmodels and observations, J. 89, 1189-1198, 1978. Tectonics,12, 1436-1450, 1993. Geophys.Res., 101, 3333-3359, 1996. Choukroune,P., andM. Seguret,Tectonics of Dixon,J. M., andS. Liu, Centrifugemodelling Bombolakis,E.G., Applicability of critical- the Pyrenees:Role of compressionand of thepropagation of thrustfaults, in Thrust wedgetheories to forelandbelts, Geology, gravity,in Gravityand Tectonics, edited by Tectonics,edited by K. R. McClay,pp. 53- 22,535-538, 1994. K• De Jongand 1C Scholten, pp. !4!-156, 70, Chapmanand Hall, New York, 1992. Boyer,S. E., Geometric evidence for JohnWiley, New York, 1973. Dixon, $. M., and R. Tirrul, Centrifuge synchronousthrusting in the southern Choukroune,P., and ECORS Team, The modelling of fold-thrust structuresin a Albertaand northwestMontana thrustbelts, ECORS Pyreneandeep seismicprofile tripartitestratigraphic succession, J. Struct. in Thrust Tectonics, edited by K. R. reflection data and the overall structure of Geol., 13, 3-20, 1991. MeClay,pp. 377-390, Chapmanand Hall, an orogenicbelt, Tectonics,8, 23-29, 1989. Farrell, S. G., G. D. Williams, and C. D. New York, 1992. Colletta, B., J. Letouzey,,R. Pinedo,J. F. Atkinson, Constraints on the age of Boyer,S. E., and P. A. Geiser,Sequential Ballard,and P. Bale,Computerized X-ray movement of the Montseeh and Cotiella developmentof thrustbelts: implications tomographyanalysis of sandboxmodels: Thrusts,south central Pyrenees, Spain, J. for mechanicsand crosssection balancing, Examplesof thin-skinnedthrust systems, Geol.Soc. London, 144, 907-914,1987. Geol.Soc. Am. Abstr.Progs., 19, 1•. 597, Geology,19, 1063-1067,1991. Geiser, P. A., The role of kinematics in the 1987. Coogan, J. C., Structural evolution of constructionand analysis of geologicalcross Burbank,D. W., and R. A. Beck, Early piggybackbasins in the Wyoming-Idaho- sectionsin deformedterranes, Spec. Pap. Plioceneuplift of the Salt Range:Temporal Utah thrustbelt, Mem. Geol. Soc.Am., 179, Geol.Soc. Am., 222, 47-76, 1988. constraintson thrust wedge development, 55-81, 1992. Geiser,P. A., andS. E. Boyer,The paradox of northwestHimalaya, Pakistan, Spec. Pap. Ctm•e,J. B., H. W. Patnode,and 1L P. Trump, the orogeniewedge and a modelfor crustal Geo•Soc. Am., 232, 113-128, 1989. Developmentof foldsin sedimentarystrata, recycling,GeoL Soc. Am. Abstr. Programs, Burbank,D. W., and 1L A. Beck, Models of Geol.Soc. Am. Bull., 73, 655-674, 1962. 19, 674, 1987. aggradationversus progradation in the Dahlen,F. A., Criticaltaper model of fold-and- Hogan, P. J., Geochronologic,tectonic, and HimalayanForeland, Geol. Rundsch.,80, thrustbelts and accretionary wedges, Annu. stratigraphicevolution of the southwest 623-638, 1991. Rev.Earth Planet.Sci., 18, 55-99, 1990. Pyrericanforeland basin, northern Spain, Burbank,D. W., andR. A. Beck,Rapid, long- Dahlen, F. A., and $. Suppe, Mechanics, Ph.D. thesis,219 pp., Univ. of Southern term rates of denudation,Geology, 19, growth,and erosion of mountainbelts, Spec. California,Los Angeles, 1993. 1169-1172,1991. Pap. GeoLSoc. Arn., 218, 161-178, 1988. Jaumt, S.C., and R. J. Lillie, Mechanicsof the Burbank,D. W., andJ. Vergis, Reconstruction Davis,D. M., andT. Engelder,The roleof salt Salt Range-Potwar Plateau, Pakistan: A of topographyand related depositional in fold-and-thrustbelts, Tectonophysics, fold-and-thrustbelt underlain by evaporites, systemsduring active thrusting, J. Geophys. 119, 67-89, 1985. Tectonics,7, 57-71, 1988. Res.,99, 20,281-20,297,1994. Davis, D. M., and R. J. Lillie, Changing Lawton,T. F., S. E. Boyer,and $. G. Schrnitt, Burbank,D. W., C. Puigdef'abregas,andJ. A. mechanical responseduring continental Influenceof inheritedtaper on structural Muioz, The chronology of Eocene collision:Active examples from theforeland variability and conglomeratedistribution, tectonicsand stratigraphicdevelopment of thrustbelts of Pakistan,J. Struct. Geol., 16, Cordilleran fold and thrust belt, western the eastern Pyrenean foreland basin, 21-34, 1994. UnitedStates, Geology, 22, 339-342,1994. northeastSpain, GeoL Soc. Am. Bull., 104, Davis, D. M., $. Suppe,and F. A. Dahlen, Liu, S., andJ. M. Dixon,Centrifuge modelling 1101-1120,1992a. Mechanics of fold-and-thrust belts and of thrustfaulting: Structural variation along Burbank,D.W., J.Vergts, J. A. Mufioz,and P. accretionarywedges, J. Geophys.Res., 88, strikein fold-thrustbelts, Tectonophysics, Bentham,Coeval hindward- and foreward- 1153-1172, 1983. 188, 39-62, 1991. imbricatingthrusting in the south-central DeCelles, P. G., Late Cretaceous-Paleocene Losantes,M. E., Aragon,s,X. Berfistegui,and Pyrenees,Spain: Timing and ratesof synorogenicsedimentation and kinematic C. Puigdef&bregas,Mapa Geolbgic de shorteningand deposition, GeoL Soc. Am. historyof the Sevierthrust belt, northeast Catalunya, 1:250,00 scale, Institut Bull.,104, 3-17, 1992b. Utah and southwestWyoming, Geol. Soc. CartogrMiede Catalunya,Barcelona, Spain, Burbank,D.W., A. J. Meigs,and N. Brozovic, An• Bull., 106, 32-56, 1994. 1989. Interactionsof growingfolds and coeval DeCelles,P. G., and G. Mitra, Historyof the Lucas,S. B., Structuralevolution of the Cape depositionalsystems, Basin Res., 8, 199- Sevierorogenic wedge in termsof critical Smith thrust belt and the role of out-of- 223,1996. tapermodels, northeast Utah and'southwest sequencefaulting in the thickeningof Byrne,D. E.,W.-H. Wang, and D. M. Davis, Wyoming,Geol. Soc. Am. BulL, 4, 454-462, mountainbelts, Tectonics,8, 655-676, 1989. •anical roleof backstopsinthe growth !995. Martinez, A., J. Vergts, and J. A. Mufioz, offorearcs,Tectonics, !2, 123-144,1993. DeCelles,P. G., M. B. Gray,K. D. Ridgway, Sequenciasde propagaci6ndel sistemade 258 MEIGS AND BURBA•: GROWTHOF THE SOUTHPYRENEAN OROGENIC WEDGE

cabalgamientosde la terminaci6ndel manto activethrust sheets, Geology, I2, 475-478, Talling,P. J., T. F. Lawton,D. W.Burba• del Pedraforca y relaci6n con los 1984. andR. S. Hobbs,Evolution of latest conglomeradossinorogtnicos, Acta Geol. Platt,$. P., Dynamicsof orogeniewedges and Cretaceous-Eocenenonmarine deposyste• Hisp., 23, 119-128, 1988. the uplift of high-pressuremetamorphic inthe Axhandle piggyback basin of central Martinez Pefia, M. B. and Pocovl, A., E1 rocks,GeoL Soc. Am. Bull., 97, 1037-1053, Utah,Geol. Soc. Am. BulL, 107, 297-315, amortiguamientofrontal de la estructurade 1986. 1995. la coberterasurpirenaica y surelation conel Pocovi,$., Estudiogeo!ogico de las Sierras Teixell,A., Estructuraalpina en la transretail anticlinal Barbastro-Balaguer,Acta Geol. Marginales Catalanas (Prepirineo de de la terminaeionoccidental de la Zoaa Hisp.,23, 81-94, 1988. Letida), Ph.D. thesis,218 pp., Univ. of AxialPirenaica, Ph.D. thesis, 252 Meigs, A. $., Sequential development of Barcelona,Barcelona, 1978a. Univ. de Barcelona,Barcelona, 1992. selectedPyrenean thrust faults, J. Struct. Pocov/,J., Estudiogeologico de las Sierras Vergts,L, Estuditecttnie del vessant sud

Geol.,in press1997. Marginales Catalanas (Prepirineo de Pirineu oriental ß i central: Evoluei6 Meigs,A. J. andBurbank, D. W., Thru• sheets Lerida),Acta Geol. Hisp., I3, 73-79, 1978b. einem•xicaen 3D,Ph.D. Thesis, 203 as a proxy for whole orogens:Testing the Price,tLA., The mechanicalparadox of large Universitatde Barcelona, Barcelona, 1993. criticaltaper wedge model with an example overthrusts,Geol. $oc. An• Bull.,I00, 1898- Vergts,L, andJ. A. Mufioz,Thrust sequenc• from the SpanishPyrenees, Geol. Soc.Am. 1908, 1988. in the southerncentral Pyrenees, Bull. Soc. Abstr.Programs, 25, A-230, 1993. Puigdef,•tbregas,C., J. A. Mufioz, and J. Geol.Ft., 8, 399-405, 1990. Meigs, A. J., $. Vergts, and D. W. Burbank, Vergts, Thrusting and foreland basin Vergts,J., J. A. Mufioz,and A. MartSnez, Ten-million-yearhistory of a thrust sheet, evolution in the Southern Pyrenees, in SouthPyrenean fold-and-thrust belt: Role Geol. Soc.Am, Bull., 12, 1608-1625, 1996. ThrustTectonics, edited by K. R. MeClay, of foreland evaporitic levels in thrust Mellere, D., I Conglomeratidi Poblade Segur: pp. 247-254,Chapman and Hall, New York, geometry,in ThrustTectonics, edited by StratigrafiaFisiea e Relazioni Tettoniea- 199Z R. McClay, pp. 255-264, Chapmanand Sedimentazione,Ph.D. thesis., 203 pp., Riba, O., Syntectonieuneonformifies of the Hall, New York, 1992. Univ. of Padova•Padova, Italy, 1992. Alto Cardchef,Spanish Pyrenees: A genetic Vergts,L, H. Mill,in,E. Roca,•. A. Mu•oz, Mellere, D., Thrust-generated, back-fill interpretation,Sediment. Geol., 15, 213-233, M. Marzo, J. Cirts, T. denBezemer, stackingof alluvial fan sequences,south- 1976. Zoetemeijer,and S. Cloefingh,Eastern centralPyrenees, Spain (La Pobla de Segur Roure,F., P. Choukroune,X. Berastegui,J. A• Pyreneesand related foreland basins: he-, Conglomerates),in TectonicControls and Mu•oz, A. Villien, P. Matheron,M. Bareyt, syn-, and post-collisional crustal-scale Signatures in Sedimentary Successions, M. Seguret,P. Camara, and J. Deramond, cross-sections,Mar. Pet. Geol., I2, 903- editedby L. E. Frostickand R. L Steel,pp. ECORS deep seismic data and balanced 915, 1995. 259-276,Blackwell Sci., Cambridge,Mass., cross sections: Geometric constraints on the Vergts,$., D. W. Burbank,and A. $. Mdgs, 1993. evolutionof the Pyrenees,Tectonics, 8, 41- Unfolding: An inverse approachto fold Misch,P., Der Bau der MittlerenSudpyren•en, 5O, 1989. kinematics,Geology, 24, 175-178,1996. Abh. Geselsch. GOttingen Math. Phys. Royden, L., and B.C. Burehfiel, Are Willett, S. D., Dynarnieand kinematicgrowth Kl. I3, 1597-1764, 1934. systematicvariations in thrust belt style and changeof a Coulombwedge, in Thrust Morley, C. K., Out-of-sequence thrusts, relatedto plate boundaryprocesses? (The Tectonics,edited by K. tL McClay,pp. 19- Tectonics,7, 539-561, 1988. western Alps versus the Carpathians), 32, Chapmanand Hall, New York, 1992. Mulugeta, G., Modelling the geometry of Tectonics,8, 51-61, 1989. Wil!ett, S., C. Beaumont, and P. Fullsack, Coulombthrust wedges, J. Struct.Geol., 10, S•ez, A., J. Vergts, L $. Pueyo,•. A. Mufioz, Mechanical models for the tectonics 847-859, 1988. and P. Burquets,Eventos evaporiticos doubly vergent compressional orogens, Mulugeta,G. andH. Koyi, Three-dimensional pale6geneosen la cuenea de antepa/s Geology,2I, 371-374, 1993. geometryand kinematicsof experimental surpireaaiea: Causas clin•ttieas- eausus Williams, G. D., Thrust tectonics in the south piggybackthrusting, Geology, I5, 1052- tecttnieas?, in Ia'bro-Gu•a Excursion 5 Ii centralPyrenees, J. Struct.Geol., 7, 11-17, 1056, 1987. CongresodeI GrupoEspa•ol del Terciaro, 1985. Mufioz, J. A., Evolution of a continental editedby F. Colombo,pp. 85, Vie, Spain, Williams, G. D., and M. W. Fischer, A collision belt: ECORS-Pyreneescrustal 1991. balancedcross-section across the Pyrenean balancedsection, in ThrustTectonics, edited Senz, J.G. and M. Zamorano, Evoluci6n orogeniebelt, Tectonics,3, 773-780,1984. by K.R. MeClay, pp. 235-246, Chapman teet6niea y sedimentaria durante el Woodward,N. B., Geologicalapplicability of and Hall, New York, 1992. Priaboniensesuperior-Mioceno inferior, en critical-wedgethrust-belt models, Geo/. $oc. Mutti, E., J. Rosell,G. P., Allen, F. Fonnesu, el frente de eabalgamientosde !as Sierras Am. Bull., 99, 827-832, 1987. andM. Sgavetfi,The EoceneBaronia tide Marginalesoccidentales, Acta Geol. Hisp., dominateddelta-shelf system in the Ager 27, 195-209, 1992. Basin, in IAS 6th Regional Meeting Sim6, A, and C. Puigdef/ibregas,Transition D. W. Burbank, Departmentof Earth Excursi• Guidebook,edited by M.D. Mila from shelfto basinon an activeslope, upper Sciences,University of SouthernCalifornia, and L Rosell, pp. 579-600, Intl. Assoc. Cretaceous,Tremp area, southern Pyrenees, Los Angeles, CA 90089-0740. (eranil: Sediment.,Lerida, Spain, 1985. in 6th European Regional Meeting [email protected]) Mutti, E., M. Seguret, and M. Sgavetti, ExcursionGuide-book, pp. 63-108, Intl. A. J. Meigs,Division of Geologicaland Sedimentation and deformation in the Assoc.of Sediment.,Lleida, Spain, 1985. PlanetarySciences, California Institute of Tertiarysequences of thesouthern Pyrenees, Stockreal, G. S., Modeling of large-scale Technology,Pasadena, CA 91125.(eranil: in American Association of Petroleum accretionary wedge deformation, $. [email protected]. edu) Geologists Mediterranean Basins Geophys.Res., 88, 8271-8287,1983. ConferenceFieM Trip Guide,1-153, Nice, Sudre, L, et al., La bioehronologie France,1988. mammaliennedu pa16og•neau Nord et au (ReceivedMay 29, 1996; Off, G. G., and P. F. Friend, Sedimentary Suddes Pyrtnies: Etat de la question,C. R• revisedNovember 12, 1996; basinsformed and earfled piggyback on Acad. Sci. Sbr.II, 314, 631-636, 199Z acceptedNovember 20, 1996.)