First direct observation of coseismic slip and seafloor rupture along a submarine normal fault and implications for fault slip history

(1)

HAL Id: insu-01557932

https://hal-insu.archives-ouvertes.fr/insu-01557932

Submitted on 6 Jul 2017

HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

First direct observation of coseismic slip and seafloor rupture along a submarine normal fault and implications

for fault slip history

Javier Escartin, Frédérique Leclerc, Jean-Arthur Olive, Catherine Mevel, Mathilde Cannat, Sven Petersen, Nico Augustin, Nathalie Feuillet, Christine

Deplus, Antoine Bezos, et al.

To cite this version:

Javier Escartin, Frédérique Leclerc, Jean-Arthur Olive, Catherine Mevel, Mathilde Cannat, et al..

First direct observation of coseismic slip and seafloor rupture along a submarine normal fault and

implications for fault slip history. Earth and Planetary Science Letters, Elsevier, 2016, 450, pp.96-

107. �10.1016/j.epsl.2016.06.024�. �insu-01557932�

(2)

First direct observation of coseismic slip and seaﬂoor rupture along a submarine normal fault and implications for fault slip history

Javier Escartín

^a^,^∗

, Frédérique Leclerc

^b

, Jean-Arthur Olive

^c

, Catherine Mevel

^a

,

Mathilde Cannat

^a

, Sven Petersen

^d

, Nico Augustin

^d

, Nathalie Feuillet

^a

, Christine Deplus

^a

, Antoine Bezos

^e

, Diane Bonnemains

^a

, Valérie Chavagnac

^f

, Yujin Choi

^a

,

Marguerite Godard

^g

, Kristian A. Haaga

^h

, Cédric Hamelin

^h

, Benoit Ildefonse

^g

, John W. Jamieson

^d

, Barbara E. John

ⁱ

, Thomas Leleu

^f

, Christopher J. MacLeod

^j

, Miquel Massot-Campos

^k

, Paraskevi Nomikou

^l

, Marine Paquet

^a

,

Céline Rommevaux-Jestin

^a

, Marcel Rothenbeck

^d

, Anja Steinführer

^d

, Masako Tominaga

^m

, Lars Triebe

^b

, Ricard Campos

ⁿ

, Nuno Gracias

ⁿ

, Rafael Garcia

ⁿ

, Muriel Andreani

^o

,

Géraud Vilaseca

^a

aInstitutdePhysiqueduGlobedeParis(CNRSUMR7154),Paris,France bEarthObservatoryofSingapore,Singapore

cLamont-DohertyEarthObservatory,Palisades,NY,UnitedStates dGEOMARHelmholtzCentreforOceanResearch,Kiel,Germany eUniversityofNantes,Nantes,France

fGéosciencesEnvironnementToulouse(CNRSUMR5563),Toulouse,France gGéosciencesMontpellier,UniversityofMontpellier,France

hUniversityofBergen,Bergen,Norway

iUniversityofWyoming,Laramie,WY,UnitedStates jCardiffUniversity,Cardiff,Wales,UnitedKingdom kUniversityoftheBalearicIslands,PalmadeMajorca,Spain lUniversityofAthens,Athens,Greece

mTexasA&MUniversity,CollegeStation,TX,UnitedStates nUniversityofGirona,Girona,Spain

oUniversityofLyon,Lyon,France

a r t i c l e i n f o a b s t ra c t

Articlehistory:

Received2January2016

Receivedinrevisedform3May2016 Accepted16June2016

Availableonline30June2016 Editor:A.Yin

Keywords:

submarinefault surfacerupture earthquake faultslip neotectonics microbathymetry

Properlyassessingtheextentandmagnitudeoffaultrupturesassociatedwithlargeearthquakesiscritical for understandingfaultbehavior and associatedhazard. Submarinefaultscan triggertsunamis, whose characteristicsaredefinedbythegeometryofseafloordisplacement,studiedprimarilythroughindirect observations(e.g.,seismiceventparameters,seismicprofiles,shipboardbathymetry,coring)ratherthan direct ones. Using deep-sea vehicles, we identify for the first time amarker of coseismic slip on a submarine faultplane along the RoseauFault (LesserAntilles), and measure its verticaldisplacement of∼⁰.9 m in situ.We alsomaprecentfissuringand faulting ofsediments onthehangingwall, along

∼^{3 km} ôf^rupture ⁱⁿ ^close ^proximity ^to ^the ^fault’s ^base, ând ^document ^the reactivation of erosion and sedimentation withinand downslope ofthe scarp.These deformationstructureswerecaused by the 2004 Mw 6.3LesSaintes earthquake,whichtriggeredasubsequenttsunami.Theircharacterization informsestimatesofearthquakerecurrenceonthisfaultandprovidesnewconstraintsonthegeometry offaultrupture,whichisbothshorteranddisplays locallylargercoseismicdisplacementsthanavailable model predictions thatlack fieldconstraints. Thismethodology ofdetailedfield observations coupled with near-bottomgeophysicalsurveying can bereadilyapplied tonumerous submarinefaultsystems, andshouldproveusefulinevaluatingseismicandtsunamigenichazardinallgeodynamiccontexts.

©²⁰¹⁶^TheÂuthors.^Published^byÊlsevier^B.V.^Thisîsânôpenâccessârticleûnder^the^CC^BY^license (http://creativecommons.org/licenses/by/4.0/).

*

Correspondingauthor.

E-mailaddress:escartin.javier@gmail.com(J. Escartín).

http://dx.doi.org/10.1016/j.epsl.2016.06.024

0012-821X/©²⁰¹⁶^TheÂuthors.^Published^byÊlsevier^B.V.^Thisîsânôpenâccessârticleûnder^the^CC^BY^license(http://creativecommons.org/licenses/by/4.0/).

(3)

1. Introduction

Seismically active faults routinely experience ruptures that propagateall thewayto theEarth’s surface.The earthquakesur- faceruptureischaracterizedbyitsextent,natureanddisplacement pattern.It resultsfromacomplexcombinationofparameters,in- cluding fault geometry and segmentation, surface geology, fault slip history, or the dynamics and geometry of the seismic rupture at depth. Past and modern coseismic fault ruptures can be readily observed in subaerial environments through ﬁeld observations,high-resolution microtopography,aerial photography, and satelliteimagery, amongother methods(e.g.,Avouacetal.,2014;

DePoloetal.,1991; Klinger,2005).Thesedetailedstudieshelpcon- strain the seismogenichistory ofthese faults, anddocument the accommodationandreleaseofstressandstrain(Bhatetal.,2007;

King,2005; RockwellandKlinger,2013).

Surfacefaultrupturestudieshavebeencriticaltoestablishscal- inglawsbetweenearthquake magnitudeandobservablessuchas maximumoraveragefaults, displacementandrupturelength(e.g., Papazachos etal., 2004; Wesnousky, 2006, 2008). These comple- ment similar scaling laws based solely on subsurface ruptures (Scholz et al., 1986; Wells and Coppersmith, 1994). These seis- micscalinglawsprimarilyrelyonstrike-slipearthquakes,yetmay dependstronglyonfaulttype(StockandSmith,2000).Inthiscon- text,normalfaultrupturesappearlargely under-represented.Ow- ingtothecomplexityoffaultrupturepropagationandsubsurface geologicalcontrols,thesurfaceruptureistypicallyshorterthanthe subsurfaceone,andthe datausedtoconstrainthesescalinglaws sufferfromsigniﬁcantscatter(e.g.,WellsandCoppersmith,1994).

Predictionsofearthquakepropertiescanthus varybyanorderof magnitudeormore,dependingonthefaultruptureparametercon- sidered.

While∼^70%ôf ^theÊarth’s^seismicity ôccursôffshore,^detailed fault surface rupture observations and associated studies to date are exclusively subaerial. Submarineruptures can also be associated withtsunami hazard, a threat that has proven to be much moredamagingforcoastalareasthantheearthquakesthemselves (Maranoetal.,2010).Seafloorobservationsofcoseismicfaultrup- turesarethusneededtodeterminewhethersubaerialobservations and associated scaling laws can be extrapolated to the marine environment.Seafloorrupture observationsare effectivelylacking owingtolimitationsimposed bytheenvironment, thetechnolog- ical requirements to conduct detailed fieldwork, and the lack of observations prior to seismic events to identify and characterize subsequentcoseismicruptures.

Submarineearthquake geologyand history is typically recon- structedfromsedimentaryrecordscontainingturbidities coredoff- shore andprimarily along active margins (e.g.,Beck etal., 2012;

Goldfinger,2011,andreferences therein).High-resolution seismic imagingofco-seismicallydepositedunitscanprovideinformation onearthquakeactivityatgreater depths(and thereforeonlonger timescales) than those probed by coring. However, these tech- niquesdonotcharacterizetheruptureinducedbyindividualearth- quakes.So far, shipboard geophysicalmethods (bathymetry, side- scan sonar images) have been used to map recently reactivated faults(Armijoetal.,2005; Cattaneoetal.,2012; Eliasetal.,2007), andto evaluate seafloor vertical displacement inthe case ofex- tremelylargeeventssuchastheTohoku-Okiearthquake(Fujiwara etal.,2011; Kodaira etal., 2012). Seafloorobservationshavealso identifiedpossiblesubmarinecoseismicscarps(Armijoetal.,2005;

Matsumotoetal., 2009; Tsuji etal., 2012), butextensiveseaﬂoor characterizations offault rupture extent andslip distribution are stilllacking.

This studypresents high-resolution geophysical data acquired inDecember 2013along theRoseau fault,an active normal fault that produced a Mw 6.3 earthquake in 2004 in the Guadeloupe

archipelago(French WestIndies,Fig. 1).Ourdataanalysisdemon- stratesan unequivocallink betweenobserved deformation structures andthis seismic eventallowing us to characterize thedis- tributionandnatureofthecoseismicfaultruptureattheseafloor, andtoidentifypossiblelinksbetweencoseismicfaultreactivation anderosionalordepositionalprocessesalongthe submarinefault scarp.We alsomeasurethemagnitudeofcoseismicdisplacement at a specific outcrop along the fault. Our results validate model predictions of fault rupture and tsunami generation in the area, andarecomparedwithsubaerialobservationsofnormalfaultrup- ture andassociated scaling laws.We also demonstratethe feasi- bilityofhigh-resolution seafloormapping,andits importance for expandingourunderstandingoffaultruptureanddynamicstothe marineenvironment.

2. Intra-arcactivefaultingandseismicity:LesSaintesGraben (FrenchAntilles)

The LesSaintes graben extends betweenGuadeloupe andDo- minicaIslands(Fig. 1)accommodatinginternaldeformationofthe LesserAntillesarcduetoobliqueplateconvergence(Feuilletetal., 2002,2011a).Thisgrabenshowsalonghistoryofinteractingvol- canic emplacement and faulting, and is bound to the southwest by the∼^{40 km}^long,^NE^dipping^Roseau^Fault^(RF),^whichîs ^seg- mentedintoseveral∼^5–15-km^long^portions^{(Fig. 1B).}Înîts^north- ern part, these are arranged asright-stepping echelons, trending N140^◦E, and crosscutting the seafloor in the vicinity of the Les Saintesarchipelago,whichishighly populatedandamajortourist destination.ThesouthernRoseauFaultsectioninsteadshowsleft- stepping echelons,withan overalltrendofN120^◦E(Leclerc etal., inpress).

ThecumulativefaultscarpheightvariesalongtheRoseauFault trace and peaks at >150 m at its center, coinciding with the most prominent echelon, hereafter termed the Roseau echelon.

Thecumulative scarpdissectstheﬂankofthevolcanicarc, which slopes towards the southwest, and captures sediments from the Les Saintes Islands and adjacent reef platform, which are then channeledalongthebaseoftheRoseauscarp.Thesesediments,to- getherwithdebrisfromthefaultscarp, makeupa>300-mthick layerwithinthehangingwallbasin(Leclercetal.,inpress).

On November 21st, 2004, the Mw 6.3Les Saintes earthquake struck the Guadeloupe archipelago, one of the strongest earthquakes to have occurred on French territory inthe last decades.

Groundshakinguptointensity-VIIIwasfeltonTerre-de-Bas(Cara etal., 2005), triggering landslides andgroundfissuring that seri- ously damaged∼^50%ôf^the ^buildings^(Feuilletêtâl., ^2011a).^The

∼¹⁰^km^deep êpicenter^was ^locatedôffshore, ∼¹⁵^km^SE ôf ^the LesSaintesplateau(Fig. 1B).Relocationoftheaftershocksequence indicatesthatthismainlyextensionalM_w6.3earthquakeruptured theRoseaufault,whichdipsat∼⁵⁵^◦ ^to^the^NEât^depth^(Bazinêt al., 2010; Feuillet etal., 2011a). Thistriggered a tsunami witha maximum run-up of 3.5 m on the nearby coasts of Les Saintes archipelago (Le Friant et al., 2008). Lacking direct observations, ad-hoc models of fault slip and tsunami sources were used to estimateaseafloorrupturewithanalong-strikeextentrangingbe- tween <10 kmand∼¹⁵^km,ând^normaldisplacements ofupto 1m,withanaverageof0.3–0.6 m(Feuilletetal.,2011a;LeFriant et al., 2008). The focal mechanisms of both the main shock and thelargestaftershocksindicateextension intheNEquadrant.The modeled rupturerakeshowsthat themain eventalsoaccommo- dated a minor left-lateral slip component (Feuillet et al., 2011a).

Aftershocksoccurredmainly northoftheoriginal hypocenter,be- neath LesSaintesplateauandalongthe northernechelons ofthe Roseaufault,inresponsetoastaticstressincreaseatthetipofthe rupture (Feuillet et al., 2011a). In particular, the strongest aftershocks,8–12kmdeepandreachingMwupto5.8, didnot trigger

(4)

Fig. 1.A)Shipboardandsatellite-derivedbathymetryoftheAntillesarc,showingtheintra-arcseismicity(epicenters<50kmdepth,thusexcludingthesubductionzone), andfocalmechanisms(ingreenfromtheHarvardCMTcatalog,inblackfromBazinetal.,2010)showingintraplatenormalfaultmechanisms.TheboxcorrespondstoB).

G: GuadeloupeI.;LS:LesSaintesI.;MG:MarieGalanteIsland;D:Dominica;M:Martinique.Theinsetshowsthegeodynamiccontext,withthetrenchparallelcomponentof shearincreasesfrom4to17 mm/yrbetweenMartiniqueandSabaowingtoconvergencebetweentheNorthandSouthAmericanplates(NAM-SAM)andtheCaribbeanone (CAR);seedetailsinFeuilletetal. (2010).B)ShipboardbathymetryofthegrabensystemfromcruiseBathySaintesbetweenLesSaintesIslandsandDominica,boundtothe southwestbytheRoseauFault.TheoutlineoftheAUVandROVmicrobathymetricsurveys(orangeandyellowoutlines,respectively),seaﬂoorphotomosaics(red)alongthe Roseaufault,andlocationofFig. 5(dashedblackbox)areshown.Theseismicityoftheareaandthefocalmechanismsincludingthe21/11/2004LesSaintesearthquake,that rupturedtheRoseauFaulttriggeringalocaltsunami(Feuilletetal.,2011a;LeFriantetal.,2008;Leclercetal.,inpress),arealsoshown.RT:RoseauTrough;CF:CocheFault;

ST:SavaneTrough.(Forinterpretationofthereferencestocolorinthisﬁgurelegend,thereaderisreferredtothewebversionofthisarticle.)

tsunamis,suggestingnosurface break.In themainshockepicen- tralarea,twoaftershocksofMw5and5.3occurredwithintheﬁrst twoweeks,andoccurredatdepthsof10and10.6 km.Onceagain, no tsunamis were triggered by these events,suggesting that the rupturedidnotreachtheseaﬂoor.

The 2004 earthquake is the ﬁrst important seismic event to haveoccurredduringtheinstrumentalperiodalongtheLesSaintes fault system. Historical reports spanning the last 200 years do not mentionother seismiceventsofsimilar size orlargerin this archipelago (see Feuillet et al., 2011b, for a review of historical seismicity in the Lesser Antilles arc). Investigations of the fault trace are thus necessaryto better apprehend the seismic behavior of this fault, and evaluate the associated hazards. Further- more,extension inintra-arcdomains iscommon, withnumerous earthquakes showing normal focal mechanisms (e.g., Fabbri and Fournier,1999; Galgana etal., 2007). Studiesof submarinefaults accommodating this intra-arc deformation can contribute to the understandingof theirdeformation historyandto improvingrisk assessmentfornearbycoastalareas.

3. Datacollectionandanalysis

To study and document the morphology, evolution, and deformation structures associated with a submarine intra-arc nor- malfault,includingrecent(coseismic) deformation,weconducted near-bottom,highresolutiongeophysicalsurveysalongthetraceof theRoseauFault,whichboundstothesouthwestthegrabenlink- ingLesSaintesandDominica,andissub-paralleltothesubduction front(Fig. 1).Wedeployedboththeautonomousunderwatervehi- cle(AUV)Abyss(Geomar,Germany)andtheremotely-operatedve- hicle(ROV)Victor6000(Ifremer,France)duringthe2013ODEMAR

cruise onboard Research Vessel Pourquoi pas? (French Oceano- graphic Cruises, http://dx.doi.org/10.17600/13030070). AUV and ROVvehiclescollectednear-bottomhighresolutionmicrobathyme- trydata(Fig. 1B),anddiveswereplannedusingpreviousshipboard multibeam bathymetry data acquired during the 2010 BATHY- SAINTES cruise (French Oceanographic Cruises, http://dx.doi.org/

10.17600/13030020).ROVsurveysalsoacquiredstillelectronicim- ages to generateseaﬂoorphotomosaics (Prados etal., 2012) over

>180,000 m²ofseaﬂoor(Fig. 1B),aswellashigh-resolutionvideo imagery from a fault surface outcrop to generatehigh-resolution terrain modelswithtexture-mappedimagery(NicoseviciandGar- cia,2013) (Figs. 2Band3).

3.1. Bathymetry

Shipboardbathymetry aroundLesSaintesislands(Fig. 1B)was processed andgriddedat 10m/pixel; cruise anddatadetails are provided elsewhere (Leclerc, 2014; Leclercet al., 2014,in press).

High-resolution,near-bottombathymetrydatawereacquiredusing 200kHzResonSeabat7125multibeamsystemsinstalledonboth underwater vehicles.TheAUV survey,conductedat∼^{70 m above} seafloor, covered an area of9.6 km²,providing detailedmapping of boththe Roseaufault scarp andtheadjacent hangingwalland footwall (Fig. 1B).ROV microbathymetricsurveyswere conducted during twodives flownat∼⁵⁰ând∼^{10 m}âbove^seafloor,^cover- ing 3.1 km² of seafloor (Fig. 1B). Data were pre-processed with a vehicle re-navigation using MBSYSTEM, and manually cleaned andadjustedusingbothMBSYSTEMandFledermausPro.AUVdata were gridded at 2 m/pixel. ROV data were gridded at different resolutions depending onsurvey altitude,from∼25 cm/pixelfor surveysat∼^{50 m above}^the^seafloor,^to∼10 cm/pixelforsurveys

(5)

Fig. 2.A)DetailoftheRoseauFaultscarp(AUVbathymetry)showingthegulliesrunningperpendiculartothefaultscarp,withtheassociateddejectionconesatitsbaseand depositingatthehangingwall.Thefaultscarpshowsstepsduetodifferentialerosionofinternallayeringoffootwallrocks(layeredvolcanicdeposits,bluethinlines).FW:

Footwall;HW:Hangingwall.B)DetailperspectiveviewfromROVbathymetryoftwosub-verticalfaultslipplanesatthebaseoftheerodedRoseauscarp,laterallybound bygulliesandtheassociateddejectioncones.ThepositionofthephotomosaicinC)andtheimagedfaultslipplaneinFig. 3areindicated(reddashedboxes).Scalebaris approximateowingtoperspective.C)Photomosaicshowingaverticalviewofasub-verticalfaultslipplane,withdebrisandrockblocsatthemouthofgulliesateachsideof thefaultslipplane.Themosaicalsoshowsthedegradedfaultscarpabovethefaultslipplane(seealsoB).(Forinterpretationofthereferencestocolorinthisﬁgurelegend, thereaderisreferredtothewebversionofthisarticle.)

at ∼^{10 m above} ^seaﬂoor, corresponding to photomosaicsurveys (seebelow).

3.2.Seaﬂoormosaics

Duringsurveysconductedby the ROVat 10mabove seafloor weusedavertically-mounted,low-lightblack-and-whitecamerato systematicallyacquire electronic still imagesoftheseafloor.Indi- vidualimageswere firstcorrected forillumination andgeometric distortion. Feature-matching betweenimages was used to calcu- latecamera motionandrenavigate theROV vehicle.Images were then projected and blended (Prados et al., 2012) to construct a geographically registered photomosaic with a pixel resolution of

∼^{10 mm}^or^better^(Figs.^1B,^2C,^and^4).

Mosaicswereinterpretedusingboththeavailableobliquevideo imageryfromtheROVandtheonboardscientiﬁcobservationsand descriptions,withmethodsdevelopedinprior studiesusingsimi- lardatasets (Barreyre etal., 2012). Features ofinterest, primarily cracks in the sedimented footwall and at the base and summit ofthefaultscarpweremanuallydigitized fromthegeoreferenced photomosaics.

3.3.Videomosaicsandthree-dimensionalterrainmodelsfromvideo imagery

Usingthehigh-deﬁnitionvideo camerasystemofROVVICTOR, we conducteda systematicsurvey ofa sub-vertical fault outcrop

∼^{20 m}^wideând∼^{5 m high}^{(Figs. 2B}ând^3).^Videoîmagery^was acquiredsub-perpendiculartothefaultplane,alonghorizontaland verticaloverlapping video transects. Video imagery was firstcor- rectedto improveimage quality (illumination, equalization,color shift),andprocessedto createa three-dimensionalterrain model oftheoutcropusingstructurefrommotiontechniques(Camposet al., 2015). The video imagery is texture-mapped to the resulting three-dimensional terrain model, allowing us to directly digitize andmeasurefeaturesandstructuresvisibleintheimagery(Fig. 3).

Theresulting video-derivedterrain modelhasaresolutionthat is better thanthat ofmodelsderived fromacousticmultibeamdata (<10 cm instead of 2 m to a minimum of 10 cm), and allows the mapping of sub-vertical and complex structuresthat cannot beproperlyimagedbyconventionalacousticmethods.

4. Resultsandinterpretation 4.1. MorphologyoftheRoseaufaultscarp

Themicrobathymetryandseaﬂoorimageryhighlightsigniﬁcant erosionandmasswastingalongthecumulativescarpoftheRoseau fault,withdevelopment ofchannelsanddepositionofmaterialat the base of the scarp, generally in the form of dejection cones.

Intensefootwallincisionisbestdevelopedwherethescarpishigh- est,andwhereits slopeaverages30–45^◦ (Fig. 2A).Sectionsofthe faultwithlesserosion,wherethescarpislower, showslopesap- proaching70^◦.Gulliesinthisarea showninFig. 2Aare 10–20m deepandspaced∼50–100 m apart,withsomeofthechannelsco- alescing.Thedebrisshedalongthescarpisemplacedatthemouth ofthesegulliesformingdejectionconesonthehangingwallalong thescarpbase(Figs. 2A,B).Thesestructureshaveareliefofupto 10–20 m,andarenotvisibleintheshipboardbathymetry(Fig. 1B).

Microbathymetry alsorevealsastair-case scarpmorphology,with laterallycontinuousstepsdecimeterstometershigh(bluelinesin Fig. 2A),whichcorrespondtolayeredvolcanicdepositswithin the footwallthathavebeenupliftedbytheRoseauFault(Leclercetal., in press), and enhanced by scarp erosion.The volcanic nature of this layering is also conﬁrmed by a single vesicular lava sample showing fresh plagioclase phenocrysts, recovered during an ROV dive,andby geologicalobservationsatthissamplinglocationand elsewherealongthescarp.

4.2. Verticalfaultslipplanes

Despitepervasivescarp degradation,theROVmicrobathymetry and photomosaics reveal several sub-vertical scarps along the

(6)

Fig. 3.(A)Video-derivedthree-dimensionalterrainreconstructionofanexposed,subverticalfaultslipplane(rightslipplaneinFig. 2B),and(B)videomosaicofthesame outcropwithinterpretationoffeaturesoverlain.Thereconstructionis∼^{20 m}^long,^with^actual^vertical^scaling^shownⁱⁿ^C.^C)^Coseismicdisplacement,measuredfromthe heightofthecoseismicslip(redline),showsamaximumof∼0.9 malongthepreservedfaultslipsurface,thathasamaximumheightof∼3 m(blackline).(D)and(E) showclose-upsofthebaseofthefaultfree-plane(locationshowninA,B,andC),showingaribbonofslipsurfaceexposedduringthe2004seismicevent(coseismicscarp) boundbyalineofsedimentadheredtotheslipplane.(Forinterpretationofthereferencestocolorinthisﬁgurelegend,thereaderisreferredtothewebversionofthis article.)

∼^{3 km} ôf^surveyed^fault^strike ^{(Figs. 2B,}^C). Êach^{of these}^scarps canreachheightsofupto10m,andextendbetween∼^{20 m}ând up to ∼50 m laterally, often lying between the mouths of gullies,andinsome casesshowa mildcurvaturethatinplane-view can havean across-fault amplitude ofa few meters (Figs. 2B, C).

The visual observations from ROV imagery of these sub-vertical planesreveal asmooth, polishedsurface (Figs. 2C and3).We in- terpret thesesurfacesas exposed andpreservedfault slip planes that have not yet been degraded by erosion. This interpretation is based on (1) the position of these slip planes at the base of the main cumulative Roseau Fault scarp, (2) their preferen- tialpreservation betweenthe mouths ofgullies,and henceaway from areas displaying the most intense erosion, and (3) their overall geometry, with a smooth, high angle surface consistent with a fault plane that is breaching the surface at a steep dip (∼^70–80^◦^).

4.3. Faultslipplanetextureandearthquake-relatedvertical displacement

The systematicROVvideo surveyofoneofthepreservedfault planes(Fig. 3)allowsustocharacterizeitsmorphologyandtexture indetail,soastoinvestigateitsnatureand,fortheﬁrsttimeina submarinefaultslipplane,itsseismichistory.Figs. 3AandBshow thevideo-derivedterrainmodelandthevideomosaicofa∼^{20 m} longsectionofoneofthesefaultslipplanes,whichshowsa∼^{3 m} high, subverticalfaultsurface.As inthecaseoftheadjacentfault slipplaneshowninFig. 2C,itislimitedateitherendbytwogul- liesandtheirassociateddejectionconesshowingrocks,debris,and indicatorsofdownslopetransportontheirsurface(Figs. 2B,C).The videomosaicshowsadistinctlight-colored linethatrunsalongthe darkerfaultplanesurface,mimickingthegeometryofthepresent- dayseaﬂoorcontactwiththisfaultplane(Fig. 2D).Whilethefault

(7)

Fig. 4.Shadedmicrobathymetry(A)andseafloorphotomosaics(BandC)fromverticallyacquiredimageryshowingcracksandfissuresdevelopingnearthebaseofthefault scarpanddeforminginduratedsediments.Thephotomosaicsalsoshowsmallscarps(<1 m inverticalrelief)dissectingtheinduratedsediment,anddippingbothtowards theRoseaufault(D)andawayfromit(E)whichcorrespondtoantitheticandsyntheticfaultsinthehanging wallsediments.(Forinterpretationofthereferencestocolorin thisfigure,thereaderisreferredtothewebversionofthisarticle.)

surfacecouldnotbesampledbytheROV,close-uphigh-deﬁnition video imagessuggest that thisclear line is sediment adhered to thefaultsurface,whichiscoherentandindurated.

Thevideoimageryalsorevealsamarkeddifferenceinthetex- tureofthefaultplaneaboveandbelowthisline(Figs. 3C,D).The surfacebelowthislineisdark,smooth, uncracked,andshowsno sedimentaccumulationonsurfaceirregularities.Bycontrast,above thislinethe surface is rough, displays numerouscracks, andhas thinsediment dusting giving an overall lighter color to the fault surface(Figs. 3C–D).Weattributethesetexturaldistinctionstodif- ferentialweathering,owingtoshorterexposuretoseawaterofthe lowerpartofthefaultplanethatislikelyarecentfaultexposure.

We exclude erosion of hangingwall material exposing the previ- ouslyburiedfault scarpasamechanismtoproduce suchfeature, as this process would obliterate the sediment line on the fault plane, which is also located between two gullies and preserved fromerosion.

Thesmoothribbon atthebaseofthe faultscarpislikelypro- ducedbyepisodicsliponthefaultratherthancontinuousaseismic slip. Thisis demonstratedby the difference inthe weatheringof thefault surface above andbelowthesediment line; thesurface doesnotshow a continuous weatheringgradient.This freshsub- marinesurface exposureis similar tothe freshlyexposed coseis- micscarpsalongsubaerialfaultsfollowingseismicevents(Boncio

et al., 2012; Cello et al., 2000; Smith et al., 2011). A coseismic origin, with a postseismic contribution discussed later, is indi- cated by the line of adhered sediment along most of this slip surface, by the geometry of this sediment line mimicking the present-day intersectionofthe fault plane withthe seaﬂoor,and by its weathering texture. Differential weathering of fault plane sectionsexposed throughsuccessive coseismic displacements has alsobeenreportedinnormalsubaerialfaults(Giaccioetal.,2003;

Wiatretal.,2015).

Wemeasureamaximumverticaldisplacementof∼0.9 m,cor- respondingtotheheightoftheearthquake-relatedscarp(Fig. 2E), using thevideo-derivedthree-dimensional terrain modeland the associatedtexture-mappedimagery(Camposetal.,2015).Thisap- parentdisplacementtapersdownto∼⁰.5 m laterallyandtowards theadjacentdejectioncone.A50%variationinfaultslipoverdis- tances ofafew metersis tectonicallyunrealistic.Instead,thede- jectionconemayhavebeenreactivatedduringorfollowingtheob- servedfaultslip,partiallyburyingthescarp.Highersedimentation ratesnear themouthof thegully feedingthiscone wouldresult in an apparent decrease ofdisplacement asobserved in Fig. 3C.

The absenceofdeformation structuresalong gullychannelsadja- centto thetwo fault slipplanesshowninFig. 2Bindicates their eﬃcientobliterationbythereactivationoferosionandsedimenta- tionalong thechannel.The ∼⁰.9 m verticaldisplacementisthus

(8)

Fig. 5.GeometryoftheRoseaunormalfaulttracewithinthesurveyarea(greenline),andassociateddistributionofcoseismicdeformationstructures.Hangingwallcracks (darkred,seeFig. 4),togetherwiththecoseismicdisplacementalongaslipsurface(redtriangle,seeFig. 3)indicatesaminimumfaultruptureof∼3 km(thickredlinewith arrowheads).Thelocationofsubverticalfaultslipplanesthroughoutthestudyareaisalsoindicated(blacksmalltriangles).Therupturemayextendtothesoutheast(dashed linewitharrowhead),andterminatetowardsthenorthwestandwithinourstudyarea(dashedredline).Modelsoffaultdisplacements(thinbluelineswitharrowheads above)basedonearthquakesources(Feuilletetal.,2011a)underpredictthemagnitudeofobservedcoseismicslip,andthelocationofthemaximumdisplacementislocated southeastoftheobserved∼^{0.9 m}^of^coseismic^slip.^(Forinterpretationofthereferencestocolorinthisﬁgurelegend,thereaderisreferredtothewebversionofthis article.)

aminimumestimate oftheactualfaultslipinthissectionofthe fault,astheearthquake-relatedexposedfaultplanemaybe partly buriedbythereactivateddejectioncone.

4.4. Hangingwalldeformationstructures

Seaﬂoor photomosaics and oblique video imagery show nu- merous near-fault damagestructuresat thesedimented hangingwall along the base of the Roseau fault scarp. The hangingwall atthe scarp base shows a dark,indurated, andubiquitously rip- pledsediment layer. Locally,light colored,unconsolidatedrippled sediments concentrate along the base of the fault scarp and in low-lying areas, such as the creases of ripples and bottoms of cracks (Fig. 4B). Along1km ofthe Roseaufault we observethat the indurated sediment is locally cracked and ﬁssured (Fig. 4B), with zones of pervasive deformation hosting dense, coalescing crack networks extending over several tens of meters (Fig. 4C).

The distribution of the cracks and ﬁssures documenting near- fault hangingwalldeformation alongthe fault traceare shownin Fig. 5.

Mostofthephotomosaicsurvey wasconductedoverthehang- ingwall in the immediatevicinity (within 10–20 m) of the fault scarp base(Fig. 1). We surveyed a zone ∼^{150 m} âlong ^the^fault strike and ∼^{110 m} perpendicular to it that displays significant sediment cracking, fissuring, anddeformation. Thesedeformation structureshereextend up to ∼^{100 m}âway ^from ^the^fault ^trace, suggestingthatthehangingwalldamagezoneisbroad,atleastlo- cally.Thelargestfracturesareseveralmeterslong,severaldecime- terswide, andup to30–50 cmdeep, andarereadily identifiable intheROVmicrobathymetry(Fig. 4A).Thesmallerfractures,which canonlybeidentified inthephotomosaics,havelengthsofafew cm,form dense clusters 1–5-mwide, and representdeformation beltsthatextendseveraltensofmeterslaterally(Fig. 4C).

Finally,northofthefissuredhangingwallareaweobservethat the indurated sediment layer within the hangingwall is cut by scarps facing both towards the fault (Fig. 4D), andaway fromit (Fig. 4E).These scarpsare found insediments atdistances ofup to∼^{50 m}^from^the^baseôf^the^Roseau^fault,ândâlong^theêntire lengthofthefaultthatwesurveyedwiththephotomosaics(Figs. 1 and4). Theirrelief isrelatively small, witha maximumofa few tensofcentimeters,andinsomecasesthey arefoundinassocia- tionwithcrackingoftheinduratedsedimentlayer(Fig. 4D).Wein- ferthesescarpstobeassociatedwithbothantithetic(Fig. 4D)and syntheticfaults (Fig. 4E)induced by recentslipalong the Roseau Fault,owingtotheir linearity,their proximitytothe RoseauFault trace, and their link to sediment cracks. While the photomosaic andvisualcoveragearelimited(Fig. 2B),theobservedcracksseem to be co-located with changes in the orientation of the Roseau

Fault trace (Fig. 5), and thus may be associated withinhomoge- neousdeformationofthehangingwallaboveacomplexfaultplane geometryatdepth.Fig. 6showsthemainfeatures describedhere andtheirlinktoco-seismicandpossiblypost-seismicdeformation.

5. Discussionandconclusions

5.1. The2004Mw6.3LesSaintesearthquakeandcoseismicvs.

postseismicdeformation

The 2004Mw 6.3Les Saintesearthquake isthe onlyplausible causeforboththe∼^{0.9 m}^vertical^slipêvent^recordedât^the^fault surface (Fig. 3), andthe hangingwalldeformation observed along the Roseau Fault trace (Fig. 4). First, the sediment line adhered to the fault surface (Fig. 3) that defines the earthquake-induced slip is an important chronological indicator, asit cannot be preserved over long periods of time (hundreds of years or more), owing to deep-sea currents, erosion, and submarine weathering.

Second,thesecurrentseﬃcientlymobilizesedimentsfromtheLes SaintesPlateautowardsthedeepestsectionsoftheRoseauTrough.

Thisalong-faultsedimenttransportisresponsiblefortheaccumu- lationofathicksedimentary sequence(>300 m)atthecenterof the RoseauFault(Leclerc etal., inpress),asindicated bytherip- pledsurface ofboth recentmobileandolderinduratedsediments (Fig. 4). While we have no accurate estimates of sedimentation rates inthearea, a hemipelagicsedimentation rateof0.5mm/yr has been estimated in a similar basin along the Lesser Antilles volcanic arc (Beck et al., 2012), and is thus a minimum for our studyarea.Therefore,thecracksinthehangingwallaswellasthe secondary syntheticandantithetic structuresarenecessarily very recent,andarenotlong-lastingstructures.

We concludethat the vertical offsetmeasured along the fault is thusrelatedto the2004earthquake.As ourobservationswere made only nine years after the 2004 event, the offset may cor- respondprimarilytoacoseismicdisplacement,thoughpotentially withanadditionalcomponentofdisplacementduetopost-seismic deformation (i.e. aftershocks and after-slip). Postseismic slip can represent up to 10–30% ofthe coseismic slip both along certain strike-slip and normal faults (Çakir et al., 2003; Cheloni et al., 2014).Assumingasimilarrangeof10–30%ofpost-seismicslipfor LesSaintes, the postseismicslipmay thereforehavecontributed

∼−⁰.1–0.2 m oftheobserved0.9,withaminimumcoseismicslip of0.7m(Figs. 6AandB).

However,thepostseismicslipinducedbyaftershockscouldalso benegligiblealongtheRoseauFaultattheseaﬂoor;the2004after- shocksequencewasconcentratedatadepthofabout10kmbelow our measurementsite, andthemain aftershocksdidnot produce any tsunami,therefore suggestingno surface break. Although af- terslipcanoccuratthesurfaceduringthepostseismicperiod(e.g.,

(9)

Fig. 6.SketchofthefeaturesobservedalongtheRoseaufaultandtheirdevelopment associatedwithco-seismicand/orpostseismicdeformation.Thefaultslipplaneis preservedbetweenmouthsofgullies,and newlyexposed duringseismicevents.

Coseismicdeformationalsoinducesﬁssuringandfaultingofsedimentsinthehang- ingwallnearfault.Erosionandsedimentationondejectionconesatthemouthof gulliesisreactivatedco- orpost-seismically.Thefaultscarpiserodedovertime, andthefootwallinternalvolcaniclayeringisrevealedbythesteppedmorphology ofthescarp.

SmithandWyss,1968),otherstudiesrevealthatafterslipismainly distributedatthe edgeof,andbelowthecoseismic rupture,such asthat documented following the L’Aquila2009 M_w 6.3normal faultearthquake (Cheloniet al., 2014). Ifthis post-seismic defor- mationpatternappliestotheRoseauFault,afterslipatthelocation ofour observations (Fig. 3) could be either extremelylimitedor absent at the surface. That postseismic deformation here makes only a minor contribution is also supported by afterslip models basedonfriction laws(Marone etal.,1991),which showpropor- tionally more afterslip occurring where surface coseismic slip is smallrelative to the deep coseismic slip. Our data cannot deter- minethepresenceorabsenceofpostseismicslipanditsamplitude here, but we conclude that it represents a proportion (probably lessthan∼^{10 cm)}ôf^the^final ^verticalôffset^that ^weôbservedât thesurface(∼^{0.90 m)}ât^this^location^{(Fig. 3).}

5.2.ConstraintsontheseismichistoryoftheRoseauFault

The vertical fault slip of atleast 0.9 m that we have imaged (Fig. 3) provides constraintson theseismichistoryof theRoseau

Faultscarp,andonthemagnitudeoftheseismiceventthatcaused it.While available empiricalrelationships betweenfault displacement and magnitude are based on subaerial observations (e.g., Bonilla et al., 1984; Wells andCoppersmith, 1994), asdiscussed further below, a displacement of almost ∼^{1 m} ^requires ^a ^large- magnitudeevent(Mw>5–6),suchastheMw6.32004LesSaintes earthquake. The association of ephemeral indicatorsof coseismic displacement to the 2004 event is also supported by estimates of recurrence intervals for seismic events of similar magnitude and with similar slip. First, if a maximum long-term slip rate of 1 mm/yr for this fault is considered, as discussed elsewhere (Leclercetal., inpress),therecurrenceintervalforseismicevents of similar magnitude and with similar slip is longer than a few hundredsofyears.Second,thereisnorecord ofseismiceventsof similarmagnitudebothininstrumentalorhistoricalrecords(Feuil- letetal.,2011a,2011b),suggestingtherecurrenceintervalmustbe afewhundredsofyearsormore.

We thus infer that the smooth ribbon at the base of the slip plane and bound by the adhered sediment line can only be related to the 2004 event, and that earlier events are not recorded owing to long-term exposure to seawater and associated weatheringofthe slipplane (Fig. 6B). Thesemarkers ofcoseismic displacementare comparableto weathering-induced pat- terns on active continental normal faults (Giaccio et al., 2003;

Wiatr et al., 2015). While we have no constraint on the nature orrateofweatheringobserved onthe slipplane,the lackofany apparent gradient in weathering above the most recent seismic event(Figs. 3D, E), which isobserved insubaerial faults (Giaccio et al., 2003), suggests that weathering takes place at timescales shorter than the recurrence interval of seismic events. The sub- verticalfaultslipplanespreservedatRoseauscarpbase(Figs. 2,3 and5)thusrecordcumulativefaultdisplacementoverseveralseis- mic cycles (i.e.,over a few thousandsof years). Thisis basedon theirheight(upto∼^{10 m),}^the^coseismic^slip⁽∼^{1 m)}^from^large- magnitude(Mw>6)earthquakes,andtheinferredrecurrencetime (0.5–1 kyr).TheselongrecurrenceintervalsattheRoseauFaultare comparableto thoseinferred forsome active, normalcontinental faults such asthe 0.7–3.1 kyrsofthe MagnolaFault(Carcaillet et al.,2008) orthe∼^{0.2 kyrs}ôf^theÎrpina^Fault^(Galliând^Peronace, 2014).

5.3. Seaﬂoorobservationvs.modelpredictionsofseaﬂoorrupture parameters

Numericalmodelsoffault rupturegive first-orderestimatesof faultdisplacementandrupturelength(Feuilletetal.,2011a;LeFri- antetal.,2008).Theseestimatesrelyonnumerousmodelassump- tions,andlackanygroundtruthing.Usingthenear-faultwalldam- age zone structures, and the location of thefault plane showing coseismic slip, wedocument rupturealong∼^{3 km}ôf^the ^Roseau Fault, as shown in Fig. 5. Our results provide ground-truth con- straintthatshouldbeusedtoupdateandimprovecurrentmodels, asthe model predictionsshow discrepancies withrespect tothe actual coseismic rupture geometry observed at the seafloor. The

∼^{3 km} ôf ^rupture îdentified ^thus corresponds to approximately one-third of the10-km long rupture predictedby models (Feuil- let etal.,2011a).

The northwestern termination of the modeled rupture lies within our studyarea, but we haveidentified norecent seafloor deformationstructuresinthevisualandphotomosaicsurveyscon- ducted(Fig. 1BandFig. 5).Thismaysuggest thattherupturedid not actually propagatethat farnorthat thesurface, andthatthe modeltsunamisource(LeFriantetal.,2008) maythereforebein- accuratealongthisportionoftheRoseaufault.Atthesoutheastern limitofoursurvey areawefinddamageextending ∼^{100 m from} thefaultscarpintothehangingwall.Thissuggeststhattherupture

(10)

is comparableto the observedcoseismic displacement (∼^{0.9 m),} these are not collocated: the fault section with predicted maximum displacementlies towards thesouthwest ofthe fault plane outcropimagedhere(Fig. 3).Instead,inthe areaoftheobserved faultslipplanes,thepredicteddisplacementsaresomewhatlower (0.3–0.6 m; Fig. 5). It is thus possible that the actual ruptureis shorterthan the ∼¹⁰ ^km^predicted^by êxisting ^models. Îf^thisîs thecase,theoveralldisplacementattheseafloorshouldbelarger than that predicted by the same models for a seismic event of a given magnitude, which is consistent with the differences between observations andmodelpredictions outlined above. While the data presented here provide some initial constraints on the 2004 Les Saintes earthquake coseismic rupture, they are of limited spatial extent. Additional observations are required to fully characterize the length and geometry ofcoseismic rupture, map the variations in widthof the deformation zone in the hangingwall, and inspect additional fault slip planes, to obtain a com- pletedistributionofcoseismicfaultdisplacementprofilealongthe Roseau Fault. Nevertheless, the limited, available seafloor observations presented here demonstrate that models of rupture and tsunami source at Les Saintes require a reassessment, and that acquiringadditionaldataalongthisandotherfaultsystemsisnec- essaryto fully understandsubmarine faultruptures and improve associatedmodels.

5.4. Comparisonwithsubaerialrupturesandscalinglaws

The deformationstructuresobserved attheseafloor alongthe RoseauFaultcanbecomparedwithstructuresassociatedwithco- seismic ruptures in subaerial faults. Such comparisons can help assesswhether deformation andruptureprocesses are similar or differ fundamentally between these two environments, owing to differences in the environmental conditions and their effects on therheology ofboththe faultandthesurroundingmaterials. For example,thepresenceofwatertrappedinpoorlyconnectedcracks andporesduringdeformationeventscouldresultinelevatedfluid pressures, weakeningthe fault zone, andpromoting seismicrup- ture. Further, the elevated water pressure acting on the seafloor (∼¹⁰ ^MPa) ^could ^promote â ^velocity ^weakening ^behavior ⁱⁿ ^the shallowest portion of the fault, by helping the consolidation of otherwiseloosenear-surfacematerial(Maroneetal.,1991; Scholz, 1998).Suchamechanismwouldhelppropagaterupturepulsesall thewaytotheseafloorinsteadofdampingorhaltingthem.

Normal subaerial faults with exhumedfootwall bedrock often display ribbons of freshly exposed fault plane following seismic events(e.g.,Cello etal., 2000; Pantostiet al.,1993; Vittori etal., 2011),asnowalsoobservedattheRoseauFaultslipplane(Fig. 3).

These features are quickly obliterated on-land owing to rainfall, erosion,andother processes modifyingthesurface atratesmuch faster than in submarine environments. Near-fault coseismic deformation ofthe Roseau Fault hangingwallis also comparableto that observed subaerially. Normal faults in volcanic areas often feature shallow sub-vertical fault planes, with the formation of open ﬁssures, fractures, andantithetic and synthetic faults along the hangingwall and next to scarps (e.g., Acocella et al., 2003;

MartelandLangley,2006).Inelasticnear-faultdeformationinsub-

mation width is consistent withthose fromobservations atsub- aerialnormalfaultsshowingtypicalrupturewidthsoverdistances of <100 m and up to ∼^{300 m from} ^the ^fault,^that âlso ^concentrate primarilyalong thehangingwall(Boncio etal.,2012). While ourdetailedsubmarineobservationsare restrictedtoa singleco- seismic rupture,they suggest that deformation processessuch as hangingwall deformation and localizationof deformation on slip fault planes(Figs. 6AandB)are,to afirst order,similarto those observed onland.Finally,theapparent reactivationoferosionand sedimentationprocesses(Fig. 6C)suggeststhatseismicitymaysig- nificantly contribute to the long-termerosion of the Roseauand nearbyfaultscarps,asobservedonland(Keefer,1994).

Fig. 7 shows relationships between seismic event magnitude and surface rupture length, as well as maximum displacement for a compilation of subaerial faults, to be compared with our observations at the Roseau Fault. Theseplots conﬁrm that exist- ingscalinglaws(WellsandCoppersmith,1994; Wesnousky,2008) showsizable scatter,likelyreﬂectinggeologicalcomplexityofthese systems. Because of this variability, rupture length and displacement can vary by a factor of 10 at a given seismic magnitude.

In comparison, the single submarine observation of the Roseau Fault seems tohave arelatively shortrupturelength, wether the modeled (Feuilletetal., 2011a; LeFriantetal., 2008) andtheob- served (this study) rupture lengthsare considered (Fig. 7A). This ruptureisassociatedwithalargeobserveddisplacement(Fig. 7B), andhenceahighdisplacement-lengthratiocomparedtosubaerial faults. Whilewecan speculatethatthishighdisplacement-length ratio may be linked to near-surface displacement facilitated by hydrostatic pressure and the presence of ﬂuids, additional data from other submarine faults is required to evaluate the robust- ness of differences between submarine and subaerial fault ruptures.

6. Conclusions

Understandingthedynamicsofcoseismicfaultruptureandthe associated inelasticprocessesrequiresdetailedstudies alongsub- marine faults. These data thus provide a ﬁrst submarine observation that can be integrated with existing compilations linking surface rupturegeometry toearthquake magnitude,that are now basedexclusivelyonsubaerialfaultrupture.Morethantwothirds ofthe Earth’sseismicitytakesplace intheoceans, withasizable proportion in proximity to coastal areas, where there is a com- binedseismicandtsunamihazard.

Withthisstudywe demonstratethat underwater vehicles can conduct geophysical surveys and geological observations at spa- tialscales similartothatofsubaerialstudies,andthatsystematic androutineseaﬂoorobservationsinsubaqueousenvironmentscan be conductedusingadvancedacoustic(Fig. 2)andoptical(Figs. 3 and 4) imaging techniques. Along the Roseau fault these methods reveal recent deformation structuresthat we can link to the submarine 2004M_w 6.3LesSaintes earthquake. We document a surface rupturelengthofatleast∼^{3 km}^that ^does^not^prolong^as farnorthaspreviouslymodeled,andwhoselengthislikelyshorter than the ∼10–15 km rupture length proposed by earliermodels.

We alsoidentify coseismic displacementat apreservedfault slip

(11)

Fig. 7.Earthquakemagnitudevs.surfacerupturelength(A)andmaximumdisplacement(B)ofsubaerialnormalfaults(redcircles)comparedwiththesubmarineobservations alongtheRoseauFaultrupture(largebluedots)andmodelestimates(smallbluedot).Subaerialobservationsrangefrominstrumentalrecords(1977andlater,largestcircles) topre-instrumentalhistoricalevents(1900andearlier,smallestcircles).Selectedmagnitudescalinglaws,labeledW&C(WellsandCoppersmith,1994) andW(Wesnousky, 2008) arealsoplottedfornormalfaults(N,solidlines)andforallfaultscombined(A,dottedlines).Thedatausedinthisplot,togetherwithcorrespondingreferences,are providedinSupplementaryTable 1.(Forinterpretationofthereferencestocolorinthisﬁgurelegend,thereaderisreferredtothewebversionofthisarticle.)

surface,withamaximumdisplacementof0.9 m.Thisissomewhat higherthanthatpredictedbymodels,andnotcollocatedwiththe predictedzone of maximumdisplacement. Ourobservations also indicatethaterosionofthefaultplanealonggullies,andconcomi- tantsedimentationondejectionconesattheirmouths,iseﬃcient andmayoccurco-seismically,post-seismically,orboth.

Thecoseismic deformationstructuresobservedattheseafloor, both on fault slip planes andon the hangingwall,are similar to those observed along subaerial ruptures. We are able to identify well preserved seafloor ruptures ∼10 yrs after their formation (Figs.2–5)owing tosubmarineerosionandweatheringratesthat arein generallessefficient thanthose operating onland. Hence, the techniques used here open the possibility of conducting additional,extensive high-resolutionunderwater surveysalong fault scarpsthat have witnessed earthquake ruptures in recent times, butthat havenotbeen characterizedtodate.Such offshorestud- ieswouldallowustobetterdocumentandenrichexistingseismic catalogsthatare usedfor semi-empiricalscaling laws(Wellsand Coppersmith,1994;Wesnousky,2006;2008),whichlackunderwa- terseismiceventsandunder-represent normalandthrustseismic ruptures. Bydoing so we could improve the assessment of seismic andtsunami hazardsassociated withearthquakes inoceanic areas.

Authorcontributions

Datawereacquiredbythecruiseparticipants(NA,AB,DB,VC, YC,MG,KH, CH,BI,JJ,BJ,FL,TL,CJM,MM-C, PN,JAO,SP,AS,MT, LT), led by JE. The experiment was planned and designedby JE, MC,FL,CD,NF,andMA.Processingofmicrobathymetry datawas conductedbyNA,MR,ASandGV,incoordinationwithSV,JE,and MC.Imagerywas processedby RP,RG, NG,andJE.Data interpretation was coordinated by JE, and involved all participants. The manuscriptwas draftedby JE, FL, CM, andJAO, withinput from therestofco-authors.

Acknowledgements

DatapresentedinthispaperwereacquiredduringtheODEMAR Cruisewith trip time andROV deployment fundedby CNRSand IFREMER (France), with AUV deployments funded by GEOMAR (Germany). CNRS provided support fordata analyses through an INSU-SYSTER granttoJE.CJM’s participationwasfundedby NERC grant NE/J021741/1 at Cardiff University. We thank X. Castrec (IFREMER) forfacilitatingpermitsandauthorizations requiredfor the ﬁeldwork, and the GENAVIR Team and ship’s crew and oﬃ- cers for all shipboard and ROV operations. Reviews provided by A. Robinsonandananonymousreviewer,inadditiontothosefrom the Editor An Yin, helped improve this manuscript. This is IPGP contribution3757.

Appendix A. Supplementarymaterial

Supplementarymaterialrelatedtothisarticlecanbefoundon- lineathttp://dx.doi.org/10.1016/j.epsl.2016.06.024.

References

Acocella,V.,Korme,T.,Salvini,F.,2003.Formationofnormalfaultsalongtheaxial zoneoftheEthiopianRift.J.Struct.Geol. 25,503–513.

Armijo, R., Pondard, N., Meyer, B., Uçarkus, G., deLépinay, B.M., Malavieille, J., Dominguez,S.,Gustcher,M.-A.,Schmidt,S.,Beck,C.,Çagatay,N.,Çakir,Z.,Imren, C.,Eris,K.,Natalin,B.,Özalaybey,S.,Tolun,L.,Lefèvre,I.,Seeber,L.,Gasperini, L.,Rangin,C.,Emre,O.,Sarikavak,K.,2005.SubmarinefaultscarpsintheSeaof Marmarapull-apart(NorthAnatolianFault):implicationsforseismichazardin Istanbul.Geochem.Geophys.Geosyst. 6,Q06009.

Avouac,J.-P., Ayoub,F.,Wei,S., Ampuero, J.-P.,Meng, L.,Leprince, S.,Jolivet, R., Duputel,Z.,Helmberger,D., 2014.The2013,Mw7.7 Balochistanearthquake, energeticstrike-slipreactivationofathrustfault.EarthPlanet.Sci.Lett. 391, 128–134.

Baker,A.,Allmendinger,R.W.,Owen,L.A.,Rech,J.A.,2013.Permanentdeformation causedbysubductionearthquakesinnorthernChile.Nat.Geosci. 6,492–496.

Barreyre,T., Escartín,J., Garcia, R.,Cannat, M.,Mittelstaedt, E.,Prados,R., 2012.

Structure,temporalevolution,andheat fluxestimatesfromthe LuckyStrike deep-seahydrothermal fieldderived fromseafloor imagemosaics.Geochem.

Geophys.Geosyst. 13,Q04007.