Sediment evolution in the mouth of the Seine estuary (France): A long-term monitoring during the last 150years

Stratigraphy,

Sedimentology

(Geology

of

Sedimentary

Basins)

Sediment

evolution

in

the

mouth

of

the

Seine

estuary

(France):

A

long-term

monitoring

during

the

last

150

years

Sandric

Lesourd

a,

*

,

Patrick

Lesueur

a

,

Ce´dric

Fisson

b

,

Jean-Claude

Dauvin

a a

NormandieUniversite´,UNICAEN,M2C,CNRS,UMR6143,24,ruedesTilleuls,14032Caen,France

b_{GIPSeine-Aval,poˆlere´gionaldessavoirs,115,boulevarddel’Europe,76100Rouen,France}

1. Introduction

Manyestuarieshavebeensubjectedtostronghuman impacts(Blottetal.,2006;Hossainetal.,2004;Lesourd etal.,2003;MorrisandMitchell,2013;PyeandBlott,2014; Thomasetal.,2002;Tonisetal.,2002;VanderWaletal., 2002;Wolanski,2006;Wolanskietal.,2001).Amongthem themacrotidalSeineestuaryisagoodexample.Duringthe lastdecades,astrongartificialreductioninthe accommo-dationspace,superimposedonnaturalvariations,hasled to a downstream shift of the sedimentation area. The comparison of old nautical charts and sedimentological data fromthe 19th century shows an increase in fine-grainedsedimentation(i.e.mud,grainsize<63

m

m)inthe outletoftheSeineRiver,aswellasintheshallowwatersof theBayoftheSeine.Themodernfine-graineddeposition extendstotheopensubtidalshallowarea(Lesourdetal.,

2001;Lesueuretal.,2003).Asaresultinthemouthofthe Seineestuary,coresshowafiningupregressivesequence thatcorrespondstoashiftfromaninnershelftermwith pebblesandcoarsesands,toanestuarinetermwithmud and fine sand (Lesourd et al., 2001). Moreover, the distribution of superficial sediment is governed by seasonalvariations,dependingonboththeriverdischarge intensityandthewaveactivity(Lesourdetal.,2003):when 55%ofthetotalmouthareaiscoveredbymudattheendof thewinter,thisamountisapproximatelyof24%attheend ofthesummer.

Theaimofthisstudyistounderstandtheevolutionof thesuperficialsedimentation,overthelast150years.The main question isto estimatetheimpactof engineering worksonthesedimentationofanestuary.Thesetofdata onsuperficialsedimentisextensiveenoughandcoherent topermit acorrelation withthehydrodynamiccontext, suchastheriverflowrateonthesedimentation.Beyond this question, this study involves a larger work of monitoring of an estuary submitted to strong human impact since themiddle of the 19th century, using an

ARTICLE INFO

Articlehistory:

Received28July2015

Acceptedafterrevision19August2015

Availableonline17December2015

HandledbySylvieBourquin

Keywords: Sedimentation Estuary Riverflows SeineRiver ABSTRACT

Theaimofthisstudyistounderstandtheevolutionofthesuperficialsedimentation,over

thelast150yearsonthemouthoftheSeineestuary.Anindicatoriscreated,calculatingthe

percentageareacorrespondingtothedifferentfaciesforeachsetofavailabledata

(mid-19thcenturyto2009).Theshiftbetweenthe1970s(16%ofsurfaceareaofmudandmuddy

sand),the1990s(about50%)and2009(5%)appearsclearly.Thedecreaseinthemuddy

areaisbalancedwithanincreaseofsandymudandmuddysandsurface.Thisevolution

could be explainedby riverflowrate activity. TheSeineestuary is anaturally

tide-dominatedestuary; however,engineeringactivitieshave increasedtheenergy ofthe

channelizedriver.Consideringthefine-grainedfractioninthesuperficialsedimentsand

itsvariations,theSeineestuaryhasshiftedtoariver-dominatedsystem.

ß2015Acade´miedessciences.PublishedbyElsevierMassonSAS.Thisisanopenaccess

articleundertheCCBY-NC-NDlicense(http://creativecommons.org/licenses/by-nc-nd/

4.0/).

* Correspondingauthor.

E-mailaddress:sandric.lesourd@unicaen.fr(S.Lesourd).

ContentslistsavailableatScienceDirect

Comptes

Rendus

Geoscience

ww w . sci e nc e di r e ct . com

http://dx.doi.org/10.1016/j.crte.2015.08.001

1631-0713/ß2015Acade´miedessciences.PublishedbyElsevierMassonSAS.ThisisanopenaccessarticleundertheCCBY-NC-NDlicense(http://

originalsetofdata,usefulformonitoringtheevolutionof theestuary.

2. Background

TheSeineisthelargestriverthatflowsintotheEnglish Channel (La Manche). It hasa length of 780km and a catchmentareaof78,000km2_{inwhich40%oftheFrench}

population and economic activityis concentrated (Paris Basin).TheSeineestuaryistheresultofthedrowningof the Weichselian incised valley during the Holocene transgression (Lefebvre et al., 1974), as is the case of manymodernestuaries(Dalrympleetal.,1992;Fairbridge, 1980;Perillo,1996).Thepresent-daymorphologyofthe downstream partof theSeine is theresultof its filling duringtheHolocene(Lesueuretal.,2003;Tessier,2012;

Tessieret al., 2010). The downstream partof theSeine estuaryisfunnel-shapedwithamouththatopenstowards thewestwithelongatedsandbarsthataretypicalof tide-dominatedestuaries (Dalrympleetal., 1992).The mean riverdischargeis400m3_{/s(1940–2014average,}

measure-mentatPosesLock,seeFig.1),withmeanextremevalues of200–2000m3_/s1_{.Themeanannualmassofsuspended}

particulatematter(SPM)intheriverdischargeisestimated tobeabout6105_{T/year(GIPSeine-Aval),withmarked}

variations over years, extreme values being 1.3105– 1.7106_{T/year (‘‘Service de navigation de la Seine’’,}

unpublished data). The modalities of sediment supply accordingtofreshwaterflowhavebeendocumentedin

Avoine(1987),Gue´zennecetal.(1999),andLesourdetal. (2003).A great partofSPM andmud of thestudyarea originates from the catchment area of the Seine River

Fig.1.Mapofsuperficialsedimentinthemouth(088Eto08;49823to49832N)oftheSeineestuary,includingpartofthemainnavigationchannelandthe

northernandsouthernchannels,basedonthe2009–2010samplingcampaigndata.Accordingtosiltandclaycontents:sand(<5%),muddysand(5–25%),

sandymud(25–75%),mud(>75%).Pebblefaciesareinterpretedasinapreviousstudy(Lesourdetal.,2001;Lesueuretal.,2003)usingdrilling,forthesame

(Dubrulle et al., 2007). The sediment cover is mostly composedoffine-grainedmaterial(siltandclay).Thetidal amplitudeatLeHavreis3mduringneaptidesand7.5m duringspringtides.Aflowvelocityupto2.5m/s atthe surface (1.60m/s on the bottom) can occur during the risingtide.Thetidalcycleissemi-diurnalwithahighwater slack close to 3h. In the Seine estuary, as in other macrotidalestuaries,thesuspendedsediment concentra-tionintheturbiditymaximum(TM)ismainlygovernedby tidalcurrents(Allenetal.,1980;Avoine,1987;Brenonand LeHir,1999;LeHiretal.,2001;Lietal.,1995;Mitchell etal.,2003,2012;SalomonandAllen,1983;Wells,1995). Thesuspendedsedimentconcentrationvariesfrom0.1to 1g/L, associatedwithfluid mud(100–300g/L),duringa semi-lunartidalcycle.ThetotalmassoftheTMvariesfrom 2105Tduringneaptideto4106Tduringspringtide (Avoineetal.,1981),withalongitudinaloscillationofthe massgravitycentreoftheTMofabout20km.Duringhigh riveroutput,a varying but largeproportion oftheSPM containedintheTMisexpelledouttotheBayoftheSeine (Avoine, 1987). Propagation of swells within the Seine estuary is highly complex and poorly documented. Complexityisduetotheunevenmorphology(sandbars, embankments)and waterdepthsthat varyrapidlywith the tide conditions. Nevertheless, the Seine estuary is mainlysubjectedtowesterlytonorth-westerlyswellsand towavesgeneratedbylocalwindsintheBayoftheSeine, withawesterlydominance.

3. Materialandmethods

Anewsetofsuperficialsedimentssamples(Table1)has beencollectedinthemouthoftheSeineestuaryduringthe years2009–2010(COLMATAGEproject,intheframework of Seine-Aval program). Sample stations were chosen accordingtoaregulargrid(1nauticalmilespacedout).The setofdataconsistsofamainsamplingmadeduringMay 2009 (300 samples). Some additional samples were realizedduringyears2009–2010.Someofthem(27)were reacquiringsamplestationstostudyseasonalvariationsof thesedimentation.Atotalof561sampleswereanalysed.It mustbenoticedthatmappingbymarinegeophysics,such

asside-scansonar,isnotrelevantinthestudyareabecause thepicturesshowgenerallymonotonousacousticsfacies (Lesourd,unpublisheddata). Samplesweretakenwitha ShipekgrabandaReineckbox-corer;severalsampleswere realizedoneachpointtoaccountforlocalfaciesvariations and/orwhenthesampleswerenotrelevant(reworkingof sedimentarylayersduringsampling,oremptygrab).

Each sample wasdescribed and photographed. Sub-samplings were realized according to the sediment structuration. In the case of coarse sediments (gravel), the sediment was sieved in a dry environment using 15 sieveswith Frenchstandard mesh sizes(i.e. AFNOR definition); the particle size distribution of the other sampleswas obtainedwitha Laser CoulterLS230.The sedimentologicalclassificationofsampleswasestablished usingboth structural(asdescribedinthegrab afterthe sampling) and textural (grain size used to reinforce structural description) characteristics. The classification isdisplayedinFig.1.

The2009map(Fig.1)hasbeenobtainedwithmanual correlation(Diesingetal.,2014),usingbathymetricdata associated with a knowledge based on more than 1500 samples realized in the Seine estuary since the 1990s. Areas were drawn using density of points, morphology(limitofbank,trough)andseasonalvariations whenastationwassampledduringtoseveralseasons.The highnumberofsamplespermitstodrawarelevantmap, butneitherthedensityofsamples(toofew)norvariations ofsedimentaryfacies(scatteredgeographicaldistribution) allowustousestatisticalanalysestocreatethemap.GIS softwarewasusedtocalculatetheareacorrespondingto thevariousfacies.

The2009data,associatedwithfourothermaps(Fig.2), forma verycomplete andoriginalsetallowinga better understandingofthestoryoftheSeineestuarysediments.

4. Resultsandinterpretation

Thedistributionofsuperficialsedimentsinthemouth oftheSeineestuary(Fig.1)showsin2009–2010a high diversity of facies. Most of the samples correspond to sandymudormuddysand,whereasveryfewofthem(7%)

Table1

SetofsuperficialsedimentssamplescollectedinthemouthoftheSeineestuaryduringtheyears2009–2010(COLMATAGEproject,intheframeworkof

Seine-Avalprogram).

Date Numberandmethodof

sampling

Source Constructionworkbackground

forthestudyarea(seeLesourd

etal.,2001)

Flowratebackground

(seeFig.3)

19thcentury Northernpart,1834,

364sounding,southernpart,

1883,9310soundings

SHOM,1834–1883 Savageestuary Unknown

1913 30,000soundings SHOM,1913 Constructionworksbut

upstream

Unknown

1977 655Shipeckgrabsand

40shortcores

Avoineetal.,1981 Justbeforetheconstructionof

majordikes

Normalflowratedynamic

1993–1996 1100Shipeckgrabsand

125shortcores,(12sampling

campaign)

Lesourdetal.,2001 Continuousconstructionworks Normalflowratedynamic

2009 561Shipeckgrabsand

10shortcores(4sampling

campaignsin2009–2010)

Thisstudy Continuousconstructionworks Afteralongperiodoflow

containmorethan75%ofsiltandclay.Consideringgrain size,mostofsamplesareamixtureofthreemainclasses (>200

m

m,[63–200

m

m],<63

m

m),involvinganunsorted stock of sedimentary material. Sediments<63

m

m are composed ofa constant proportionof coarsesilts ([63– 20

m

m],around30%),finesilts([20–4

m

m],50%)andclay (<4

m

m,20%).Fourmodesarerevealedbythegrainsize laser analysis (18, 38, 97 and 200

m

m), systematically detected,withvariablerelativeproportions.

This2009map (Fig.1)revealsarepresentationofall facies,withascatteredgeographicaldistribution.Ageneral overviewshows:

thewide sandbarson thetwosidesoftheendofthe navigationchannel;

alargegeographicalrepresentationofsandymudfacies (25to75%ofsiltandclay);

anareaofmuddysandfaciesjuxtaposingcoarsesandin theshallowerwateroftheBayoftheSeine.

Thenorthernchannelischaracterizedby high-scatte-red facies, featuring reworking of both seabed and mudflats(originatedintheformationofanewchannel). The seabedof thenavigation channelis also underthe influenceofpermanentengineeringworks(i.e.dredging) andshowsaninconstantpatternoffacies,essentiallyshell fragmentsandmudpebbles,asdiscussedlater.Anumber of samples correspond to pebble and gravel, generally Flandrianandante-Flandrianmaterial,difficulttosample withgrabs,andmorelocallytheycorrespondtooutcrops ofMesozoicrocks.Akindofspecificsedimentiscomposed

ofverystiffancientblackmudcharacterizingtheseabedof formerchannels,inthenorthernandsouthernmarginsof theestuary.Consideringonlythegrainsize,thisfaciesis fine-grained (more than 85% of silt and clay) but neverthelessmustbeconsideredasoutcrops.Abovemost of this facies, variable and temporary sedimentation includingfluidmudorsandripplescanbefound.

Themudfacies(siltandclay>75%)ispoorlyexposed, comparativelytotheformer1996study(seethemapin

Lesourdetal.,2001).Mudcanbefoundaroundthelarge sandbanksandwithintroughs.However,consideringthe visualdescriptionandnotonlythegrainsizeanalysis,silt andclaycanbesampledaslittlemudpebblesofverysoft mud,layeringontheseabed,whateverthefacies. These soft mud pebbles were not included in the grain size measurements,andcouldbefoundalloverthestudyarea, asdiffusemud deposits.Theyprobably correspondtoa residueoffluidmuddepositsastheresultoftheexpelling anddepositionofturbiditymaximumduringfloodevents. Thesandymudfacies(25to75%ofsiltandclay)was verywellrepresented onthewhole mouthoftheSeine estuary in 2009–2010. This facies occupies the former place of the mud facies, especially in theNorth of the estuary,whereitreplacesmudandcoverscoarsesandand muddysandareas.

Themuddysandfacies(5–25%ofsiltandclay)isalso largely represented on this map, contrasting with the situationin1996,whenveryfewsamplescorrespondedto thisfacies.Morethanarecoveringoftheformerfacies,this appearsasasiltandclayenrichmentofthesand,mostlyon theexternal partof theestuary. So,muddy sandfacies appears:

Fig.2.SedimentaryfaciesevolutionoftheSeineestuaryoverfiveperiodsoftime(seeFig.1forgrainsizeclassification)Redline:delimitationofcommon

areaonwhicharecalculatedpercentageareascorrespondingtothedifferentfacies(mud,sandymud,muddysandandsand).SeeTable1forthesourceof

data.Accordingtosiltandclaycontents:sand(<5%),muddysand(5–25%),sandymud(25–75%),mud(>75%).Data:SHOM(1834–1883);SHOM(1913);

inthewesternpartoftheestuary,widelyovertheformer areaofmuddysand;

on the northwestern part of the estuary where the correspondingsurfaceareasincrease;

alongthenortherncoast.

Theformerstudy(Lesueuretal.,2003;Lesourdetal., 2003)showedthatthemuddyarea(i.e.muddysandand sandymud)didnotpasstothewestovertheGreenwich meridian;thislimit isupto28Waccordingtothenew sampling.

Thesandfacies(lessthan5%ofsiltandclay)ispoorly represented in the mouth of the Seine estuary, mainly extendingatshallowdepths,onbothsidesoftheendofthe dikes.Sandformsthetopofwidebanks,thenorthernone beingmainlycomposedofthixotropicsandandprobably suppliedbytheKannickdepositofsedimentdredging.

Some of the 2009 facies are very specific and new, comparedtothe1996map(Fig.2).Itisthecaseofshells layers,withsamplesthatmaycontainupto99%inweight ofshells,eitherasfragmentsorentire.Someofthemare foundattheseabedofthenavigationchannelandnorthern channel;othersareoverlappinglocallyaseabedofgravels andpebbles.Thesedeposits,easilytransported,originate intheerosionandsortingofsedimentbytidalcurrentsand swells,butalsounderthedredgingworksofthenavigation channel and Port 2000 works. Another specific facies correspondstomudpebbles.Theyweredescribedinthe former 1996 situation (Lesourd et al., 2003), but very locallyandaslowerfrequencies,andlayeringabovethe surface of other sediments. They were foundjust after strong gales that had eroded stiff mud deposits. In the presentstudy,theyappearasanindividualfacies,dueto theirwideextensionoverthewholestudyareaandtheir high amounts in the corresponding sediments: mud pebbles representup to 30% in weight of the samples. Theyarepebblesofstiffmud,mixedwithamuddysandor sandymud.Theseheterometricpebblesarefoundinthe navigation channel and in the northern channel, and, dependingonthelocation,aretheresultof:

dismantling of the seabed by dredging works (in the navigationchannel);

erosionofdepositsbyswells;

erosionofstiffmudofthenorthernmudflatassociated with theshiftingof thenorthernchannel (due tothe embankmentsofLeHavrePort2000works).

Theevolutionofsuperficialsedimentsfrom1834–1883 to1993–1996 was described and discussed by Lesourd etal.(2001).Thecontributionofournewdatahighlights the more recent evolution and the relative impact of hydrodynamicparameters(essentiallyriverflowrates)on thesedimentation(Fig.2).Inthemiddleof19thcentury, theSeineestuarywasstillconsideredasa ‘‘savage’’(i.e. pristine) estuary, because it was not yet affected by dredging or the effects of major port works. In 1913, construction works are still localised upstream. During year1977and after,constructionsare importantin the estuary(major dikes finished in 1980). 1993–1996 and

2009situations are alsocharacterizedby strong human impact.Inthemiddleofthe19thcentury,about65%ofthe superficialsedimentsweresands,transitingoveroutcrops of gravel and pebble, especially in the northern area. Muddy facies,poorly represented, wereassociated with troughs.In1913,sandsoccupiedthemostpartofthestudy area; in thenorthern area, sand ridges covered former gravel and pebble outcrops. Mud was found only in troughs.Accordingtothe1977situation,gravelandpebble outcropshavebeencoveredbysandandsandymud;the mostfrequentfacieswerestillsandy,but thetotalmud area had increased. In the 1993–1996 situation, sand formedwide sandbanks. Muddyand sandymud facies wereprevailing, spreading over sand,aroundthebanks andatthebottomofthechannels.Theincreaseinmuddy facies was thenexplained by themodificationof water circulation, according to the construction of new dikes (Lesourdetal.,2003).The2009situationcorrespondsto theremovalofthefine-grainedsediments,thedecreasein mud being balancedby theincrease in mixedmaterial, sandymudandmuddysand;therehasbeenalsoaclear westwardextensionofmuddysand.

TheSeineRiverflowrateshavebeenmeasuredfromthe 1940s.Fig.3showsthenumberofdayswithaflowrate exceeding700m3_{/s,whichisaroughthresholdestimate}

fromwhichflow ratecouldbehighenoughtoinducea significantsedimentationinthemouth(Gue´zennecetal., 1999).Theflowratemustbestrong,butalsolongenough toinduce‘‘estuarinefloods’’(Garnaudetal.,2002,2003), expellingtheTMandassociatedfluidmudovertheareaof themouth.Duringtheendofthe1970s,thehydrological contextwasastrongandprolongedlowflowrateperiod (e.g.,onlyafewdayswithflowrate>700m3_/sin1977).

Duringthemid-1990s(1996map,Fig.2),thenumberof dayswithriverfloodswassignificant(until140in1995), whereasbefore2009,thenumberofdayswithhighflow rate was moderate (max 77), so that no strong fine sedimentation occurred during a period of about nine years.

For each set of available data, mid-19th century, 1913, 1977, 1993–1996 and 2009, the percentage of areascorrespondingtothefivemainfacies(mud,sandy mud,muddysand,sandandgravel,andpebble)overthe same total area of the mouth (Fig. 2)were calculated (Fig.4).Itfirstshowstheverygoodconsistencyofthis sedimentologicalindicatorwiththeevolutiondescribed with maps, even with the oldest charts based on the tallow sounding method (19th century, 1913). The proportionofgravelandpebble,hardground,andsand remained constant up to 1977. Nevertheless, from 1913 to 1977, a slight variation occurred between mud and sandy mud (due to the distinction between mud and sandy mud donesince the 1997 study). The shiftbetweentheendofthe1970s(16%ofsurfacearea of mudandmuddy sand)andthe1990s(about50% in surfaceofmudandmuddysand)appearsclearly.Itmust benoticedthatthesiltationoftheSeineestuaryisstill nowadaysimportant, butwhilethe1993–1996 muddy areascorrespondedtothicklayers(10cmtolocallyupto 1m)ofmudcoveringasandyandgravellysubstratum, mostsamplesin2009correspondtoamixofsandand

mudwithoutanyinternalstructure.So,thedecreasein thetotalmuddyareawasfromabout40%in1996to5% in2009,balancedbyanincreaseinsandymud(from14% to24%)andmuddysand(from6to22%)surfaces.During

this time, the total area corresponding to sand has remainednearlythesame.Thereisalsoaslightincrease in hardground surfaces,corresponding tooutcropping stiffmud.

Fig.3.Annualoccurrenceofflood/highflowrateeventsoverthe1941–2011period.Blackline:numberofdayswhentheflowrateexceeded700m3

/s,

roughestimatethresholdfromwhichtheflowratecouldbehighenoughtoinduceasignificantsedimentationinthemouth.Thisnumberisanapproachto

measuretheflowrateduringayear.Greyline:numberofdayswhentheflowrateexceeded1500m3_{/s,whichisasignificantlystrongflowratefortheSeine}

estuary.

Fig.4.Percentagesofareascorrespondingtothedifferentfacies(mud,sandymud,muddysandandsand)intheirproportionofthecommonestuarymouth

area(reddelimitation,Fig.2),calculatedforeachsetofavailabledata(19thcentury/around1850,1913,1977,1993–1996and2009).Accordingtosiltand

claycontents:sand(<5%),muddysand(5–25%),sandymud(25–75%),mud(>75%).Sandymudfaciesisnotdiscriminatedfor1834/1883and1913,and

5. Discussion

Superficialsedimentationofestuariesevolves accord-ing to multi-decennial and seasonal time-scales, as describedintheHudsonRiver(Woodruffetal.,2001)or in theHumber-Ouse(Uncles et al., 1998). In the lower Seineestuary(Lesourdetal.,2003),superficialsediments havebeenshowntobesubmittedtosignificantseasonal changesduringtheyears1993–1996:mudandsandymud covered55%ofsurfaceduringwinter,whentheturbidity maximum commonly shifted toward themouth during highriverflows,inducingmuddeposition,whereas24%of the surface was characterized by mud and sandy mud duringthesummerperiod.Attheendofthewinter, fine-grainedsedimentcanbereworkedandscatteredoverthe estuary,apartbeingtransportedupstream(channelsand feedingtheTM),anotherpartbeingexpelledtotheBayof theSeine.

Thenewsetofsamplescollectedin2009–2010includes anumberofreacquiringsamplingstations(RSS).Seasonal variationswereverylightforbothgrainsize(%<63

m

m) andfacies.Thisinvolvesaweakimpactofmudsuppliesby theriverduring few highflowrate periods.Areaswith strong seasonal evolution are seabed composed of stiff mud,occasionallycoveredwithsandorfluidmud.Areas withslightvariationsareessentiallylocalizedaroundsand banks,suggestingamovingboundaryofthesebanks,with avariationoffaciesfromsandymudtomuddysand.

Theevolution ofsuperficialsedimentsisanindicator characterizingthemorphological evolution ofthe Seine estuary.Maps haveshown aclearevolution from sand-dominated seabed, as generallyin estuaries (Dalrymple etal.,1992;Perillo,1996)toseabedmainlycomposedof sandymud,thenmud.Thisevolutioniscorrelatedwiththe anthropogenicevolutionofestuary(Lesourdetal.,2001; MorrisandMitchell, 2013;Thomasetal.,2002;Vander Waletal.,2002;Wolanski,2006).Theevolutionillustrated bythe2009dataischaracterizedby:

astrongdecreaseinthetotalmudfaciesarea(facieswith morethan75%ofsiltandclay);

ashiftfrommudandsandymudtomuddysand;

a decrease in sandy area balanced by an increase in muddysandarea;

theextensionofthemuddysandareaby4kmbeyond theformerknownboundaries.

The fine-grained sediment did not disappear in the studyarea,buthasspreadovertheestuaryandhasbeen incorporated in the sandy fraction. The silt and clay contentoftheSeineestuaryisalsostillsignificant.

Theratiobetweenthemud-dominatedarea(surfaceof mud and sandy mud) and the sand-dominated area (surfaceofsandandmuddysand) isagoodindicatorof thetemporalchangesofthesuperficialsediment(Fig.5).It is calculated over a common area on thefive maps,as shown in Fig. 2. Before 1977, this ratio was around 0.25.Thechangebetween1977and1993–1996appears veryclearly,witharatioincreasingupto1.5. Themost recentdata(2009)correspondtoaratioof0.54,illustrating

thepersistenceofthesiltandclaycontentinthesediment andaspreadingofmudoverthearea.Siltandclaycontent increasebysettlingontheseabedandmixingbycurrent andswell.Stiffmudpebblesarespecificsourcesofsiltation when they rest on the top of the seabed and are transported by tidal currents and swells. Eroded by abrasion,theyreleasea fine-grainedmaterial,mainlyin thewaterbutalsoinsidethesediment.Asobservedinthe mostrecentsamples,theycanbeincorporatedintosandor muddysand.Trappedmudballsarealsomixedwithinthe sediment,leadingtoaninternalmudsupply.Observations ofmanygrabcontentsshowpebblesbeingeroded(friction wear), sothat theyappear as‘‘ghosts’’of mud pebbles, their boundaries being erased when integrated in the sandymatrix.

TheSeineRiverflowshavebeenmeasuredattheupper limitoftheestuaryfrom1941uptotoday(GIPSeine-Aval). Duringtheendofthe1970s,flowratesweregenerallyvery low(Fig.3),whichcouldexplaintheremainingofsandy areas,eventakingintoaccountthefactthatengineering workwasalreadyfinishedupstreambeforethisperiod.For theyears1993–1996,periodsofhighflowratesoccurred frequently, although 2009 data were collected after differentconditions;superimposedonseasonalvariations, long periods (about 9years) of low river flows (when storage of fine-grained sediments occurred within the estuary)weredisruptedbyfewandshortperiodsofhigh river discharges (when mud supply increased in the downstreampartoftheestuary).Beforeandduringyears 1993–1996,flowrateswerestrongenoughtodisplacethe turbidity maximum downstream, to expel it from the navigation channeland totrigger a highsedimentation rateofmudintothemouthoftheestuary.So,highsupply ofmudinvolvedadepositionofthicklayersoffine-grained sediment.Undertheseconditions,thesedepositsarethen reworkedanddispersedallovertheestuary,accordingto seasonalvariations,asdescribedabove.Whensuppliesof estuarinefine-grainedsedimentsaretooweak,reworking

Fig.5.Ratioofmuddyarea(mudandsandymud)tosandyarea(sandand

muddysand),calculatedforthe19thcentury,1913,1977,1993–1996

processes by waves and tidal currentsdominate. These otherconditionsleadtothemixingofthelowquantityof mudwithsands.ThispatternwasshownintheinnerBay oftheSeine(Garnaudetal.,2002).Thethinpebblesofvery softmudoftenobservednowadaysarealsoalightversion of fine-grained sedimentation processdescribed for the years 1990s. Theimportance of highriver flowsversus tidaldynamicshasalsobeendescribedinotherEuropean estuaries (e.g., Blott et al., 2006; Mitchell et al., 2012; Thomas et al., 2002), Western Australian estuaries (Wolanski et al., 2001) or subtropical estuaries (Van MarenandHoekstra,2004;Wangetal.,2013).

The present-day distribution of sedimentary facies illustrates a spreading of fine-grained sediment, beyond formerboundaries,butalsoontheseabedoftheBayofthe Seine,wheresedimenthadalwaysbeen,untilnow,described ascoarsesandorgravellysand.Therepartitionofsuperficial sediment in 2009 looks like the repartition of facies in summerduringthe1990s,i.e.afterthereworkingofthemud depositedduringthepreviouswinterconditions.Thefacies corresponding to 2009 are also the product of the long reworkingofdepositsofthepreviousyearswithmoderate riverflows.Suchanobservationsuggestingaglobalerosion oftheestuaryarereinforcedbyotherobservations:

thepresenceofnumerousshellfragments, characteriz-ingstrongreworkingandsortingofsediment;

the presence of thixotropic sands, attesting to an unstabledeposit;

thepresenceofoutcropsofblack,old(dozensofyears) and verystiff mud, showinga striping ofpresent-day sediments;

thepresenceofmud pebblesbeingerodedwithinthe sediment;

theabundanceofmudpebblesabovetheseabed,evenif mostofthemareassociatedwithdredgingworkinthe navigationchannelanddismantlingofnorthernmudflat inthenorthernchannel(Cuvilliezetal.,2009).

TheSeineestuaryisanaturallytide-dominatedestuary, accordingtotheclassificationofDalrympleetal.(1992); however, engineering activities, including building of embankments and dikes, have induced the infilling of lateralareas,thestrongreductionoftheavailableamount of space and has led to a downstream shift of the sedimentation area and increased the energy of the channelized river. Nevertheless, nowadays, the role of theriverhasincreasedanditsconsequenceisanexpelling oftheTMandhighmudsupplyinthemouthoftheSeine estuary.DuetotheexpellingofTM,thebeachesoftheBay oftheSeinearedistalpartsoftheSeineestuarinesystem (Dubrulleetal.,2007).Whentheriverflowsarehighfor long enough, they provide mud supply and deposition everywinter,and especially duringthefirstmajor high flow event. When theriver flows are low, mud supply remainsweak,andreworkingbywavesandtidalcurrents dominates. Then, the superficial sedimentation is also characterized by erosionalfeatures, includingmixing of sand and mud, mud pebble production, sorting of sediments.

Wepointoutthefactthatitisdifficulttoidentifythe influence of man-made versus natural influence on the sedimentology.Allthestudies(e.g.,Avoineetal.,1981)on long-termevolutionofmorphologicalcharacteristicshave shownthatanthropogeniceffectsaredominantonnatural evolution.Furtherworkisneededtoclearlydeterminethe influence of construction work and dredging versus hydrologicalinfluencealone onthedistributionof sedi-menttypes,butafirstapproachistoconsidermaintenance dredgingasabackgroundnoiseonsedimentation.

6. Conclusion

The classification of superficial sediment is an effectiveindicatortomonitortheevolutionofanestuary froma morphosedimentological pointof view.Prior to 1977 studies, the Seine estuary was globally made of sands,ascommonlydescribedinnaturalestuariesunder temperate climate. By comparison with this former situation, in 1996, a shifting to a mud-dominated sedimentation was described. This evolution was explained by the modification of river flow rate dynamics induced by new engineering works (dikes and dredging) in the navigation channel that had increasedthepowerofwinterriverflowrates, increas-ing the expelling and deposition of the turbidity maximum in the mouth of the estuary. According to themost recentsamplingcampaign (2009),data show anobviousreductionofthetotalsurface ofmudfacies area,balancedbyalargespreadingoffine-grained(silt and clay) sedimentation over the whole Seine estuary mouthandanincreaseinthesiltandclaycontentinthe proximal marine sandy deposit, suggesting an ‘‘intra-sedimentarysilting’’.

During the three described periods, the hydrological conditions were different: low river flow rates in the 1977study,mediumtohighflowratesforthe1993–1996 study,longperiodoflowflowratesintherecentstudyof 2009.Whentheflowrateishighandlongenough(usually each year), mud dominates over the mouth. On the contrary, when the dynamic of tide and swell is not dominatedby flow ratepower, estuaryrecoversa tide-dominatedsedimentarydynamics,involvingadispersalof mud,evenifthesiltandclayfractioninthesandremains high. Theimportanceofriver flowratein thesediment dynamics appears also very significant, even dominant since muddy feature of superficial sedimentation is controlled by the presence/absence of significant flow rates.

TheSeineestuaryisagoodexampleofaman-altered estuary,sincetherespective influenceofthethreemain hydrodynamicparameters(flowrate,tideandswell)have been modified. Modifications of morphology induce modifications ofsedimentary dynamics,in a proportion thatthesedimentationinthemouthoftheestuaryisnow mainlyinfluencedbyriverflows,asaprevalentlong-term signal on tidal currents, even in a megatidal context. Considering the fine-grained fraction in the superficial sedimentsanditsvariationsalongthelast150years,the Seineestuaryhasshiftedfromatide-dominatedtoa river-dominatedsystem.

References

Allen,G.P.,Salomon,J.-C.,Bassoulet,P.,DuPenhoat,Y.,DeGranpre,C.,

1980.Effectsoftidesonmixingandsuspendedsedimenttransportin

macrotidalestuaries.Sediment.Geol.26,69–90.

Avoine,J.,1987.SedimentexchangesbetweentheSeineestuaryandits

adjacentshelf.J.Geol.Soc.144,135–148.

Avoine,J.,Allen,G.P.,Nichols,M.M.,Salomon,J.-C.,Larsonneur,C.,1981.

SuspendedsedimenttransportintheSeineestuary,France.Effectof

made-manmodificationsonestuary-shelfsedimentology.Mar.Geol.

40,119–137.

Blott,S.J.,Pye,K.,VanderWal,D.,Neal,A.,2006.Long-termmorphological

changeanditscausesintheMerseyEstuary,NWEngland.

Geomor-phology81,185–206.

Brenon,I.,LeHir,P.,1999.ModellingtheTurbidityMaximumintheSeine

Estuary(France):identificationofFormationProcesses.Estuar.Coast.

ShelfSci.49,525–544.

Cuvilliez,A.,Deloffre,J.,Lafite,R.,Bessineton,C.,2009.Morphological

responsesofanestuarineintertidalmudflattoconstructionssince

1978 to 2005: the Seineestuary (France). Geomorphology 104,

165–174.

Dalrymple,R.W.,Zaitlin,B.A.,Boyd,R.,1992.Estuarinefaciesmodels:

conceptualbasisandstratigraphicimplication.J.Sediment.Petrol.62,

1130–1146.

Diesing,M.,Green,S.L.,Stephens,D.,Lark,R.,Stewart,H.A.,Dove,D.,2014.

Mappingseabedsediments:comparisonofmanual,geostatistical,

object-basedimageanalysisandmachinelearningapproaches.

Con-tinentalShelfRes.84,107–119.

Dubrulle,C.,Lesueur,P.,Boust,D.,Dugue´,O.,Poupinet,N.,Lafite,R.,2007.

Sourcediscriminationoffine-graineddepositsoccurringonmarine

beaches:theCalvadosbeaches(easternBayoftheSeine,France).

Estuar.Coast.ShelfSci.72(1–2),138–154.

Fairbridge,R.W.,1980.Theestuary:itsdefinitionandgeodynamiccycle.

In:Olausson,E.,Cato,S.L.I.(Eds.),Chemistryandbio-geochemistryof

estuaries.JohnWiley&Sons, pp.1–35.

Garnaud,S.,Lesueur,P.,Lesourd,S.,Poupinet,N.,Brun-Cottan,J.C.,2002.

Caracte´risationdesvariationslithologiquessaisonnie`resdes

se´di-mentsdelabaiedeSeineorientale;impactdescruesdelaSeine

(France).C.R.Geoscience334,323–330.

Garnaud,S.,Lesueur,P.,Clet,M.,Lesourd,S.,Garlan,T.,Lafite,R.,

Brun-Cottan,J.C.,2003.Holocenetomodernfine-grainedsedimentationon

amacrotidalshoreface-to-inner-shelf setting (eastern Bayof the

Seine,France).Mar.Geol.202,33–54.

Gue´zennec,L.,Lafite,R.,Dupont,J.P.,Meyer,R.,Boust,D.,1999.

Hydrody-namicsofsuspendedparticulatematterinthetidalfreshwaterzone

ofamacrotidalestuary(theSeineestuary,France).Estuaries22(3A),

717–727.

Hossain,S.,Eyreb,B.D.,McKeec,J.J.,2004.Impactsofdredgingondry

seasonsuspended sedimentconcentration in theBrisbane River

estuary,Queensland,Australia.Estuar.Coast.ShelfSci.61,539–545.

Lefebvre,D.,Huault,M.-F.,Guyader,J.,Giresse,P.,Hommeril,P.,

Larson-neur,C.,1974.Leprismealluvialdel’estuairedelaSeine:synthe`se

se´dimentologique,stratigraphiqueetpale´oge´ographique.Bull.d’Inf.

Ge´ol.Bass.deParis39,27–36.

LeHir,P.,Ficht,A.,SilvaJacinto,R.,Lesueur,P.,Dupont,J.P.,Lafite,R.,

Brenon,I.,Thouvenin,B.,Cugier,P.,2001.Finesedimenttransportand

accumulationsatthemouthoftheSeineestuary(France).Estuaries

24(6B),950–963.

Lesourd,S.,Lesueur,P.,Brun-Cottan,J.C.,Auffret,J.P.,Poupinet,N.,Laignel,

B.,2001.Morphosedimentaryevolutionofamacrotidalestuary

sub-jectedtohumanimpact;theexampleoftheSeine(France).Estuaries

24(6B),940–949.

Lesourd,S.,Lesueur,P.,Brun-Cottan,J.C.,Garnaud,S.,Poupinet,N.,2003.

Seasonalvariationsinthecharacteristicsofsuperficialsedimentsina

macrotidalestuary(theSeineinlet,France).Estuar.Coast.ShelfSci.58

(1),3–16.

Lesueur,P.,Lesourd,S.,Lefebvre,D.,Garnaud,S.,Brun-Cottan,J.C.,2003.

HoloceneandmodernsedimentsintheSeineestuary(France):a

synthesis.J.Quat.Sci.18(3–4),339–349.

Li,Z.H.,Nguyen,K.D.,Brun-Cottan,J.C.,Martin,J.M.,1995.Numerical

simulationoftheturbiditymaximumtransportintheGironde

estu-ary(France).Oceanol.Acta17(5),479–500.

Mitchell,S.B.,Couperthwaite,J.S.,West,J.R.,Lawler,D.M.,2003.

Measur-ing sediment exchangerates onan intertidalbank at Blacktoft,

HumberEstuary,UK.Sci.TotalEnviron.314–316,535–549.

Mitchell,S.B.,Uncles,R.J.,Akesson,L.,2012.Observationsofturbidityin

theThamesestuary.WaterEnviron.J.26,511–520.

Morris,R.K.A.,Mitchell,S.B.,2013.Haslossofaccommodationspacein

the Humber estuary led to elevated suspended sediment

concentrations? Int.J.Front.Construct.Eng.2(1),1–9.

Perillo,G.M.E.,1996.Definitionsandgeomorphologicclassificationsof

estuaries.In:Perillo,G.M.E.(Ed.),Geomorphologyandsedimentology

ofestuaries.Elsevier, pp.17–47.

Pye,K.,Blott,S.J.,2014.ThegeomorphologyofUKestuaries:theroleof

geological controls, antecedent conditions and human activities.

Estuar.Coast.ShelfSci.150,196–214.

Salomon,J.C.,Allen,G.P.,1983.Roˆlese´dimentologiquedelamare´edans

lesestuairesa` fortmarnage.CompagnieFranc¸aisedesPe´troles,Notes

etMe´moires18,35–44.

Tessier,B.,2012.Stratigraphyoftide-dominatedestuaries.In:Principles

oftidalsedimentology,pp.109–128.

Tessier,B.,Delsinne,N.,Sorrel,P.,2010.Holocenesedimentaryinfillingof

atide-dominatedestuarinemouth.Theexampleofthemacrotidal

Seineestuary(NWFrance).Bull.Soc.geol.France181,87–98.

Thomas,C.G.,Spearman,J.R.,Turnbull,M.J.,2002.Historical

morphologi-calchangeintheMerseyEstuary.Cont.ShelfRes.22,1775–1794.

Tonis,I.E.,Stam,J.M.T.,VandeGraaf,J.,2002.Morphologicalchanges

of theHaringvliet estuary afterclosure in1970. Coast. Eng. 44,

191–203.

Uncles,R.J.,Stephens,J.A.,Harris,C.,1998.Seasonalvariabilityofsubtidal

andintertidaldistributioninamuddy,macrotidalestuary:the

Hum-ber-Ouse,UK.In:Black,K.S.,Paterson,D.M.,Cramp,A.(Eds.),

Sedi-mentaryProcessesintheIntertidalZone.Geol.Soc.London,Spec.

Publ.139,pp.211–219.

VanderWal,D.,Pye,K.,Neal,A.,2002.Long-termmorphologicalchange

intheRibbleestuary,NorthwestEngland.Mar.Geol.189,249–266.

VanMaren,D.S.,Hoekstra,P.,2004.Seasonalvariationofhydrodynamics

andsedimentdynamicsinashallowsubtropicalestuary:theBaLat

River,Vietnam.Estuar.Coast.ShelfSci.60,529–540.

Wang,Y.,Dong,P.,Oguchi,T.,Chen,S.,Shen,H.,2013.Long-term(1842–

2006)morphologicalchangeandequilibriumstateoftheChangjiang

(Yangtze)estuary,China.Cont.ShelfRes.56,71–81.

Wells,J.T.,1995.Tide-dominatedestuariesandtidalrivers.

Geomorphol-ogyandsedimentologyofestuaries.In:Perillo,G.M.E.(Ed.), Dev.

Sediment.53,179–205.

Wolanski,E.,2006.Theevolutiontimescaleofmacro-tidalestuaries:

examplesfromthePacificRim.Estuar.Coast.ShelfSci.66(3–4),

544–549.

Wolanski,E.,Moore,K.,Spagnol,S.,D’Adamo,N.C.,Pattiaratchi,C.,2001.

Rapid,human-inducedsiltationofthemacro-tidalOrdRiverestuary,

WesternAustralia.Estuar.Coast.ShelfSci.53,717–732.

Woodruff, J.D., Geyer, W.R., Sommerfield, C.K., Driscoll, N.W., 2001.

SeasonalvariationofsedimentdepositionintheHudsonRiver