<%BANNER%>

Muddy Sea-Floor Response to Wave Action, Atchafalaya Shelf, Louisiana, USA

Permanent Link: http://ufdc.ufl.edu/UFE0044592/00001

Material Information

Title: Muddy Sea-Floor Response to Wave Action, Atchafalaya Shelf, Louisiana, USA
Physical Description: 1 online resource (91 p.)
Language: english
Creator: Sahin, Cihan
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2012

Subjects

Subjects / Keywords: atchafalaya -- cohesive -- concentration -- current -- floc -- mud -- nearshore -- ocean -- rheology -- sediment -- turbulence -- wave
Civil and Coastal Engineering -- Dissertations, Academic -- UF
Genre: Coastal and Oceanographic Engineering thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The interaction between muddy sea-floor and the hydrodynamics is investigated based on wave, current, sediment and acoustic backscatter observations made in 2006 and 2008 on the muddy Atchafalaya Shelf, Louisiana. A method to estimate the vertical structure of suspended sediment concentration from acoustic backscatter observations is developed using 2006 data. Acoustic estimates of suspended sediment concentration are used to re-examine the relation between sediment stratification and floc size in a cohesive sedimentary environment. Concentration measurements by optical backscatterance sensors at two vertical levels are used to calibrate the backscatter intensity of the acoustic profiler. In spite of the complexities due to the rapidly changing flow and cohesive sediment properties, the acoustic estimates of sediment concentrations agree well with the observations. Estimated vertical suspended sediment concentration profiles and the measured current profiles are used to calibrate a one-dimensional-vertical (1DV) boundary layer numerical model for combined wave-current flow on muddy beds. The effect of the floc size on the vertical structure of the suspended sediment concentration profile is investigated using the acoustic and numerical estimates of concentrations, and the floc size measurements. For similar  flow conditions, smaller flocs result in more mixed profiles with higher concentration in the upper water column and lower near-bed concentration. This observation based on field measurements supports the previously published numerical results. Wave, current, acoustic backscatter and suspended sediment concentration measurements (both single-point and vertical profiles estimated by conversion of acoustic backscatter data) collected in 2008 are used to investigate wave-current-cohesive sediment interaction on the muddy Atchafalaya inner shelf. During an energetic storm, bed state follows a cycle of dilation due to fluidization, erosion, deposition with fluid mud formation and consolidation. A one-dimensional-vertical cohesive sediment transport model is calibrated using current and concentration profiles to estimate the physical parameters that could not be measured directly, e.g., bottom stresses. Estimated bed position and computed bottom stresses suggest that the critical erosion threshold is in the range of 0.3 Pa to 0.5 Pa. The study site is impacted by a sediment-laden fresh water plume coming from the Atchafalaya River mouth. Bed density evolution during the storm is estimated from vertical sediment exchange between the water column and the bed excluding the duration of passage of a sediment-carrying water front. The values are in the range of 1,030 kg/m$^{3}$ to 1,200 kg/m$^{3}$ and indicate that the bed density increases during the erosion phase and decreases during deposition. At the end of the storm, it shows a steady increasing trend during hindered settling and exceeds the space-filling value during consolidation. Both the critical erosion shear stress and bed density values are consistent with the results of laboratory tests on samples from the experimental site. The applicability of these results for a larger population of storms is also investigated. The results generalize the estimates of bed yield stress and bed density evolution, allow for a statistical model for the bed reworking cycle and represent a first step towards a forecasting model for wave-bed coupling in muddy environments.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Cihan Sahin.
Thesis: Thesis (Ph.D.)--University of Florida, 2012.
Local: Adviser: Sheremet, Alexandru.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2012
System ID: UFE0044592:00001

Permanent Link: http://ufdc.ufl.edu/UFE0044592/00001

Material Information

Title: Muddy Sea-Floor Response to Wave Action, Atchafalaya Shelf, Louisiana, USA
Physical Description: 1 online resource (91 p.)
Language: english
Creator: Sahin, Cihan
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2012

Subjects

Subjects / Keywords: atchafalaya -- cohesive -- concentration -- current -- floc -- mud -- nearshore -- ocean -- rheology -- sediment -- turbulence -- wave
Civil and Coastal Engineering -- Dissertations, Academic -- UF
Genre: Coastal and Oceanographic Engineering thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The interaction between muddy sea-floor and the hydrodynamics is investigated based on wave, current, sediment and acoustic backscatter observations made in 2006 and 2008 on the muddy Atchafalaya Shelf, Louisiana. A method to estimate the vertical structure of suspended sediment concentration from acoustic backscatter observations is developed using 2006 data. Acoustic estimates of suspended sediment concentration are used to re-examine the relation between sediment stratification and floc size in a cohesive sedimentary environment. Concentration measurements by optical backscatterance sensors at two vertical levels are used to calibrate the backscatter intensity of the acoustic profiler. In spite of the complexities due to the rapidly changing flow and cohesive sediment properties, the acoustic estimates of sediment concentrations agree well with the observations. Estimated vertical suspended sediment concentration profiles and the measured current profiles are used to calibrate a one-dimensional-vertical (1DV) boundary layer numerical model for combined wave-current flow on muddy beds. The effect of the floc size on the vertical structure of the suspended sediment concentration profile is investigated using the acoustic and numerical estimates of concentrations, and the floc size measurements. For similar  flow conditions, smaller flocs result in more mixed profiles with higher concentration in the upper water column and lower near-bed concentration. This observation based on field measurements supports the previously published numerical results. Wave, current, acoustic backscatter and suspended sediment concentration measurements (both single-point and vertical profiles estimated by conversion of acoustic backscatter data) collected in 2008 are used to investigate wave-current-cohesive sediment interaction on the muddy Atchafalaya inner shelf. During an energetic storm, bed state follows a cycle of dilation due to fluidization, erosion, deposition with fluid mud formation and consolidation. A one-dimensional-vertical cohesive sediment transport model is calibrated using current and concentration profiles to estimate the physical parameters that could not be measured directly, e.g., bottom stresses. Estimated bed position and computed bottom stresses suggest that the critical erosion threshold is in the range of 0.3 Pa to 0.5 Pa. The study site is impacted by a sediment-laden fresh water plume coming from the Atchafalaya River mouth. Bed density evolution during the storm is estimated from vertical sediment exchange between the water column and the bed excluding the duration of passage of a sediment-carrying water front. The values are in the range of 1,030 kg/m$^{3}$ to 1,200 kg/m$^{3}$ and indicate that the bed density increases during the erosion phase and decreases during deposition. At the end of the storm, it shows a steady increasing trend during hindered settling and exceeds the space-filling value during consolidation. Both the critical erosion shear stress and bed density values are consistent with the results of laboratory tests on samples from the experimental site. The applicability of these results for a larger population of storms is also investigated. The results generalize the estimates of bed yield stress and bed density evolution, allow for a statistical model for the bed reworking cycle and represent a first step towards a forecasting model for wave-bed coupling in muddy environments.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Cihan Sahin.
Thesis: Thesis (Ph.D.)--University of Florida, 2012.
Local: Adviser: Sheremet, Alexandru.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2012
System ID: UFE0044592:00001


This item has the following downloads:


Full Text

PAGE 2

2

PAGE 3

3

PAGE 4

IwouldliketothankmyadvisorDr.AlexSheremetforbeingagreatmentorandverysupportivethroughoutmydoctoralstudy.Ihavealwaysfeltveryluckytobehisstudent.IwouldalsoliketothankIlgarSafaknotonlyforhisincrediblehelpaboutmyworkbutalsobeingagreatfriend.Thismuchworkcouldnothavebeendonewithouthishelp.IamgratefultoDr.AshishMehtaformanyconstructiveandinsightfuldiscussionsonthetopicofthisdissertation.IwouldalsoliketothankDr.Tian-JianHsufromUniversityofDelawareforsharinghisnumericalmodelfornumericalsimulations.Thisstudyhasconsiderablybenetedfromthismodel.IalsopresentmythankstomycommitteemembersDr.DonaldN.Slinn,Dr.ArnoldoValle-Levinsson,andDr.JohnJaegerfortheirsupport,andcontributionsthroughoutmystudy.Iwanttothankallthestudentsinthecoastalgroup,especiallymyofcematesTracyMartz,MiaoTianandUriahGravois,formakingthisprogramfunandafriendlyenvironment.IacknowledgetheCouncilofHigherEducation,TurkeyandtheYildizTechnicalUniversityforthescholarshipofferedduringmydoctoralstudy.Lastly,butnotleast,Iwishtothankmywonderfulfamilyforsupportingandencouragingmeineverystageofmylife. 4

PAGE 5

page ACKNOWLEDGMENTS .................................. 4 LISTOFTABLES ...................................... 7 LISTOFFIGURES ..................................... 8 ABSTRACT ......................................... 10 CHAPTER 1INTRODUCTION ................................... 12 2FIELDEXPERIMENT ................................ 18 2.1ExperimentalSite ............................... 18 2.2Instrumentation ................................. 21 3OBSERVATIONSOFSUSPENDEDSEDIMENTDYNAMICS .......... 25 3.1Motivation .................................... 25 3.2AcousticBackscatterIntensityConversiontoSSC ............. 26 3.2.1ConversionAlgorithm ......................... 26 3.2.2FieldApplication ............................ 32 3.2.3Results ................................. 36 3.3NumericalModeling .............................. 37 3.3.1GoverningEquations .......................... 37 3.3.2Application ............................... 41 3.4EffectofFlocSizeonConcentrationProles ................. 42 4COUPLEDWAVE-COHESIVEBEDDYNAMICS ................. 48 4.1Motivation .................................... 48 4.2Observations .................................. 48 4.3ReconstructionofWater-columnProcesses ................. 55 4.3.1ConversionofAcousticBackscattertoSSCprole ......... 55 4.3.2NumericalSimulations ......................... 56 4.4Results ..................................... 58 4.4.1CriticalShearStressforErosion ................... 58 4.4.2MassBalanceConsiderations ..................... 60 4.5VariationwithDifferentStormConditions ................... 65 5CONCLUSIONS ................................... 72 APPENDIX ATHEORETICALBACKGROUNDTOACOUSTICMETHODOLOGY ...... 77 5

PAGE 6

........................ 80 CYIELDSTRESSMEASUREMENTS ........................ 83 REFERENCES ....................................... 85 BIOGRAPHICALSKETCH ................................ 91 6

PAGE 7

Table page 2-1Locationoftheinstrumentedplatforms[ Safak 2010 ] .............. 23 C-1Yieldstressestimatesfordifferentsedimentvolumefractions .......... 84 7

PAGE 8

Figure page 2-1Theapproximatelocationofthetripodsdeployedin2006and2008ontheAtchafalayaShelf. .................................. 19 2-2CongurationofinstrumentsinA)2006;B)2008. ................. 22 3-1Schematicoftheapplicationofacousticstomeasuresedimentconcentration. 26 3-2Timeevolutionofsignicantwaveheight,currentspeedproles,verticallyaveragedcurrentdirections,normalizedacousticbackscatterintensity,suspendedsedimentconcentrationandmeanocsize .......................... 34 3-3A)Thehistogramofsuspendedsedimentconcentration.B)Variationofktwithsuspendedsedimentconcentration. ...................... 36 3-4A)Thehistogramofocsizeobservations.B)Variationofktwithocsize. .. 37 3-5Timeevolutionofmeanocsizeandtheoptimumocsize,suspendedsedimentconcentrationatthelocationoftheOBSsandPC-ADPderivedsuspendedsedimentconcentrationproles. .......................... 38 3-6Timeevolutionof:A)suspendedsedimentconcentrationestimates(PC-ADPbackscatter);B)suspendedsedimentconcentrationcalculationsofthemodel. ............................................. 43 3-7Timeevolutionof:A)currentspeedmeasurements(PC-ADP);B)speedcalculationsofthemodel. ..................................... 44 3-8A)TwoobservedSSCverticalprolesandcorrespondingmodelsimulations.B)Correspondingobservedandsimulatedcurrentspeeds.Verticalprolesof:C)theRichardsonnumber;D)thenormalizedturbulenceintensity. ..... 46 4-1EvolutionofwavefrequencyspectrumandpropagationdirectionduringthestormofMarch3rdto5th,2008. .......................... 49 4-2StormofMarch3rdto5th,2008. .......................... 51 4-3ThevariationofA)salinityandB)temperaturewithdifferentwaterdepths. .. 52 4-4EvolutionofbedpositionindicatorsandstateduringthestormofMarch3rdto5th,2008. ..................................... 53 4-5ComparisonofSSCvaluesestimatedusingthePCADPacousticbackscatterandopticalobservations. .............................. 56 4-6ModelsimulationsforthestormofMarch3rd-5th,2008. ............ 59 8

PAGE 9

...................................... 62 4-8AsummaryofthebedreworkingcycleduringthestormofMarch3rdto5th,2008. ......................................... 64 4-9Idealizedsketchofbedreworkingduringthestorm. ............... 65 4-10StormofFebruary26thto28th,2008. ....................... 67 4-11EvolutionofbedpositionindicatorsandstateduringthestormofFebruary27thto28th. ..................................... 68 4-12AsummaryofbedreworkingcycleduringthestormofFebruary27th. ..... 69 4-13Observedmass(water-column)/volume(bed)uxvs.numericallysimulatedbedstress. ...................................... 70 4-14AsummaryofthebedreworkingcycleduringthestormofFebruary27th,2008. 71 B-1SampleofPC-ADPobservationsonMarch3rd,13:00hours,duringaperiodwhenthebedcanbeconsideredasstationary. .................. 81 B-2SampleofPC-ADPobservationsonMarch4th,12:00hoursUTM,atthemaximumerosionaldepth .................................... 82 C-1Upper-Binghamyieldstressasafunctionofthesolidsvolumefractionfordifferenttypesofsediment. .................................. 84 9

PAGE 10

10

PAGE 11

11

PAGE 12

WinterwerpandvanKesteren 2004 ]. Mehta [ 2002 ]givesoneofthemostcomprehensivedenitionsofmudasfollowing:sedimentwatermixturecomposedofgrainsthatarepredominantlylessthan63minsize,exhibitingarheologicalbehaviorthatisporo-elasticorvisco-elasticwhenthematrixisparticle-supported,andishighlyviscousandnon-Newtonianwhenitisinauidlikestate.Physico-chemicalcharacteristicsofne-grainedcohesivesedimentparticlescauseaggregationoftheseparticlesformingtheunitsknownasocs.Thisprocessiscalledocculationwhichcanalsobeenhancedbytheorganicmatterfoundinwater.Hence,cohesivesedimentparticlesareexibleocswhichcanaggregateorbreak-down,dependingonvariousexternalfactors(suchasturbulentshear,concentrationandsalinity).Thephysicalpropertiesofocdiffersignicantlyfromthepropertiesofindividualparticlesformingtheocssincelargewatercontentofocsmakesthemhaveopenstructureswithdensitiesslightlylargerthanthatofwater.Becauseoftheir 12

PAGE 13

1996 2000 southwestcoastofIndiaandEastCoastofChina; Cacchioneetal. 1995 theAmazonDelta;Traykovskietal., 2000 2007 EelRiverinNorthernCalifornia,andthePoDelta; Allisonetal. 2000 AtchafalayaShelf,GulfofMexico;andmanyothers)suggestthatastrongcouplingexistsbetweenhydrodynamicsandcohesivebeddynamics.Bottomstressesinducedbyenergeticwavescanresultinbulkstressesthatexceedtheyieldthresholdandreworkthebedsedimentthroughacombinationofprocessessuchasbedliquefaction,uidization(swellingduetomixingwithwater),erosion,anddeposition.Underhindered-settlingconditions,adenseandviscouslayerofuidmudcanforminthevicinityofthebed,andinducesubstantialwaveenergydissipation[ JiangandMehta 1996 SheremetandStone 2003 Allisonetal. 2005 Sheremetetal. 2005 Winterwerpetal. 2007 Sheremetetal. 2011 ].Fortheinvestigationofwave-mudinteraction,themuddyseaoorhasbeenmodeleddependingonitsrheologicalresponsetowaveforcing.Themostcommondescriptionofthestateofthecohesivebedusedinwave-sedimentinteractionstudies,i.e.,Newtonianviscous-uid(e.g., Gade 1958 DalrympleandLiu 1978 Ng 2000 ),hasbeencriticizedforover-estimatingmudviscosity[ MaaandMehta 1990 ].Alternativemodelsproposedtocorrectthis,suchassingle-ormulti-layeredvisco-elasticmodels[ HsiaoandShemdin 1980 MaaandMehta 1990 Fodaetal. 1993 ],orvisco-plasticones[ LiuandMei 1989 1993 ChanandLiu 2009 ],areassumedtoprovideaphysicaldescriptionclosertotheeldreality.Visco-elasticmodelscanaccountforbothliquidandelasticphasesofmud;visco-plasticmodelsprovideabetterrepresentationofhigh-densitymudphasesduringtheincipientstagesofbed 13

PAGE 14

JainandMehta [ 2009 ]suggestedthedomainofapplicabilityofdifferentrheologicalmodelsbasedonthePecletnumber(theratioofrateofadvectionofowtoitsrateofdiffusionwithinthebed)andthebeddensity(orthesolidsvolumefraction).Assumingthatthebedisconsolidatedbeforethestorm,startingformviscoplastic,theapplicablemodelchangesthroughviscoelasticsolid,viscoelasticuidandtheviscousuidwithdecreasingbeddensityduringthestorm.However,supportiveobservationaldataremainscarce.Onereasonistheconsiderabledifcultyinobservingdirectlyandcontinuouslytheevolutionofbed-sedimentstateduringinterestingevents(storms).Anotheristhefactthatmethodologiesforin-situobservationsofcoupledhydrodynamicsandsediment-transportprocessesarefarfrommature.Inordertoachieveabetterunderstandingoflargescalecirculationandsedimenttransportprocesses,estimatingverticalstructuresofsuspendedsedimentconcentrationforvaryingowconditionsisessential.Fieldexperimentsonwave-cohesivesedimentcoupling[e.g, Kinekeetal. 2006 Jaramilloetal. 2009 Safaketal. 2010 Sheremetetal. 2011 ]monitoredowandsedimentmainlyusingopticalsensorscalibratedforestimatingthelocalsuspendedsedimentconcentration(SSC),andacousticsensorsmainlyforrecordingtheowvelocity(assinglepointmeasurementsorverticalproles).OpticalsensorsprovideafewSSCmeasurementpointsthatareingeneralnotdenseenoughforasatisfactorycharacterizationofthesedimentcontentinthewatercolumn.Acousticprolershavebeenusedsuccessfullyinrecentyearsforestimatingtheverticalstructureofsuspendedsedimentconcentration(SSC)insandyenvironments.Theyprovidehightemporalandspatialresolution,aredeployableinhighenergyconditions(whentraditionalwatersamplingbecomesdifcult)anddonotdistorttheow,asthemeasurementsarecollectedatdistance[ Lynchetal. 1991 Thorneetal. 1993 Thostesonetal. 1998 ThorneandHanes 2002 ].Toobservenear-bedsedimenttransport,theinstrumentistypicallymountedatabout1-2meterabovebed(mab), 14

PAGE 15

ThorneandHanes 2002 ].Theanalysisiscomplicatedbythestrongdependencyoftheacousticscatteringprocessesoninstrumentcharacteristics,aswellasenvironmentalfactors,suchassedimentconcentration,structure,andspecicsofsediment-soundinteraction,andothers.Effortstoapplythismethodologyinmud-dominatedenvironmentshavebeenhamperedbytheadditionalcomplexitiesofcohesivesedimentcharacteristics(e.g.,variableparticlesizeanddensityduetotheocformationandbreakup)thatarealsostronglycorrelatedtoowturbulenceandamountofsedimentinsuspension.Recentstudies[ Gartner 2004 HoitinkandHoekstra 2005 ]focusedontheperformanceofacousticprolersindilutecohesivesedimentconcentrations(uptotheorderof0.1kg/m3).Itisstillunclearhowaccuratethesemethodsareinhigherconcentrationsand/orhighconcentration-gradientconditions(e.g.,lutocline,uidmudlayer). Hamiltonetal. [ 1998 ]usedanacousticprolertoestimateSSCinsolutionsofupto10g/L;however,theirestimateswerenotcomparedwithanyindependentsourceofinformationonconcentration.Inmuddyenvironments,verticalprolesofSSCaretypicallyestimatedusingnumericalmodels[e.g., Hsuetal. 2009 Winterwerp 2001 2002 ]thattakeasinputasmallnumberofpointSSC(forexample,derivedfromcalibratedopticalsensorsdeployedatdifferentheightsabovethebed).Thesemodelsareessentialforstudyingow-relatedparametersthataredifculttomeasuredirectly,suchasnear-bedturbulentstresses.Thequalityandresolutionoftheobservationaldatausedtoforcethesemodelsis,ofcourse,amajorconcern.Forexample, Safaketal. [ 2010 ]showedthatalimitedverticalresolution(1-2pointmeasurements)canbematchedwiththemodelwithdifferentverticalproles,i.e.,usingdifferentmodelparameters(e.g.,differentsettlingvelocityorocsize).Thechoiceofmodelparameterstocalibratethemodel 15

PAGE 16

Safaketal. [ 2010 ]studiednumericallytheeffectofocsizeontheverticalSSCproleandinferredthatdecreasingocsize,whichresultsinadecreaseinsettlingvelocity,producesmoremixedconcentrationprolewithsmallernear-bedsedimentconcentration.Sincetherelationshipbetweenocsizeandsettlingvelocity,andtheassociatedattenuationofturbulencethroughsediment-induceddensitystraticationarepartofthemodelclosures,thisresultstillneedsconrmationbasedonmorecomprehensiveeldobservations.Thisstudyrepresentsanefforttoassembleobservationsofnear-bedsedimentandhydrodynamicsmadein2006and2008intoareconstructeddescriptionofwave-current-sedimentinteractionontheAtchafalayainnerShelf,Louisiana,USA(Chapter 2 ).Informationonvariationofverticalsuspendedsedimentconcentrationprolesisoneofthekeyelements,howevermissinginpreviousstudies,todrawmoreclearresultsonwave-muddyseabedcoupling.Therefore,amethodtoestimateverticalprolesofSSCfromtheacousticbackscatterintensityofcurrentprolersincohesivesedimentaryenvironmentsinrelativelyhighconcentrations(upto10kg/m3)isdeveloped,presentedandvalidatedinChapter 3 .TheimportanceofhavinghighresolutionSSCproleestimatestocalibratenumericalmodelsforobtainingmorereliablenumericalresultsisalsoemphasizedinthissection.Bothestimatedconcentrationprolesandnumericallycomputedparametersareusedtogethertoinvestigatetheroleofocsizeonsedimentstratication.InChapter 4 ,theevolutionofbedstateandpropertiesthroughoutenergeticstormsisinvestigatedbytakingtheadvantageofhavingverticalSSCinformation,togetherwithobservationsofwaves,owvelocities,andsea-bedposition.Physicalparametersthatcannotbemeasureddirectly,suchasturbulentshearstressesarecalculatedusingaone-dimensional-vertical 16

PAGE 17

Hsuetal. 2009 ]thatwascalibratedusinghighresolutionowandSSCproles.Sedimentcharacteristicparameters(e.g.,yieldstressandbulkdensity)obtainedfrompreviouslaboratorytestsonsamplesfromtheexperimentalsite[ Robillard 2009 ]areusedtovalidatethendings(e.g.,criticalshearstressforerosion,beddensityattheonsetoferosion).Theimplicationsoftheresultsarediscussedinrelationtobedresponsetohydrodynamicforcingandtheimportanceofsedimentresuspensionrelativetoadvection.Asarststeptowardsbuildingastatisticalmodelforthebedreworkingcycleandaforecastingmodelforwavebed-couplinginmuddyenvironments,thevariationoftheprocessforanotherstormisalsoinvestigated.Chapter 5 summarizestheresultsofthisstudy,anddiscussespossiblefuturecontributionstocohesivesedimenttransportliteratureinlightoftheseresults. 17

PAGE 18

2-1 )in2006and2008.Theseexperimentswerepartsofalarger-scopestudyofhydrodynamicsandsedimenttransportprocessesinshallowmuddyenvironments.Thewideandshallow(the10-misobathisabout40kmoffshore)Atchafalayashelfhasbeensubjectofseveralstudiesrelatedtowave-cohesivesedimentinteractionduetoitssedimentarypropertiesand,energeticandvariabilewaveclimate[ Allisonetal. 2000 SheremetandStone 2003 Allisonetal. 2005 Sheremetetal. 2005 2011 Drautetal. 2005 Kinekeetal. 2006 Jaramilloetal. 2009 Safaketal. 2010 Safak 2010 Sahinetal. 2011 2012a b ].Thesub-aqueousfeatureonthemuddyinnershelffrontingAtchafalayaBayisdenedasaclinoformofupto3-mthickmudlayer.Theclinoformextendsouttothe8-misobathattensofkilometersoffshore[ NeillandAllison 2005 ].Theinnershelfreceivesabout30%ofthedischargeoftheMississippiRiver,i.e.,approximately84millionmetrictonsofsediment,annually[ Mossa 1996 ],witharepresentativegrain-sizeatthesiterangingbetween2and7m,with17%ne-sandcontent[ Safaketal. 2010 Sheremetetal. 2005 Allisonetal. 2000 2005 ].InspiteofrisingsealevelsandthefactthatmostoftheLouisianacoastlineiseroding,landaccretesverticallyontheshelfduetothissedimentdischargeoverthelastfewdecades[e.g., Drautetal. 2005 ].Sedimentcharacteristicsovertheshelfvarysignicantlyshowingalarge-scalealong-shelfningofAtchafalayaRiverderivedsediments.ThebedsedimentconsistssandyandclayeysiltsintheEastneartheAtchafalayadredgedshipchannel(silt/claycontentofgreaterthan3%);whileintheWestandalongthecheniercoast,thecompositionisdominatedbysiltyclayswithlessthan5%sandandcoarsesilt 18

PAGE 19

Theapproximatelocationofthetripodsdeployedin2006and2008ontheAtchafalayaShelf.Thexmarksthelocationoftheinstrumentedplatforms. (silt/claycontentabout0.2-0.5%)[ NeillandAllison 2005 Drautetal. 2005 ]. WalkerandHammack [ 2000 ]observedthatcoarsematerials(sandandcoarsesilt)aredepositedclosertotheriverinesourceduringresuspensionevents,whilewavesandcurrentstransportednersedimentsfurtherwestward,withtheonlyexceptionoftheperiodscharacterizedwiththecoldfrontpassagesduringwhichsurfaceplumesedimentsareadvectedoffshorebypost-frontalnortherlywinds. Jaramilloetal. [ 2009 ]presentedtheresultsofthesedimenttypedistributionanalysisbasedonthesamplesbytheNavalResearchLaboratory,StennisSpaceCenter(NRL)on3-5May2006usingasurfacegrabsampler.Theirresults(Figure2in Jaramilloetal. [ 2009 ])supportedthequalitativepictureinFigure 2-1 thattheShipShoalareaandclosetotheTrinityShoaltypicallyoffshoreofthe6-misobatharedominantlysand(upto95%),overallthesurcialsedimentontheAtchafalayaShelfintheexperimentalareaappearstobemixtureof 19

PAGE 20

Allisonetal. 2000 ].ThisactivitycoincideswiththeperiodofhighsedimentdischargeoftheAtchafalayaRiver.Theseperturbationscangenerateswellsinexcessof1-mheightlastingforseveraldays.Inshallowwater,suchintensewaveactivitycausessignicantvariationsofthebedstatethroughoutastorm,inasequenceofbreakingdownstratication,triggeringbedliquefaction,increasingsedimentresuspensionandturbiditythroughoutthewatercolumn[ Allisonetal. 2000 ],followedbytheformationofwave-dissipativeuidmudlayerswithinone-mnearthebedduetosettling,andnallyconsolidationtoasoftbed.Thesehighconcentrationuidmudlayerswereobservedtobetransportedovertheshelfbyvariousmechanisms:westwardbyresidualcurrentsofabout10cm/swhichmayaccountforadvectionofmorethanhalfofthesedimentdischargeintotheshelf[ WellsandKemp 1981 ];onshorebycoastalupwelling[ Kinekeetal. 2006 ];andoffshoreintheformofaturbiditycurrentreaching5cm/s,whichismaintainedinsuspensionbywave-inducedturbulence[ Jaramilloetal. 2009 ].Twocompetingprocessesareassumedtogeneratetheseows:advectionofsedimentbytheplumeoftheAtchafalayaRiver;andlocal,wave-inducedbedreworking.Therelationbetweenthesetwomechanismsisnotwellunderstood. 20

PAGE 21

3 ).Thisdatasetwasselectedtotestthemethod,becauseitcontainstheinformationaboutsizedistributionsofsuspendedsedimentswhichisneededtoexecutetoSSCcalculationalgorithm,andallowsforinvestigationofthecapabilityofacousticstoestimatetheocsize.Detaileddiscussionontheexperimentsiteandinstrumentationabout2006datasetcanbefoundin Jaramilloetal. [ 2009 ].Here,instrumentationusedinthisstudywillbedetailed.Instrumentplatformwasdeployednearthe5-misobathbetweenFebruary28th-March14th('T2'in Jaramilloetal. [ 2009 ],locationcorrespondingto'Platform3'inFigure 2-1 ).TheschematicofthecongurationoftheinstrumentedplatformisshowninFigure 2-2 a.Near-bedcurrentvelocitiesweremeasuredusingadownward-pointing,1500-kHzPC-ADP(Pulse-CoherentAcousticDopplerProler,manufacturedbySontek/YSI)whichusesthreebeamsoriented15oofftheverticalaxis.ThePC-ADPsampledat2-Hzin60binsof2-cm,followinga10-cmblankingdistancein10-minburstsevery30-min.ThePC-ADPalsologgedSSCobservationscollectedbytwosynchronizedOBS-3s(OpticalBackscatteranceSensors,D&AInstruments,CampbellSci.)mountedat50and75cmab(cmabovethebed).TheOBS-3swerecalibratedinthelaboratoryusingsedimentandwatersamplescollectedattheexperimentsite.SizedistributionsofsuspendedsedimentswereestimatedbasedonobservationsusingaLISST-100XType-C(LaserInSituScatteringTransmissometer,SequoiaScientic)mountedat120cmabwhichestimatessizedistributionsofsuspendedparticles(ocsandprimary)at32classrangesbetween25-500m.Theinstrumentrecordedthesizedistributionofsuspendedsedimenteveryminute(averageof100,2-Hzsamples)in30-minburstseachhour.ThePC-ADPpressuretimeseriessegmentsof10-minlengthwerede-trendedandde-meaned,thendividedinto128sblockswith50%overlap,andtaperedusinga 21

PAGE 22

CongurationofinstrumentsinA)2006;B)2008.Circlesmarkthelocationofthesamplingvolumesofpointmeasurements,arrowsindicatethedirectionofacousticsignals. Hanningwindow.Theresultingspectrahaveapproximately17degreesoffreedom.Thesignicantwave-heightHswasestimatedbasedontherelationH2s=16Rf2f1S(f)dfwhereSisthepowerspectraldensityofseasurfaceelevationatfrequencyf,estimatedusingstandardspectralanalysis.Swascorrectedfordepthattenuationusingthelinearwavetheorytheory,withahigh-frequencycutoffdenedbyadepthattenuationofwavevariancelargerthan95%.Aspectraltailproportionaltof5wasaddedtocoverthehighfrequencyrange.Swell(longwaves)andsea(shortwaves)bandsweredistinguishedbyusingacutofffrequencyoffc=0.2Hz,e.g.,forswellbandf1=0.0078Hz,f2=fc;forseabandf1=fc,f2=0.992Hz.Representativeowandsedimentstatistics(meanspeedanddirectionofcurrents,wavespectrum,signicantwaveheight,andmeanconcentration)werecalculatedfor30-minintervals,resultingina281-pointtimeseriesofmeanvaluesforPC-ADPandOBSdata.ThemeasurementintervalfortheLISSTdatawas1-hour(atotalof141mean-valueobservations). 22

PAGE 23

Locationoftheinstrumentedplatforms[ Safak 2010 ] PlatformLatitude(North)Longitude(West)Depth(m) 129O11.815'91O36.731'7.4229O13.439'91O34.807'5.0329O15.574'91O34.267'3.8429O22.238'91O46.922'4529O33.759'92O33.800'4 Inthe2008experiment,veinstrumentedplatformsweredeployedonFebruary22nd,andcollecteddataforsix-weeks.Platforms1-3werealignedinacross-shoretransectbetween4and7.5-mwaterdepths.Thelocationofplatform4wasinanalongshoretransectwiththeshallowplatform3.Platform5wasatabout4-mwaterdepthnearFreshWaterBayouabout60mileswestoftheAtchafalayaBay.Theexactlocationoftheplatforms,andthemeanwaterdepthattheplatformlocationsaregiveninTable 2-1 .Moredetailsofthefullexperimentaregivenby Safak [ 2010 ]and Safaketal. [ 2010 ].Becausethegoalofthisstudyistoexaminetheresponseofsea-oortowaveaction,wefocushereonobservationscollectedbyasingleinstrumentedplatform,platform3,locatednearthe4-misobath(Figure 2-1 ).Thelocationoftheinstrumentedplatformcoincideswiththatofthe2006measurements.ItisassumedherethatthesedimentpropertiesneededforSSCcalculationsandnumericalsimulations(e.g.,ocsizerange,fractaldimension,seevaluesinSection 3.2.1 andSection 3.3.2 )werepracticallythesameinthetwoexperiments.Theverticalstructureoftheowvelocityintherstmab(meterabovethebed)wasobservedusingaPC-ADP(Pulse-CoherentAcousticDopplerProler,Sontek/YSI),thatsampledat2-Hzcontinuouslyin27binsof3.2cmwitha30-cmblankingdistance.DirectSSCobservationswereprovidedbyanOBS-5(OpticalBackscatteranceSensors,D&AInstruments,CampbellSci.)thatrecorded2-minaveragesofturbidity,andanOBS-3(D&AInstruments,CampbellSci.)samplingsynchronouslywiththePC-ADP.Upperwater-columncurrentsandsurfacewaveswereobservedusinga1200-kHz,ADCP(AcousticDopplerCurrentProler,TeledyneRDInstruments).Currentproles 23

PAGE 24

3 andChapter 4 ,respectively. 24

PAGE 25

3.2 )inrelativelyhighconcentrations(upto10kg/m3).ThebackscattermeasurementsofPC-ADPmountedat1.1-meterabovebed(mab)areusedinconversion.ThesignalreectedbysedimentsuspendedatdifferentelevationsisrelatedtosedimentconcentrationasillustratedinFigure 3-1 .Conversionofacousticbackscatterintensitytomassconcentrationisanindirectmethodandnotstraightforward.Acousticsignalstrengthneedstobecorrectlycompensatedfortransmissionlossesfrombeamspreadingandattenuationthatdependonrange,concentrationandotherenvironmentalcharacteristics(e.g.,salinity,temperature,pressure).Inthepresentedmethod,detailedexplanationforcalculationsoftransmissionlossesduetowaterandsedimentattenuation,thereforethesedimentconcentration,isgiven.Thehigh-resolutionSSCproleestimatesareusedinSection 3.3 toconstrainawave-currentboundarylayermodelforcohesivebeds[ Hsuetal. 2009 ].Themodelusesatime-dependentRANS(Reynolds-averagedNavier-Stokes)formulationbasedonatwo-equationk"closure.Suspendedsedimentdynamicsaremodeledusingadvection-diffusionequationcoupledwiththeowequations,thataccountsforsediment-induceddensitystraticationeffects.Thesedimentphaseisdenedinthemodelbyocdensity,andthusbytheprimaryparticlesize,ocsize,andocfractaldimension(assumedconstant).SettlingvelocityismodeledusingtheStokeslawwithhinderedsettlingeffectincorporated. 25

PAGE 26

Schematicoftheapplicationofacousticstomeasuresedimentconcentration. InSection 3.4 ,bothacousticandnumericalresultsareusedtoinvestigatetheroleoftheocsizeonverticalSSCproletoconrmthepreviouslypublishednumericalresults[ Safaketal. 2010 ]. 3.2.1ConversionAlgorithmForsphericalparticles, ShengandHay [ 1988 ],andlater Thorneetal. [ 1993 ]; ThorneandHanes [ 2002 ]showedthattheverticalproleofsuspendedsedimentconcentration,SSC(r)canbedeterminedfromthebackscatterintensityofanacousticproler(inthisstudy,thePC-ADP,seeFigure1b)as 26

PAGE 27

Downingetal. 1995 ],ksrepresentsthescatteringpropertiesofthesediments,ktisasystemconstant,denotestheattenuationcoefcient,ffistheformfunctionthatdescribesthebackscatteringcharacteristicsofparticlesinsuspension,aistheradiusofthesedimentinsuspension,anddenotesthesedimentdensity.Theangularbracketsrepresenttheaverageovertheparticle-sizedistribution.Theattenuationcoefcient=w+shasawatercomponentwandasedimentcomponents.Waterattenuationisafunctionofacousticfrequencyf(kHz),watertemperatureT(oC),depthz(m),andsalinityS(psu)(e.g., FrancoisandGarrison 1982a,b KayeandLaby 1986 ):

PAGE 28

c(1+.025T),(3)P2=11.37104z+6.2109z2,f2=8.1710f81990=(273+T)g 8.684.9371042.59105T+9.11107T21.50108T3,(3)ForT>20oC,A3=1 8.683.9641041.146105T+1.45107T26.50108T3,withP3=11.83105z+4.91010z2.Sedimentattenuationcanbecalculated[ Urick 1948 ThorneandHanes 2002 ]as: ThorneandHanes 2002 ]: 4haihi,(3)withthenormalizedtotalscatteringcross-section.Theparametersandffcanbecalculatedusingthesimpliedexpressionsgivenbasedonthemeasurementsoftheseparametersforsedimentsuspensionsandindividualirregularlyshapedparticlesfromseveralsources[ ShengandHay 1988 CrawfordandHay 1993 Thorneetal. 1993 28

PAGE 29

, 2008 ].Theviscousabsorptioniscalculatedasfollowing[ Urick 1948 ]: 4a1+1 w,=1 21+9 2a,=r Urick 1948 Haet.al. 2011 ].Consideringthefactthatthemeasuredocsarelargerthan100mthroughoutthedataset,scatteringcomponentmaybeamplieddependingonthechangeinocsize.However,thesizeofsingleparticlesformingtheocsareverysmallcomparedtotheacousticwavelength(particlesizeofDp=5m,ka=0.015)sothattheycancontributesignicantlytotheviscouscomponentofattenuation.Therefore,incalculations,bothcontributionsofocstothescatteringcomponentandsingleparticlestotheviscouscomponentweretakenintoaccount.Thesedimentabsorptionscanbecalculatedinprinciple,iftheSSCproleisknown.However,theSSCproleisnotknown,andestimatingitistherealgoal. 29

PAGE 30

Becauseinobservations(Section 3.2.2 )near-bedsedimentconcentrationsexceeded1kg/m3,inthisinvestigationsistakenintoaccount.DeterminingtheproleSSC(r)thenbecomesanimplicitproblem,andaniterativeapproach(e.g., ThorneandHanes 2002 )hastobeusedtosolveit,e.g.,SSCandsarecalculatedsequentiallyateachcell.Basedontheapplicationsinmuddyenvironments,thereisnodirectevidencewhethertheacousticbackscatterrespondstoocsassinglelargerparticles,ortheindividualprimaryparticlesinsidetheocs. FugateandFriedrichs [ 2002 ]suggestedthattheacousticresponseforresuspendedaggregatesdependsmostlyonthesizeandshapeoftheconstituentgrainsratherthanthesizeorshapeoftheoc. Gartner [ 2004 ]estimatedsuccesfullytheSSCprolesusingthesizeofaggregatedsuspendedsolidsasscatterers.Recently, Haet.al. [ 2011 ]showedthatPC-ADPdidnotperformwellforthesuspensionswithgrainsizeintheorderof1mandsuggestedthatthePC-ADPlikelyrespondstoocsasawholeboundinbottomboundarylayerwheretheocserodedfromthebedaremoredenseandrobustthantheocsinthewatercolumn.Inthisstudy,acousticinversionweredoneforbothtreatingtheocsasawholebound,andtheindividualparticlesinsidetheocasscatterers.Betterresultswereobtainedwhenassumingaggregatingsuspendedsolidsaslargersingleparticles,andthisassumptionwerefollowedthroughoutthispaper.Notethatthisassumptionstillneedstobeconrmedwithlabororatoryexperiments,hovewerithasnotbeendoneyetduetothediffucultyofcontrollingthesizeoftheocswhichrequirescontrollingseveralparametersatthesametimesuchasfallvelocity,concentrationandturbulencelevel.Thisproblemisasubjectofaveryextensiveresearchwhichisbeyondthescopeofthestudy.Here,theassumptionmadeisjustiedbythegoodresults(Chapter 3.2.3 ). 30

PAGE 31

3 ),theslantrangeisdeterminedforeachmeasurementbinasafunctionoftransducerangle,binsizeandblankingdistance.Theneareldcorrectioniscalculatedfollowing Downingetal. [ 1995 ];thewaterabsorptionfactorwistabulated(e.g., FrancoisandGarrison 1982a,b ).Ifsedimentsizeinformationisavailable,thedensityofmudocsisestimatedasfollowing Kranenburg [ 1994 ]: where,wandsarethedensitiesofmudocs,waterandprimarysedimentparticles,andDfandDpareocandprimaryparticlediameters,respectively.Theexponentisafunctionofthefractaldimensionnfoftheoc.Equations( 3 ),( 3 )and( 3 )canthenbeusedtocalculatetheparametersks,s(setting,e.g., Standardmethodsfordeterminingtheinstrumentconstantkt[ ThorneandHanes 2002 Betteridgeetal. 2008 ]areeitherperformingafullelectronicandacousticcalibrationofthesystem,whichrequirespecialequipment,orconductingextensivemeasurementsinahomogeneoussuspensionwithknownsedimentconcentrationsandscatteringcharacteristics,again,requiringaspeciallaboratorysetup.Intheabsenceofthemeanstoperformthesetests,weproposehereanoptimizationapproachthatseekstoidentifythevalueofktthatbestreproducesaselectedsetofobservations.TheoptimalvalueofktresultsinSSCvalues(equation 3 )thatminimizestheRMSerror 31

PAGE 32

3 )usingtheoptimalvalueofkt. Theapproachhasshownlimitationsandambiguitiesinpreviousapplications.Single-frequencyinstrumentscannotdistinguishparticle-sizefromSSCvariability[ Gartner 2004 ].Theacousticbackscatterhasmaximafor(thusisbiasedtoward)certaingeometriesandsuspendedparticles.Air-bubble backscatterdominatesthatofsedimentparticlesofsamesize [ Hamiltonetal. 1998 Libickietal. 1989 Traykovskietal. 2007 ];mostefcientbackscatterisachievedforparticleswithka=O(1).Theshapeofthescatteringparticlesisnotimportantifka1(Rayleighscatteringregime; ThorneandMeral 2008 ),butbecomessoka!1approachesunity.Applicationinmuddominatedenvironmentsaddsmorecomplexities.Forexample,verticalvariationofthesedimentsize(mudocs)ismuchmoresignicantthansandyenvironments,sinceaggregationandocbreak-upcantakeplaceimmediatelydependingonowconditions(e.g.,turbulencelevel).Inuseofsinglefrequencyinstruments,thissignicantchangeinocsizecanbeinterpretedasachangeinsedimentconcentration.Theseambiguitiesconstrainourabilitytofullyrepresentocdynamics,andourinterpretationoftheresults.Inthefollowing,weassumethattheocsizeisinvariantwiththedepth,andwillinterpretthisvalueasanequivalent,insomesenseverticallyaveraged,ocsize. 2.2 ).Observed30-minmeanwaveheights,currentspeedanddirection,backscatterintensityproles,andSSCobservationsareshowninFigure 3-2 .UntilMarch8th,17:00,theswellenergywasmuchlowerthantheenergyofseas(Figure 3-2 a).Thelocationofthemaximumbackscatterintensitywassteadyduringthisperiod(intenseblacklineinFigure 3-2 d),suggestingnegligiblebedreworkingbywavesandcurrents.However,asmallamountof 32

PAGE 33

3-2 aande),andcurrentspeedanddirection. (Figures 3-2 bandc).AnincreaseintheswellenergystartingonMarch8thappearseventuallytohavetriggeredsomebedevolution(slightchangeinmaximumbackscatterintensitylocation,Figure 3-2 d),togetherwithincreasingSSCvalues.ThesizedistributionestimatedfromtheLISSTmeasurementswasbimodalwithadominantoc-modearound230mandaweakeroc-modepositionedaround50m[ Safaketal. 2010 ].Themeanocsizeextractedfromtheocsizedistributionvariedbetween 100mand300mwiththeaveragearound200m(Figure 3-2 f). Amajorsourceofuncertaintyinthisstudyisduetothelimitedscopeofapartoftheavailabledataset,i.e.,measurementofsedimentsizedistributionatasinglepointabovethebed.QuestionablefunctionalityoftheLISSTatrelativelyhighnear-bedconcentrationsisthereasonofnothavinghighspatialresolutionofsizedistributionnearthebed.Itisverylikelythatocsizevariesthroughoutthewatercolumnduetovaryingturbulenceintensityandsedimentavailability,bothofwhicharerelativelyhighnearthebed[ DyerandManning 1999 Winterwerp 2002 Hilletal. 2001 ,andmanyothers].Thiswouldrequiretheinvestigationontherelationshipbetweenocculation,sedimentconcentrationandturbulentow.Anintermediatesteptowardsevaluatingthisthree-wayinteractioniscurrentlybeingtakenbasedonelddatacollectedovertheAtchafalayaShelfduringsimilarowconditions[ Safaketal. inpreparation ].Theapplicationoftheabovealgorithmtoeldobservationsrequiresinformationaboutsedimentstructure.SomeparameterscanbederivedfromLISSTobservations:forexample,LISSTestimatessuggestanaverageofmeanocsizeofapproximately200m(Figure 3-5 a, Safaketal. 2010 ).Others,suchastheocfractal-dimensionneededtoevaluateequation 3 ,cannotbemeasureddirectlyandwereinferred(nf=2.3forDf=200m, KhelifaandHill 2006 Safaketal. 2010 ).Theone-hour 33

PAGE 34

TimeevolutionofA)signicantwaveheightatthesurfaceinthesea(f>0.2Hz,thinline)andswell(f0.2Hz,thickline)bands;thePC-ADPmeasurementsofB)currentspeedproles,C)verticallyaveragedcurrentdirections;D)normalizedacousticbackscatterintensity(dashedlinescorrespondtotheelevationsatwhichtheOBS-3sweresampling);(e)suspendedsedimentconcentrationmeasuredbytheOBS-3s(thickline:50cmab,thinline:75cmab);F)meanocsizemeasuredbytheLISST. 34

PAGE 35

3.3 ).However,moreimportantly,itisimposedinthecontextofagenerallackofeldobservationsaboutthespecicsofocdynamics(e.g.,unknownocfractaldimension;orthesingle-pointsizedatacollectedat120cmab,outsidethePC-ADPprolingrange).Whiletheassumptionofvertically-constantocsizeisprobablyunrealistic(seethediscussionabove),itcanstillbeinterpretedasaneffective-sizevalue.Thetime-variabilityoftheocsizeistakenintoaccount(Figure 3-2 f).Thesystemconstant,kt,doesnotdependonrange,particlesizeandconcentration[ Betteridgeetal. 2008 ].SSCmeasurementsattwoverticallevelsweredividedintogroupsof0.1kg/m3(Figure 3-3 a).Foreachgroup,theoptimumktgivingtheminimumerrorbetweenmeasurementsandcalculationswasdetermined(equation 3 ,Figure 3-3 b).ThevalueofktfordifferentSSCclassesdoesnotshowasignicanttrendandliesaroundtheaveragevalueof44dBm3=2withtheexceptionofthevaluecorrespondingtoSSC=0.2kg/m3.Thatvaluewasconsideredasanoutlierduetomeasurementerrorandwasnottakenintoaccountincalculationofkt.Dependencyofktondifferentocsizeclasses(25mintervals)isseeninFigure 3-4 .Thereisagainnosystematicvariation.Thevalues,consistentwiththeSSCdependency,liebetweenthe10%ofthesameaveragevalue.Thesystemconstantktdidnotshowadependencyonrange,either(see'x'sandcirclesinFigures 3-3 band 3-4 b).Therefore,thisconsistentmeanvalueofktwasusedinSSCcalculations. 35

PAGE 36

A)Thehistogramofsuspendedsedimentconcentrationforatotalof281mean-valueobservationsat50cmab(black)and75cmab(gray)B)Variationofktwithsuspendedsedimentconcentration.Circlesand'x'sdenotethevaluesat75cmaband50cmab,respectively.Thesolidlineistheaveragedktvalueovertheallconcentrationclassesandthedashedlinesare10%differencefromtheaveragevalue. 3-5 bandc,correlationcoefcientr=0.87and=0.14kg/m3).ThecirclesinFigure 3-5 adenotetheoptimumocsizesprovidingthebestSSCcalculations(circlesinFigure 3-5 bandc).Itisencouragingthatoptimumocsizevaluesareinthesameorderofmagnitudeandshowasimilartrendwiththemeanocsizesmeasured.ThisresultsuggeststhattheeffectiveocsizeDfandSSCcanbeestimatedwhenin-situparticlesizemeasurementsare 36

PAGE 37

A)Thehistogramofocsizeobservations.B)Variationofktwithocsize.Circlesand'x'sdenotethevaluesat75cmaband50cmab,respectively.ThesolidlineisthesameaveragevaluewithFigure 3-3 bandthedashedlinesare10%differencefromtheaveragevalue. notavailable.Figure 3-5 dshowsthetimeevolutionoftheverticalprolesofsuspendedsedimentconcentration.Althoughthebackscatterstructureitselfisoftenconsidered(andinterpretedas)ameaningfulrepresentationoftheSSCprole(e.g., Jaramilloetal. 2009 ),comparingFigures 3-2 dandFigure 3-5 dshowsthatthebackscatterintensityisagoodindicatorofthetheoverallamountofsedimentinsuspension,butnotoftheverticaldistributionofsediment. 3.3.1GoverningEquationsTheobservationsweremodeledusinga1DVwave-phase-resolving(verticaldomainatasinglepointonthehorizontalplane)bottomboundarylayernumericalmodel 37

PAGE 38

TimeevolutionofA)meanocsize(continuousline)andtheoptimumocsizeyieldingthebestconcentrationestimates(circles).SuspendedsedimentconcentrationatthelocationoftheOBSsB)75cmandC)50cmab(OBScontinuousline;PC-ADPestimatesusingmeanocsize'x's;PC-ADPestimatesusingeffectiveocsizecircles);D)PC-ADPderivedsuspendedsedimentconcentrationproles. developedby Hsuetal. [ 2009 ].Themodel,atime-dependentRANS(Reynolds-averagedNavier-Stokes)formulationbasedonatwo-equationk"closure,hasbeenappliedsuccessfullyinrecentyearstocohesivesediments(seee.g., Hsuetal. 2007 2009 Safaketal. 2010 SonandHsu 2011 ).Themomentumbalanceisbetweenfree-streamhorizontalpressuregradient(prescribedasowforcingduetowavesandcurrents)andmomentumtransportbyuidshearstresses(bothviscousandturbulent).Theowmomentumequationsforcross-shelfuandalongshelfvvelocities 38

PAGE 39

Hsuetal. [ 2007 ]): Hsuetal. [ 2007 ]): @t=@ @z(1)Tp11gzt @z+Tp 39

PAGE 40

@z+t @z, @t=C1 @z+t @zC22 @z, whereTListheturbulenteddytimescale.Thenumericalcoefcientsusedinthek"closureare:C=0.09,C1=1.44,C1=1.92,C3=1.20,k=1.00,=1.30,C=1.00.Therstthreetermsintheequationsrepresentshear-inducedturbulenceproduction,diffusion,andturbulentdissipation.Thelasttwotermsrepresenttheeffectsofsediment(mostlydamping)oncarrieruidturbulenceduetosediment-densitystraticationandinteractionbetweenuidandsedimentvelocityuctuationsthroughviscousdrag.Thesedimentphaseisdenedinthemodelbyocdensity(equation 3 ),andthusbytheprimaryparticlesize,ocsize,andocfractaldimension(assumedconstant).SettlingvelocityismodeledusingtheStokeslawwithhinderedsettlingeffectincorporated: Hsuetal. [ 2007 ]: 40

PAGE 41

Hsuetal. [ 2007 ]: 4k3 2 3-2 eand 3-5 d).Fornumericalsimulationpurposes,thebottomlocationwasdenedasthatofthemaximumacousticbackscatterintensity.Initspresentimplementation,modelassumeszero-shearatthetopboundary;tosatisfythisrequirementandalsotoallowthemodeltomatchSSCestimatesatthetopmostPC-ADPbin(sometimesexceeding0.6kg/m3)thecomputationaldomainwasextendedfromthebedtoapproximately30cmabovetherstPC-ADPbin(overallspanof1.2m).The2-cmverticalresolutionusedisequaltothePC-ADPbinheight.Testrunswitha1-cmverticalresolutionshowednomodelsensitivitytogridsize,suggestingthatthecurrentboundary-layerowdominatestheprocesses,wellresolvedata2-cmgridsize. Sedimentavailabilityfromthebed ( parametero wasadjustedtomatchestimatedSSCvalues.Becausepreviousapplicationsdidnotshowmodelsensitivitytocriticalshearstressvalues[ Safaketal. 2010 Hsuetal. 2009 ],thisparameterwassettoc=0.4Pa.Thisisalsowithintherangeof0.05-1.1Pasuggestedby Hsuetal. [ 2007 ]and Hsuetal. [ 2009 ],andusedby Safaketal. [ 2010 ]forsimulationsofsedimenttransportatthesamegeographiclocation.ThemedianocsizemeasuredbytheLISSTwereusedinthesimulations. Numericalrunsusedafastrelaxation-timemethodrecentlyimplementedinthemodelthatgeneratesacurrentprolewithauser-deneddepth-averagedvelocity(e.g., 41

PAGE 42

TheprolesofSSCandcurrentvelocityarecalculatedbasedontheoptimal(ks,0)pair,i.e.,thevaluesthatyieldthebestagreementbetweencalculatedandthePC-ADPderivedverticalprolesofSSC.Forthemodeled13-hrinterval,overall,thenormalizedRMSerrorinthemodelcalculationsofsuspendedsedimentconcentrationvariesbetween6-20%withanaverageof11%;theerrorinthecurrentspeedcalculationsisbetween3-16%withanaverageof8%(Figure 3-6 and 3-7 )TheresultssupporttheassumptionthatSSCprolesderivedfromacousticbackscatterintensitycanhelpvalidatenumericalmodelsandimplicitly,improvemodelestimatesofquantitiesthataredifculttoobservedirectly,suchasthenear-bedturbulentstresseld. 42

PAGE 43

Timeevolutionof:A)suspendedsedimentconcentrationestimates(PC-ADPbackscatter);B)suspendedsedimentconcentrationcalculationsofthemodel. Figure 3-8 ashowsanexampleoftwoprolescorrespondingtotwodifferentmeanocsizes(250m,and160m)butwiththesameSSC=4kg/m3atapproximately12cmab.Theverticalprolesareinagreementwiththeconclusionsofpreviousstudiesthatsmallerocs,forsimilarowconditions,resultinhigherSSCintheupperwatercolumnwithmoremixedprole[ Safaketal. 2010 DyerandManning 1999 ].Accordingtoequation 3 ,settingvelocityscaleswithD3nff.Hence,withatypicalfractaldimensionofaround2,settlingvelocityincreasesalmostlinearlywithocsize.Largerocssettlefaster,thusleadingtolowerconcentrationsintheuppercolumn.ThedependencyshowninFigure 3-8 awasconsistentlyobservedthroughtheentiremeasurementpointsanalyzed. 43

PAGE 44

Timeevolutionof:A)currentspeedmeasurements(PC-ADP);B)speedcalculationsofthemodel. AcloserexaminationofthestructureofthecurrentboundarylayersuggeststhatthetransitiontoalargerSSCgradientwiththedepthisassociatedwithacriticalchangeinsediment-induceddensitystratication.WhilenumericalsimulationresultsforthetwocasesshowninFigure 3-8 exhibitverticalprolesofvelocitythatarequalitativelysimilar(Figures 3-8 aandb),thecomputedgradientRichardsonnumberdifferssharply(Figures 3-8 c).Forthecaseofsmalleroc(trianglesandthinlineswithocsizeof160m),theRichardsonnumbernearthebedisaround0.2andincreasesupward.Ontheotherhand,forthecaseoflargeroc(circlesandthicklineswithocsizeof250m),thenearbedRichardsonnumbervalueismuchlarger(around0.68).However,itrstdecreasesto0.5ataroundz=5cmabovethebedandthenincreaseswithdistancefromthebedbutatalowerratecomparedtothatofsmallerocsize. 44

PAGE 45

@z z0,forturbulentboundarylayer(uisthefrictionvelocity,isVonKarmanconstantandz0=ks=30)andtheRouseproleforsedimentconcentration z,(3)(risareferenceconcentration,histhedepthofow,=ws 3 ,yields z1.(3)ThisresulthighlightsthecriticaleffectofontheRiprole.For<1,thisexpressionpredictsRi(z)increasingwiththeheightabovethebed.Basedon_ws 45

PAGE 46

A)TwoobservedSSCverticalproles(estimatedfromacousticbackscatter)(circles:250m,March,8that21:00hoursUTC;andtriangles:160m,March,4that18:30hoursUTC),andcorrespondingmodelsimulations(thickline:250mandthinline:160m).B)Correspondingobservedandsimulatedcurrentspeeds.Verticalprolesof:C)theRichardsonnumber;D)thenormalizedturbulenceintensityobtainedfromnumericalsimulations. IntheexamplesshowninFigure 3-8 ,owconditionsaresimilar,butthedifferenceinthesizeoftheocsisapparentlyenoughtocausethesignicantdifferenceinRiproles.Rousenumberscorrespondingtothesmall-ocandlarge-occasesare=0.62(u=0.94cm/s,ws=0.24cm/s)and=0.86(u=1.27cm/s,ws=0.45cm/s),respectively.Here,ourestimateofRousenumberdoesnotexplicitlyaccountfortheeffectofsedimentondampingtheturbulence,suchasthroughareducedKarmanconstant.However,qualitativelythecasewithlargerochasaRousenumberthatissignicantlylargerthanthatofsmalleroc.Hence,thelarge-occaseshowsanear-bedRivaluearound0.6,indicatingthatthedampingofturbulenceduetosedimentinducedstraticationisimportant.Theturbulentintensity(p 3-8 d.Basedontheseproles,itisclearthatalthoughthecaseoflargerocismoreenergetic,relativelyhighturbulenceonlyexistswithinrst5cmabovethebed,whichisapproximatelyassociatedwiththewaveboundarylayerthickness.Abovethewave 46

PAGE 47

3 ,lowerRichardsonnumbersareexpectedforhigherturbulencelevels.Figures 3-8 canddshowthatlargeroccasehasbothhigherRiandhigherturbulenceintensitynearthebed.However,theverticalconcentrationterm@ @zisalsolargerforthelargeroccasewhichreversesthebehaviorandresultsinhigherRinearthebed(around0.6).NotethatRiandturbulentintensityestimatesshownherearebasedonnumericalsimulationsthatassumetheocsizeindependentofheightabovethebed.Therefore,verticalvariationsoftheocsizethatalsoaffecttheshapeoftheconcentrationprolearenotaccountedfor. 47

PAGE 48

3.2 ,provideimportantcluesonthesea-oorbehaviortosignicantwaveactionatdifferentstagesofthestorm.Sedimentconcentrationestimatesalsoallowforreliablyconstraininga1DVnumericalmodelpresentedinSection 3.3 which,inturn,providesnearbottomshearstressesinducedbywavesandcurrents.Differentbehaviorsofsuspendedsedimentandbedpositiontothebottomshearstresshelpdistinguishthecontributionsoflocalresuspensionandsedimentadvectiontothesedimenttransport.Inlightofdifferentpiecesofobservationsandanalysis,thethresholdvalueofshearstresstoinitiatebederosion,andtheevolutionofbeddensitythroughoutthestormarealsodetermined.Thesendingsareconrmedbythesedimentcharacteristicparametersobtainedindependentlyfrompreviouslaboratorytests.Thevariabilityoftheprocess(e.g.,bedstates,beddensity,criticalshearstressforerosion)foralargerpopulationofstormsisalsoinvestigated. 4-1 a,c).Theeventwasprecededbyaweekofrelativelycalmconditions(4-swavesrarelyexceeding0.5m).Thefrontproducedwaveheightsreaching1.3m,with1-mheightswellsthatpropagatedconsistentlynorthwardthroughMarch4th,despitetheshiftinwinddirection.Aswindsweakenedtoabout5m/s,rapiddecayinwaveactivityoverallfrequency 48

PAGE 49

EvolutionofwavefrequencyspectrumandpropagationdirectionduringthestormofMarch3rdto5th,2008.A)Windspeedanddirection(color-coded),andsignicantwaveheightintheshort-wave(f>0.2Hz,blue)andswell(f0.2Hz,red)bands;B) Normalizedspectraldensity;C)Peakdirectionofeachspectralband.Inthedirectionconventionused,Nmeansowing(orpropagating)northward. bandsreducedthesignicantheightfromapproximately0.9-mheightto0.1-minsixhours.Thisphenomenonwasobservedbeforeandshowntoberelatedwithwavedissipationinducedbythemuddybed[ SheremetandStone 2003 Sheremetetal. 2005 Jaramilloetal. 2009 Sheremetetal. 2011 ].ThestrongestcurrentscoincidedwiththewindshiftingdirectiononMarch4th(Figure 4-2 ).TwocurrentpulsescanbeidentiedinthePC-ADPobservations(Figure 4-2 c,d).TherstpulseowingtowardWSWwasobservedonMarch4thfrom4:00to10:00hours,associatedwitha1-mdropinthemeanwaterlevel(Figure 4-2 b). 49

PAGE 50

Jaramilloetal. 2009 ],thispulseislikelyrelatedtotheushingofthestormsurgeproducedbytheatmosphericcoldfrontandthechangeinthewinddirectionafterfrontalpassage.Asecond,strongerpulseowingtowardSSWfollowed(March4th12:00toMarch5th00:00hours),associatedwithanincreaseinthesurfaceelevationandareturntothenormaltidalcycle.Thispulsetransportedfresh,sediment-ladenwaterlikelyassociatedwiththeAtchafalayaRiverplume(salinitylessthan5psu,watertemperature13oC,andSSCupto10kg/m3)thatdisplacedseawater(30psusalinity,temperature18oC,SSCapproximately4kg/m3;Figure 4-2 eandf).Thewaterfrontresultingfromthecollidingmassespassedovertheexperimentalsitewithinapproximatelythreehours,illustratedbytherapiddropinsalinityandtemperature(Figure 4-2 f);SSCvaluesshowweakervariationwithperhapsanincreasebyafactorof2.StartingfromMarch4th18:00hourswaveactivitydecayedandsedimentsettlingandadvectionclearedthewatercolumnrapidly,withSSCvaluesdecreasingfrom10kg/m3toalmostnilinsixhours(Figure 4-2 e).Thefactthatsalinityandtemperaturedropsignicantlyduringthestrongcurrentactivityperiodsisstartling,andreliabilityofthevaluesneedstobechecked(especiallyforalmostzerosalinityvalues).Thesalinityandtemperaturemeasurementsatotherstationsalsoshoweddecreasingvaluesduringthisperiod(Figure 4-3 ).ThevaluesarenotaslowastheonesatPlatform3agreeingwithexpectations,sincethatplatformistheshallowestandtheclosestonetoAtchafalayaRivermouth.Also,thedailystagedatacollectedclosetotherivermouth(AtchafalayaBayatEugeneIslandstation)bytheUSArmyCorpsandEngineersindicatesasignicantincreaseinriverdischargeonMarch4th(notshownhere).Thisconrmsafreshwaterincometothesystem(likelysediment-laden)duringstrongcurrentperiods.ThecurrentvelocityprolecanbeusedinconjunctionwiththePC-ADPacousticbackscatterobservations(Figure 4-4 )toassessthedynamicbehavioroftheseabed. 50

PAGE 51

StormofMarch3rdto5th,2008:A)Windspeedanddirectionandsignicantwaveheights(short-wave:blueandswell:red).VerticalstructureofmeancurrentrecordedbytheB)ADCPandC)PC-ADP;D)DirectionofPC-ADPmeancurrents;e) SSCobservedbytheOBS-3locatedat18cmab;F)Salinity(blue)andtemperature(red)at55cmab.LocationsoftheinstrumentsareshowninFigure 2-2 ). 51

PAGE 52

ThevariationofA)salinityandB)temperaturewithdifferentwaterdepths. Becausestrongreectionsaretypicallyassociatedwithsharpdensitygradients,thepositionofthebedisestimatedhereasthesmoothedpositionofthenear-bedlocalbackscattermaximum.Thisdenitionincludessharplutoclinesofhigh-densityuid-mudlayers(e.g.,March5th,Figure 4-4 c),butexcludesweakmaxima(e.g.,March4th21:00),whenthebedcannotbereliablyidentied.Theelevationsofzero-meanvelocity(ZMV)andzero-RMSvelocity(ZRV)canbeusedtoinvestigatethedepthofpenetrationofsteadycurrent(hydrodynamicdepth)andoscillatorymotions(wavepenetrationdepth),respectively(e.g., Jaramilloetal. 2009 ).Theresults,showninFigure 4-4 c,areinformative.AtnoononMarch3rd,thebedpositioncoincideswithZMVandZRV,suggestingasolidbed.Aswaveactivityincreases,theverticallystretchablebedlevelrises(dilation,March3th18:00toMarch4th01:00hours).Thisimpliesthatpriortodilationthesolidbedwouldbemoreappropriatelydescribedasastationarysoftbed,anephemeral 52

PAGE 53

EvolutionofbedpositionindicatorsandstateduringthestormofMarch3rdto5th,2008.A)Windspeedanddirection(color-codedthickline),andsignicantwaveheights(short-wave:blueandswell:red); B)Salinity(blue)andtemperature(red)at55cmab(sameasinFigure 4-2 f); C)normalizedPC-ADPacousticbackscatterintensity.Thelinesrepresentlocationsof:maximumbackscatterintensity(triangles);zeromeanhorizontalvelocity(stars),andzeroRMShorizontalvelocity(circles).Thesmoothedestimateofthebedpositionismarkedbythecontinuousthickline. statewithouthorizontalmovement.Followingdilation,thebedlevelfallscontinuouslyforabout10-hours(erosion,March4th02:00to12:00hours).BedaccretionbeginsonMarch4that12:00hoursandcontinuesuntilearlymorningofMarch5th.Beddilationoccursintandemwiththedownwarduxofwaterintothebed(alsocalleduidization)andareductioninbeddensityasthesedimentmasswithinthedilatingbeddoesnotchangeeitherduetoerosionfromthebottomordepositionfromthetop.Sincethedensityisbelowthespace-llingvalueatwhichthebeddevelopsacontinuousparticlematrix,thebedsurfaceisalutoclinemarkingtheuppersurface 53

PAGE 54

4-2 )thebedissoftandstationary;itdilateswithincreasingwaveenergyanderodescontinuouslythereafterthroughnoon,March4th.Thedeposition(accretion)phasecoincideswithadecreaseinwaveactivity.IntheafternoonofMarch4th,withthearrivalofasecondcurrentpulsecarryingfresh,sediment-ladenwater(Figures 4-2 and 4-4 ),SSCincreasesto10kg/m3andremainsapproximatelyconstantuntilearlymorningofMarch5th.Attheendofthestorm,thepersistenceofoscillationsinthe5-cmthicksurciallayersuggeststheformationofahindered-settlinguidmudlayer. 54

PAGE 55

4.3.1ConversionofAcousticBackscattertoSSCproleSuspendedsedimentconcentrationproleswereestimatedusingthemethodexplainedinSection 3.2 .Sincetherewasnoobservationonthesedimentocsizedistributionforthedatasetdiscussedhere,aslightlydifferentcalibrationprocedurewasusedintwosteps:a)aprocedureforestimatingthePC-ADPsystemconstantkt,andb)anoptimizationsearchforadepth-independent,effectiveocsize,correspondingtothebesttbetweenbackscatterandopticalSSCestimates.ThePC-ADPbackscatterwascalibratedfortheinstrumentconstantktusingindependentOBSmeasurementsmadebetweenFebruary22ndandMarch3rdattwoheightsabovethebed(theOBS-5stoppedfunctioningafterMarch3rd).BecauseSSCvalueswerelow(orderof1-2kg/m3;notshown)duringthatperiod,andnoocsizeobservationscouldbemade,theprocedureusedaconstantocdiameterDf=200m.Thisisacceptablebecauseinpreviousapplications[e.g., Sahinetal. 2012a ],thesystemconstantktdidnotshowsensitivitytoocsize.Thisisalsoconsistentwithpreviousstudiesatthesite[ Safaketal. 2010 ],thatindicatedavariabilityrangeforDfbetween100mand350mwithanaveragevalueofaround200m.FortheMarch3rdtoMarch5thperiodthesearchrangefortheeffectiveocsizewasbetween50mand350m.TheSSCproleandtheocsizeforeachburstoverthedurationoftheexperimentincludingthemajoreventofinterestbetweenMarch3rd-5thweredetermined.Thedeviation(RMSerror)betweenthecalculated 55

PAGE 56

ComparisonofSSCvaluesestimatedusingthePCADPacousticbackscatterandopticalobservations:A)OBS3,at18cmab,andB)OBS-5at10cmab. concentrationsatthelocationsoftheOBSsandthemeasurementswas0.37kg/m3withcorrelationcoefcientr2=0.92(Figure 4-5 ). 3.3 developedby Hsuetal. [ 2009 ].Thestandardsimulationprocedure[e.g., Safaketal. 2010 ]startsthemodelfromaninitialstateofrestwithzeroSSCandseeksasteadystatematchingwiththeobservedmeanowstructureandSSCprole.Sedimentismadeavailabletothesimulationdomainatthelowerboundary;theresuspensioncoefcient[e.g.,equation20in Hsuetal. 2009 ]wasusedasafreeparametertocontrolsedimenterosion.Thesimulationdomainwasdenedtospanthebottommeterabovebed,withthebedposition(Figure 4-4 c)asthebottomboundary,theupperboundaryatapproximately 56

PAGE 57

Hsuetal. 2007 2009 Safaketal. 2010 ].Themodelhasbeenusedasaninvestigativetool:ifitcanbetunedtoreproducetheobservations,itsinnerbalancecouldbeusedtodrawinferencesconcerningthenon-observablephysics.However,initspresentimplementationthemodelcannotbeexpectedtofullyreproduceobservationsmadeundernon-stationaryconditions.TwoexamplesofsuchconditionsarethetimesegmentwithbeddilationfromMarch3rd,18:00to24:00hours,andarrivalofthesediment-ladenwaterfrontonMarch4thfrom12:00to18:00hours.Theformernon-stationaryconditionissimplyanuninterestingevent,withaweakowvelocity(orderof1cm/s,Figure 4-2 b-d).ThemodelisunabletoproducetheturbulencerequiredtosustaintheobservedlowSSCvalues(Figures 4-4 c, 4-6 b),andthestationarysolutionpracticallycapturestotheentiresedimentmasssettledonthebed.Thelattereventismorechallengingbecauseitisclearlyunsupportiveofthemodelassumptionofverticalsedimentbalancewithabottomboundarysedimentsource.Thefrontalpassageisequivalenttoasuddenemergenceofasedimentsourcehighinthewatercolumn.Duringthisperiod,evenifthenumericalsimulationsarenotobviouslywrong,unlesstheerrorscanbeestimated,thesimulationsshouldbetreatedwithcaution,asthemodellikelyover-estimatestheturbulentstresses.However,theremarkablespatialuniformityattheexperimentalsite(e.g.,bottomslopeslessthan0.001overtensofkm)suggeststhatsedimentuxconvergencemaybenegligibleingeneral,withthepossibleexceptionofwaterfrontsproducedbythe 57

PAGE 58

4-2 )supportstheinferencethatnon-zerosediment-uxconvergencerarelyoccursatthesite,andhasashort(orderofhours)lifespan.Basedontheseconsiderations,themodelwasappliedovertheentirestormduration.Overall,thenumericalreconstructionoftheSSCverticalstructureagreeswellwiththeestimatedvalues(Figure 4-6 b-c).NormalizedRMSerrorrangewasbetween4and60%(18%average).ThemodelalsocapturedtheverticalowvelocitystructureswellwithnormalizedRMSerrorrangeof6to56%(23%average).Theerrorswerelessthan10%formostofthesimulatedperiod,thesimulationswithlargeerrorsmostlycorrespondtotheperiodofbeddilation.Thesimulationscanbeexpectedtobevalidfortheentireduration,withthepossibleexclusionoftheweaklynon-stationarysegmentonMarch3rd18:00hours,andwater-frontpassagearoundMarch4th12:00hours. 4.4.1CriticalShearStressforErosionSurfaceerosionoccurswhenlargelayersofsedimentareerodedandmobilizediftheshearstressforerosionisexceededbystressesinducedbywavesandcurrents.However,sedimentresuspensioncanbeobservedwhenstressesarelowerthenthecriticalshearstress.Thisisknownasentrainmentanditoccursincaseofuidmudwhennon-turbulentmudlayerisentrainedbytheupperturbulentwaterlayer[ WinterwerpandvanKesteren 2004 ].Theevolutionoftheobservedbedposition(e.g.,Figure 4-4 cand 4-6 b-c)andthebottomstresscalculatedbythemodelinthevicinityofthebed(Figure 4-6 d)areconsistent,despitebeingdynamicallyunconnected(inthemodel,theboundaryisprescribedanddoesnotevolveduringsimulation).DuringthelowcurrentsandwaveactivityperiodbeforeMarch4th,shearstressvaluesarelessthan0.3Pawithanaverageof0.22Pa.StressvaluesbegintoincreasearoundmidnightMarch4thwith 58

PAGE 59

ModelsimulationsforthestormofMarch3rd-5th,2008.A)Windspeedanddirection,andsignicantwaveheights(short-wave:blueandswell:red);B)SSCverticalstructureestimatedfromPC-ADPbackscatter;C)Numerically-simulatedSSCverticalstructure;D)TurbulentReynoldsstressatbedlevel(denedasthecontinuouslineinFigure 4-4 b). 59

PAGE 60

Migniot [ 1968 ]fortheupperBinghamyieldstressy(stressdeterminedbyextrapolatingthelinearportionofastressstrainow-curvetothestressaxis)andthecriticalshearstressforerosions: Thisrelationshipisappliedhereasitisbasedonviscometricdatafromawiderangeofmarinemudsandsomenepowders.Thebeginningofthebederosionperiodsuggestsacriticalshearstressforerosion0.3Pas0.5Pa,whichcorrespondstoayieldstressof1.1Pay1.95Pa(equation 4 ).Thesevaluespermitaback-estimationofthebeddensityandocsizeattheonsetoferosion.Thecorrespondingvolumefractionforsolidsof0.075vs0.085(equation C ,seealsoFigure C-1 ),yieldsabeddensityof1,145kg/m3bed1,160kg/m3,withaocsizerangeofbetween175mand205m(equation4in Safaketal. 2010 ).Thesevaluesagreewiththelaboratorytests( Robillard 2009 ;AppendixB),aswellaswiththe2006LISSTparticlesizeobservationsgivenin Safaketal. [ 2010 ]. Letterand 60

PAGE 61

, 2011 ]suggestthatunderaconstantturbulentstressappliedtothebed,thesuspended-sedimentmassapproachesanequilibriumvalueatwhichthedepositionanderosionuxespracticallybalanceoutprovidedtheentirecohesivesedimentsizerangepresentisincludedintheanalysis.Dampeningofturbulencebythesuspendedsediment,andthefactthattheyieldstressaswellasthecriticalshearstressforerosionincreasewithdepthinthestratiedbedassedimentisremoved,arefactorsthatmayacceleratetherealizationofanear-equilibriumstateintheeld.Thenetsedimentmassuxcalculatedasd dtRDSSC(z)dz,whereSSC(z)istheestimatedverticalSSCprole(Figure 4-6 b)andDisthemodeldomain(approximatelytherstmab),isplottedinFigure 4-7 aagainstthesimulatedbottomstress(Figure 4-6 d).Here,concentrationsabovethePC-ADPrangewerenottakenintoaccountastheSSCintheuppercolumnisnegligiblecomparedtothatintherstmab.Inthishysteresis,thereissomemassuxbetweenMarch3rdat18:00hoursandMarch4th03:00hourswhenthebottomstressislessthan0.3Pawhichislikelyrelatedtotheentrainmentofdilatedsoftbed.ThenetuxincreasessignicantlywithincreasingbottomstressonlyafterMarch4th04:00hours,whenthebedstress'0.5Paisintherangeofthecriticalshearstressforerosion(seeSection 4.4.1 ).ThenetuxreachesamaximumonMarch4that12:00hourswhentheseabedreachesitslowestposition(Figure 4-4 ).Thenetuxcancels,i.e.,depositionbalanceserosion,asthebedstressreachesitsmaximumvalue;thenthecyclereverses,withthewatercolumnlosingsuspendedsedimentmassbydepositionandthestressdecreasingduringthewaningphaseofthestorm.Figure 4-7 bshowstherelationbetweentheevolutionofthesimulatedbedstressandthenetbed-volumeuxdh dt,wherez=h(t)isthebedposition(Figure 4-4 c).Thesigninsuresthatpositiveuxesraisethebedposition(addsedimenttothebed).Thedependencyisconsistentwithexpectations,witherosionoccurringasthestressincreasesandaccretionasthestressdecreases. 61

PAGE 62

Observedmass(water-column)/volume(bed)uxvs.numericallysimulatedbedstress.Thehysteresisshowingcurvesareparametrizedbytime,witharrowspointingtothedirectionoftimeaxis.Colorsmarktimeintervalsidentiedasdominatedbybed/water-columnmassexchange(red:bedloss,blue,bedgain)orbylateraluxconvergence(green:advectivegain). ThefactthatthetwocurvesinFigure 4-7 havedifferentshapesissignicant.Undertheassumptionthatthemassuxesarestrictlyverticalandtheonlysourceofsedimentisthebed(knowntobeincorrectduringthewaterfrontpassage),thetransferfunctionthatmapsthetwopanelsofFigure 4-7 ontoeachotheristhebeddensity.Theshapeswouldbeidentical(withsignsreversed),ifthedensityofthemobilizedlayerswasconstant.Thedifferentshapes,therefore,offersomeinsightintothelikelyvariabilityinthedensityofthemobilizedbedlayers.However,thephysicalinterpretationofFigure 4-7 isnottrivial.First,thecurvesusesimulatedbedstressesthatarenotrealisticeverywhere;forexample,thestressislikelyoverestimatedduringwaterfrontpassage.Second,thedensityestimatederivedbytakingtheratioofthetwocurveshaszeroswhenthenumerator(water-columnux)cancelsandsingularitieswhenthedenominator(bed-volumeux)cancels.Thesepointshavetobediscarded.Ifmassexchangewerestrictlybetweenthewatercolumnandthebed,thenumeratorandthedenominatorwouldcancelsimultaneously.Thisisclearlynot 62

PAGE 63

4-7 ,andlargelyagreeswiththeproposedinterpretation.Notethatasexpectedtheadvection-dominatedgreenregionspansthetimeintervalbetweenthezerosofthetwouxes.Withinthedomainofvalidityofthemodel,theassumptionthatsedimentmassbalanceisdominatedbyverticalexchangebetweenthewater-columnandbedpermitsaroughestimationofthedensityofthebedlayerseroded/depositedduringthestorm(Figure 4-8 d).Thevaluesindicateasofteningofthebedunderwaveactionduringtheerosionstage(March4th00:00to12:00hours),andasteadyincreaseindensitystartingwithhinderedsettling,pastthegellingpointduringconsolidationbasedonthemeasuredcharacteristicvaluesofAtchafalayamudatthegellingpoint(Gel=1100kg/m3[ Robillard 2009 Sheremetetal. 2011 ],Figure 4-8 d).Thebeddensityattheonsetof 63

PAGE 64

AsummaryofthebedreworkingcycleduringthestormofMarch3rdto5th,2008.A)Windspeedanddirection,andsignicantwaveheights(short-wave:blueandswell:red); B)Salinity(blue)andtemperature(red)at55cmab ;C)VerticalproleofSSCestimatedbasedonthePC-ADPbackscatter;D)Densityoftheremoved/depositedbedlayersestimateddirectlyastheratioofthemass/volumeuxes(Figure 4-7 ).InD),theredcirclesmarkthedensityvaluesindependentlyestimatedfromtheobservedyieldstressrange.ThedashedlinemarksthegelationdensityforAtchafalayamud[ Robillard 2009 ].Bluerectanglescovertheperiodswherethenumericalmodelusedinthisanalysisislikelyinvalid. erosionandvaluesindependentlyestimatedfromtheobservedyieldstressrange(redcirclesinFigure 4-8 d)areremarkablyclose.Figure 4-9 showsanidealizedsketchofbedstateevolutionduringthestorm,startingfromuidizationwithdownwardwaterux,andleadingtobederosion,sediment 64

PAGE 65

Idealizedsketchofbedreworkingduringthestorm. resuspension,touid-mudformationthroughhinderedsettling,andeventuallytodewateringandconsolidation. 4-4 c.Thissuggeststhatthesea-oorgoesthroughasimilarbedreworkingcycleunderstrongwaves,andallowsforgeneralizingtheresultspresentedinthischaptersofar.AnotherstorminvestigatedtookplaceinlateFebruaryatthesameobservationlocation(Platform3,4-mwaterdepth,Figure 2-1 )withthestormdiscussedintheprevioussection.ThestormstartswhenwesterlywindschangetonortherlyonFebruary26that12:00hours(Figure 4-10 a).Seasreachingalmost1-mheightwereproducedbythestorm.Differentfromthestormdiscussedinprevioussections,theenergyofswellwasinsignicantthroughoutthestorm(Figure 4-10 a).Withdecreasingwindspeedto5m/s,signicantwaveheightdroppedtoabout0.1-mattheendofthestorm.ThePC-ADPcapturedthreesignicantcurrentpulsesrsttwoofwhichtookplaceonearlyFebruary27thowingtowardsSSW.Thethirdstrongercurrentpulseowing 65

PAGE 66

4-10 bandc).Duringthersttwopulses,signicantdropintemperatureandsalinity,andincreaseinSSCwereobservedsuggestingafastfreshwater,likelysediment-loaded,incometotheexperimentalsite(Figure 4-10 eandf).AfterFebruary27that10:00hours,withstartofcurrentdirectionchangeslowlytowesterly,thetemperatureandsalinityvaluesstartedincreasing,SSCstarteddecreasing.Duringthepeakofthethirdpulseonthesamedayat12:00hours,SSCvaluesincreasedsignicantlyagain,probablyasaresultoflocalresuspensionduetoincreasingbottomstressinducedbyhighcurrentspeed.Thefocusintherestofthissectionwillbeaone-dayintervalbetweenFebruary27th-February28thwithstrongestseasandhighestconcentrationsmeasured.Theestimatedbedpositionbasedmaximumbackscatterintensity(blackcurveinFigure 4-11 b)showsasimilarresponsetowaveandcurrentaction.DifferentfromtheobservationsmadeduringtheeventdiscussedinSection 4.2 ,sinceswellenergyislowthroughoutthestorm,beddilationinthebeginningofthestormdoesnotoccur.Thebedstartsbeingerodedbetween02:00hours-05:00hoursonFebruary27th.Depositionstartsfollowingthedecreaseincurrentspeeduntilthestartofthethirdstrongcurrenteventat10:00hourswhichcausessignicantresuspensionofthebedsediment.Withtheendofthethirdcurrentpulse,depositionandconsolidationareobserved.Thebedpositionisrecordedatabout4-cmaboveitsinitialpositionattheendofthestormduringuid-mudformationandconsolidationstages.Theseinterpretationsareinagreementwiththepreviousobservations(Section 4.2 )thattheevolutionofseaoorthroughoutaone-daystormeventgoesthroughthefollowingcycle:mobilizationandresuspensionofbedsedimentthroughpossibleliquefactionanderosion,whichisfollowedbysettling(uidmudformation)andconsolidationaswaveenergydecays. 66

PAGE 67

StormofFebruary26thto28th,2008:A)Windspeedanddirectionandsignicantwaveheights(short-wave:blueandswell:red).VerticalstructureofmeancurrentrecordedbytheB)ADCPandC)PC-ADP;D)DirectionofPC-ADPmeancurrents;E)SSCobservedbytheOBSslocatedat18cmab(blue)and10cmab(red);F)Salinity(blue)andtemperature(red)at55cmab.LocationsoftheinstrumentsareshowninFigure 2-2 ). SuspendedsedimentconcentrationsestimatedusingthebackscatterinversionmethodagreewellwiththeonesmeasuredbytheOBSswiththeRMSerrorof0.62kg/m3andcorrelationcoefcientr2=0.88(Figure 4-12 b,seethecorrespondingtimesegmentinFigure 4-5 ).ThebottomstresseswerecalculatedbythenumericalmodelusingthesamecalibrationapproachexplainedinSection 4.3.2 (Figure 4-12 c).Figure 4-12 suggestthattheamountofthesedimentinthewatercolumniscontrolledbythevariationsinthebottomstress.Theshearstressvaluesarelessthan0.3Paduringlow 67

PAGE 68

EvolutionofbedpositionindicatorsandstateduringthestormofFebruary27thto28th.A)Windspeedanddirection(color-codedthickline),andsignicantwaveheights(short-wave:blueandswell:red);B)normalizedPC-ADPacousticbackscatterintensity.Thesmoothedestimateofthebedpositionismarkedbythecontinuousthickline.;C)Salinity(blue)andtemperature(red)at55cmab(sameasinFigure 4-10 f). wave-currentactivityperiods.Thetwosignicantresuspensioneventscoincidewithtwoclearpeaksinthemodeledbottomstress,withamaximaataround0.6Pa.ThersteventappearstohavetriggeredthebederosionprocessthatstartedinthemorningofFebruary27th,whenthesimulatedshearstressisintherangeof0.3Pato0.5Pa.ThiserosionalcriticalshearstressrangeremarkablysupportsthendingssuggestedinSection 4.4.1 .ThevariationofthenetsedimentmassuxwithbottomstressshowsasimilarbehavioruntilFebruary27that09:00hourstothatinMarch3rdtoMarch5thstorm.The 68

PAGE 69

AsummaryofbedreworkingcycleduringthestormofFebruary27thA)Signicantwaveheight(short-wave:blue,swell:red);B)verticalstructureofmeancurrentrecordedbythePC-ADP;C)normalizedPC-ADPacousticbackscatterintensity;D)SSCverticalstructure;E)Bottomshearstressatbedlevel. behaviordifferswhenthethirdcurrentpulsestartswhichtriggerstheseconderosionevent.Becauseoftheerosioncausedbythethirdpulse,afterFebruary27that07:00hours,massuxvaluesbecomepositiveagain,untilthethirdcurrentpulsedecaysat12:00hours.Afterthat,similartothepreviousresults,depositionaluxisobservedattheendofthestorm(Figure 4-13 a).Therelationbetweenthesimulatedbedshearstressandthenetbed-volumeuxisseeninFigure 4-13 b.Again,untilthestartofthethirdcurrentpulse,thebehaviorisconsistentwithexpectations,erosionwithincreasingstressanddepositionwithdecreasingstress.Thirdpulsecausestheaccretionofthebedforashortduration(one-twohours),thendepositionstartsagain.ThefactthatthetwocurveshavedifferentsignsassurestheeffectofsedimentadvectionbetweenFebruary27that04:00hoursto05:00hours,and09:00hoursto 69

PAGE 70

Observedmass(water-column)/volume(bed)uxvs.numericallysimulatedbedstress.Thehysteresisshowingcurvesareparametrizedbytime,witharrowspointingtothedirectionoftimeaxis.Colorsmarktimeintervalsidentiedasdominatedbybed/water-columnmassexchange(red:bedloss,blue,bedgain)orbylateraluxconvergence(green:advectivegain). 10:00hours,againspanningthetimeintervalbetweenthezerosoftwocurves.Thethirdgreenpartneedstobeseperatedbecausethebedlevelfallsduetoconsolidationofthesettledmaterialratherthanerosion,thereforelateraluxconvergencedoesnotseemimportantduringthisperiod(seealsotemperatureandsalinityvaluesinFigure 4-11 c).Timeevolutionofthebeddensityisestimatedbasedontheverticalmassbalancebetweenwatercolumnandbedduringthedomainofvalidityofthemodel,e.g.,excepttheperiodsofadvective-gainandnon-stationaryconsolidation-dominatedtimesegment(Figure 4-14 ).Similartopreviousresults(Section 4.4.2 ),thevaluesindicateasofteninginearlystagesoferosion,andincreaseindensityafterFebruary27th10:00hourssincethebedlayersbecomestiffer(criticalshearstressisalsoincreasing)withdepth.Thebeddensityvaluesshowasteadyincreasewithhinderedsettling,pastthegellingpoint(gel=1100kg/m3)duringconsolidation.Insummary,thebedevolutionshowsasimilarbedreworkingcycleduringbothstorms(stormsofMarch3rdto5thandFebruary27thto28th).Thevariationofthe 70

PAGE 71

AsummaryofthebedreworkingcycleduringthestormofFebruary27th,2008.A)Windspeedanddirection,andsignicantwaveheights(short-wave:blueandswell:red); B)Salinity(blue)andtemperature(red)at55cmab ;C)VerticalproleofSSCestimatedbasedonthePC-ADPbackscatter;D)Densityoftheremoved/depositedbedlayersestimateddirectlyastheratioofthemass/volumeuxes(Figure 4-13 ).InD),bluerectanglescovertheperiodswherethenumericalmodelusedinthisanalysisislikelyinvalid. bottomstressdrivenbywavesandcurrentsplaysamajorroleinmobilizationofthebedsediment.TheresultssupportthesuggestedrangeofcriticalshearstressforerosiongiveninSection 4.4.1 .Theresultsareencouragingforbuildingastatisticalmodelforthebedreworkingcycle(liquefaction,erosion,uid-mudformationandconsolidationprocesses)andrepresentsarststeptowardaforecastingmodelforwave-bedcouplinginmuddyenvironments. 71

PAGE 72

ThorneandHanes [ 2002 ]forsandtocohesivesedimentaryenvironments.Theprocedureevaluatesaninstrumentconstantusinganinversemethodthatseekstominimizetheerrorbetweentheestimatesandindependentmeasurements(e.g.,optical)atafewnumberofspatialpoints.Thealgorithmisvalidatedusingmeasurementsofowandbackscatterintensityfromanacousticproler(PC-ADP),opticalmeasurements(OBS-3)ofSSC,andoc-sizedistributionobservations(LISST)collectedin2006onthemuddyinnershelffrontingtheAtchafalayaBay,Louisiana,USA.Thedifferencesbetweenestimatedprolesandopticalmeasurementscanbeattributedtolimitationsofthemethodmostofthemrelatedtoyetunresolvedphysicsoftheinteractionbetweensoundandthefractalgeometryofcohesivesedimentocs.Theassumptionthattheocsizeisindependentofheightisprobablythestrongestspecicconstraintimposedonthealgorithmproposedhere.However,thisisinessenceanacknowledgmentoffactorsthatarebeyondthescopeofthisstudy:theoc-size/concentrationambiguityassociatedwithsingle-frequencyacousticobservations;lackofdirectobservationsofverticalvariabilityofocsize;numericalmodelingconstraints.Inprinciple,thisconstraintcanberemovedbyusingdatacollectedwithamulti-frequencyinstrument,byimplementingtheocculationmodelsintoboundarylayermodels[e.g., Winterwerp 2002 ]orwhenmoreadequatesediment 72

PAGE 73

Safaketal. 2010 ].Theoutcome,however,isarangeofsedimentstraticationcongurations,possiblyspanningdifferentinterpretationsofow-sedimentinteraction.EstimatesofSSCprolesbasedonobservationseliminatethisambiguity.Here,weusetheobservedSSCprolestoconstrainauni-dimensionalbottomboundarylayermodel,andusethenumericalsimulationstoinvestigatetherelationshipbetweensedimentstraticationandtemporalocsizevariability.ObservedSSCproleswithsimilarnear-bed(10-15cmab)concentrationsbutcorrespondingtodifferentoc-sizesconrmthetrend(hypothesizedby DyerandManning 1999 ,basedonlaboratoryandeldobservations,andby Safaketal. 2010 basedonnumericalsimulations)thatlowconcentrationspromoteocculation,andsedimentstraticationincreaseswiththeocsize.NumericalsimulationsshowthatacriticaltransitionintheproleofgradientRichardsonnumbercanoccur,relatedtoincreaseddampingofowturbulencebysedimentinduceddensitystratication.Thissupportspreviouslypublishednumericalresults[ Safaketal. 2010 ]. 73

PAGE 74

4-8 a);themeancurrentwasdominatedbythestorm-generatedweaksurgeandthesubsequentushingofthebay;halfwaythroughthestorm,thesitewasimpactedbyasediment-ladenfreshwaterplumefromtheAtchafalayaRivermouth(Figure 4-8 b-c).Thesefeaturesarecharacteristictofrontalpassages,butassuchtheyareexpectedtobeadeningpartofwave-sedimentinteractionduringthesestorms.Intheabsenceofdirectobservationsofbedstates,weanalyzedthehydrodynamicmanifestationofbedreworking.Togetherwiththeacousticbackscatterintensity,theverticalproleofowvelocity(meanandvariance)provideimportantcluesonthelocationandthemotionofthebed,andsuggestasequencesofstagesthatcouldbedescribedas:1)uidization/dilation,2)erosion;3)deposition/accretion;4)uidmudformation;5)consolidation.Atthepeakofactivity,asurciallayerofthebedofabout20-30cmthicknessoscillateswiththewaves(velocities10-20cm/s)andslidesdownslopeat5-10cm/s(Figure 4-9 ).The1DVmodelof Hsuetal. [ 2009 ]isusedtoinvestigatethoseprocessesthatcouldnotbeobserveddirectly.Themodelseeksanequilibriumstateforthehydrodynamic/sedimentarysystembasedonmassexchangebetweenthebedandthewatercolumn.Weareabletotunethemodelwithbothhydrodynamicdata(meanandoscillatoryows)andsedimentdata(verticalprolesofSSC).Themodelperformspoorlyduringnotablynon-stationaryconditions(e.g.,weakbutsteadysettling)andislikelyinvalidduringthepassageofthewaterfront.However,thenaturalspatialuniformityofthesiteandconditionssuggestthatthenumericalsimulationsarereliable 74

PAGE 75

4-8 d).Basedonthegoodnessoftheoverallperformanceofthemodel,thesimulationsareusedtogaininsightintothebedreworkingprocesses.Withinthedomainofvalidityofthemodel,theassumptionthatsedimentmassbalanceisdominatedbyverticalexchangebetweenthewater-columnandbedpermitsaroughestimationofthedensityofthebedlayerseroded/depositedduringthestorm.Thevaluesindicateasofteningofthebedunderwaveactioninearlystagesoferosion,andincreaseindensityduringerosionsincethebedlayersbecomestifferwithdepth.Thebeddensitydecreaseswithstartofdepositionduetoaccumulationofsoftmudonthebedandasteadyincreaseindensitystartingwithhinderedsettling,pastthegellingpointduringconsolidationbasedonthemeasuredcharacteristicvaluesofAtchafalayamudatthegellingpoint(Gel=1100kg/m3[ Robillard 2009 Sheremetetal. 2011 ],Figure 4-8 d).Thebeddensityattheonsetoferosionandvaluesindependentlyestimatedfromtheobservedyieldstressrange(redcirclesinFigure 4-8 d)areremarkablyclose.Calculatednear-bedshearstressessuggestavalueforthecriticalstressforerosionintherangeof0.3Pato0.5Pa.Thesecond,one-daylong,stormisalsostudiedtochecktheapplicabilityoftheseconclusionsforalargerpopulationofstorms.Thegeneralconditionsaresimilartotherststormwiththeexceptionofincreasedswellenergyinthebeginningofthestorm,thereforethebeddilationwasnotobservedduringthisstorm.Anotherdifferenceinthesecondstormisoccurrenceofthethirdstrongcurrentpulsetriggeringthesecondresuspension/erosionevent.However,resultsonthebottomstress-resuspensionrelation,criticalshearstressforerosionandbeddensityvariationthroughoutthestormremarkablysupportthendingsobtainedfromtheinvestigationoftherststorm.Theresultsallowforbuildingastatisticalmodelforthebedreworkingcycle(liquefaction, 75

PAGE 76

76

PAGE 77

ThorneandHanes [ 2002 ]willbepresented.Thescatteredpressure,Ps,fromasingleparticleisdescribedas: A into A : A overthevolumeofinsonication,=r2sindddr,givestheroot-mean-squarepressure,Prms: 77

PAGE 78

Thorneetal. [ 1997 ]asfollowing:I()==2Z02J1(x) 2x 12x 2"1(3.832)2+4"2+6"21(3.832)4

PAGE 79

A and A intoequation A ,rearrangingandexpressingintermsofrecordedvoltage;weobtain[ ThorneandHanes 2002 ] 79

PAGE 80

B-1 and B-2 .Thesamplesarerepresentativeof:1)pre-stormconditionsonMarch3rdat13:00hours,(Figure B-1 -comparewiththepositionofthebedinFigure 4-4 c),and2)bed-mobilizationconditionsattainedwhenthewaterfrontpassedoverthesiteonMarch4th,12:00hours.Thedataarerelativelynoisy,butthewaveformsrecordedareconsistentatallthemeasurementsbins,andstandardmeasuresofquality(ping-to-pingcorrelationandsignal-to-noiseratio)throughouttheprolingrangearewithintherecommendedvaluesgivenbythemanufacturer.Thevariancesmaybeslightlyoverestimatedduetooccasionalspikesintroducedbythealgorithmforvelocityambiguityremoval.Pre-stormconditionsarecharacterizedbyaclearcutoffdepthforhorizontalmotion(Figure B-1 b).Incontrast,bed-mobilizationconditionsinFigure B-2 bsuggestthatwavemotionpenetratedwellbelowtheestimatedbedposition. 80

PAGE 81

SampleofPC-ADPobservationsonMarch3rd,13:00hours,duringaperiodwhenthebedcanbeconsideredasstationary.a)Ping-to-pingcorrelation(blue)andsignal-to-noiseratio(red);verticallinesmarktherecommendedacceptable-valuethreshold;circlesmarkthelocationofthebinsplottedontheright.b)Proles(20-minaverages)ofthemeanandRMSvelocity.c-g)Samplevelocitytime-seriesatthebinlocationsmarkedonpanelsa-b. 81

PAGE 82

SampleofPC-ADPobservationsonMarch4th,12:00hoursUTM,atthemaximumerosionaldepth(Figure 4-4 c).a)Ping-to-pingcorrelationandsignal-to-noiseratio.b)Proles(20-minaverages)ofthemeanandRMSvelocity.c-h)Samplevelocitytime-seriesatbinlocationsmarkedbycirclesonpanelsa-b. 82

PAGE 83

Robillard [ 2009 ]conductedonsamplesfromtheexperimentalsite.Here,weusethestrain-stressow-curveobtainedby Robillard [ 2009 ]fromrheometricoscillatorytests(10-speriod)forsolidsvolumefractions0.0543
PAGE 84

Yieldstressestimatesfordifferentsedimentvolumefractions solidsvolumefractionyieldstress(pre-storm)(Pa)yieldstress(post-storm)(Pa) 0.05430.300.380.1256.246.460.21489.8482.00 FigureC-1. Upper-Binghamyieldstressasafunctionofthesolidsvolumefractionfordifferenttypesofsediment(reproducedfrom Migniot 1968 ).Atchafalayamudestimatesaremarkedbyblackcircles. Robillard [ 2009 ],forAtchafalayamud,usedacontrolled-stressrheometer(AR2000ex)whichcanprovidehighlyaccuratestress-strainrateresolutions. 84

PAGE 85

Allison,M.A.,G.C.Kineke,E.S.GordonandM.A.Goni(2000),DevelopmentandreworkingofaseasonalooddepositontheinnercontinentalshelfofftheAtchafalayaRiver,Cont.ShelfRes.,20,2267-2294. Allison,M.A.,A.Sheremet,M.A.Goni,andG.W.Stone(2005),StormlayerdepositionontheMississippi-AtchafalayasubaqueousdeltageneratedbyHurricaneLiliin2002,Cont.ShelfRes.,25,2213-2232. Barnes,H.A.(1999),Theyieldstressareviewor''everythingows?,J.Non-NewtonianFluidMech.,81,133-178. Betteridge,K.F.E.,P.D.Thorne,andR.D.Cooke(2008),Calibratingmulti-frequencyacousticbackscattersystemsforstudyingnear-bedsuspendedsedimenttransportprocesses,Cont.ShelfRes.,28,227-235. BuscallR.,P.D.AMills,J.W.GoodwinandD.W.Lawson(1998),Scalingbehavioroftherheologyofaggregatenetworksformedfromcolloidalparticles,J.Chem.Soc.,FaradayTrans.1,84(12),4249-4260. Cacchione,D.A.,D.E.Drake,R.W.Kayen,R.W.Sternberg,G.C.Kineke,andG.B.Tate(1995),MeasurementsinthebottomboundarylayerontheAmazonsubaqueousdelta,Mar.Geo.,125,235-257. Chan,I-C.,andLiuP.L.-F.,(2009),ResponsesofBingham-plasticmuddyseabedtoasurfacesolitarywave,J.FluidMech.,vol.618,pp.155-180. Crawford,A.M.,andA.E.Hay(1993),Determiningsuspendedsandsizeandconcentrationfrommultifrequencyacousticbackscatter,J.Acoust.Soc.Am.,94,(6),2247-2254. Dalrymple,R.A.,andP.L.-F.Liu(1978),Wavesoversoftmuds:Atwo-layeruidmodel,J.Phys.Oceanogr.,8,1121-1131. Downing,A.,P.D.Thorne,andC.E.Vincent(1995),Backscatteringfromasuspensionintheneareldofapistontransducer,J.Acoust.Soc.Am.,97(3),1614-1620. Draut,A.E.,G.C.Kineke,D.W.Velasco,M.A.Allison,andR.J.Prime(2005),InuenceoftheAtchafalayaRiveronrecentevolutionofthechenier-plaininnercontinentalshelf,northernGulfofMexico,Cont.ShelfRes.,25,91-112. Dyer,K.R.,andA.J.Manning(1999),Observationofsize,settlingvelocityandeffectivedensityofocs,andtheirfractaldimensions,J.Sea.Res.,41,87-95. Foda,M.A.,J.R.Hunt,andH.-T.Chou(1993),Anonlinearmodelfortheuidizationofmarinemudbywaves,J.Geophys.Res.,98,C4,7039-7047. 85

PAGE 86

Francois,R.E.,andG.R.Garrison(1982b),Soundabsorptionbasedonoceanmeasurements.PartII:boricacidcontributionandequationfortotalabsorption,J.Acoust.Soc.Am.,72,1879-90. Fugate,D.C.,C.T.Friedrichs(2002),DeterminingconcentrationandfallvelocityofestuarineparticlepopulationsusingADV,OBSandLISST.Cont.ShelfRes.,22,1867-1886. Gade,H.G.(1958),Effectsofanon-rigid,impermeablebottomonplanesurfacewavesinshallowwater,J.Mar.Res.,16,61-82. Gartner,J.W.(2004),EstimatingsuspendedsolidsconcentrationsfrombackscatterintensitymeasuredbyacousticDopplercurrentprolerinSanFranciscoBay,California.Mar.Geo.211,169-187. Ha,H.K.,J.P.-Y.Maa,K.ParkandY.H.Kim(2011),Estimationofhigh-resolutionsedimentconcentrationprolesinbottomboundarylayerusingpulse-coherentacousticDopplercurrentproles.Mar.Geo.279(2011),199-209. Hamilton,L.J.,Z.Shi,andS.Y.Zhang(1998),Acousticbackscattermeasurementsofestuarinesuspendedcohesivesedimentconcentrationproles.J.Coas.Res.,14(4),1213-1224. Hill,P.S.,G.Voulgaris,andJ.H.Trowbridge(2001),Controlsonocsizeinacontinentalshelfbottomboundarylayer,J.Geophys.Res.,106,C5,9543-9549. Hoitink,A.J.F.,andP.Hoekstra(2005),ObservationsofsuspendedsedimentfromADCPandOBSmeasurementsinamud-dominatedenvironment.Coas.Eng.,52,103-118. Hsiao,S.V.,andO.H.Shemdin(1980),Interactionofoceanwaveswithasoftbottom,J.Phys.Oceanogr.,10,605-610. Hsu,T.-J.,P.A.Traykovski,andG.C.Kineke(2007),Onmodelingboundarylayerandgravity-drivenuidmudtransport.J.Geophys.Res.,112,C04011,doi:10.1029/2006JC003719. Hsu,T.-J.,C.E.Ozdemir,andP.A.Traykovski(2009),High-resolutionnumericalmodelingofwave-supportedgravity-drivenmudows.J.Geophys.Res.,114,C5,doi:10.1029/2008JC005006. Jain,M.,andA.J.Mehta(2009),Roleofbasicrheologicalmodelsindeterminationofwaveattenuationovermuddyseabeds,Cont.ShelfRes.,29(2009),642-651. 86

PAGE 87

Jiang,F.,andA.J.Mehta(1996),MudbanksofthesouthwestcoastofIndia.V:Waveattenuation,J.CoastalRes.,12(4),890-897. Jiang,F.,andA.J.Mehta(2000),Lutoclinebehaviorinhigh-concentrationestuary,J.Wtrwy.,Port,Coast.andOceanEng.,ASCE,126(6),324-328. Kaye,G.W.C.,T.H.Laby(1986),Tablesofphysicalandchemicalconstants,Longman,UK,pp.477. Khelifa,AandP.S.Hill(2006),Modelsforeffectivedensityandsettlingvelocityofocs,J.Hyraul.Res.,44(3),390-401,doi:10.1080/00221686.2006.9521690. Kineke,G.C.,E.E.Higgins,K.Hart,andD.Velasco(2006),Fine-sedimenttransportassociatedwithcold-frontpassagesontheshallowshelf,GulfofMexico,Cont.ShelfRes.,26,2073-2091. Kranenburg,C.(1994),Onthefractalstructureofcohesivesedimentaggregates,Est.Coas.ShelfSci.,39,451-460. Letter,J.V.,Jr.,A.J.Mehta(2011),Aheuristicexaminationofcohesivesedimentbedexchangeinturbulentows,Coast.Eng.,58,779-789. Libicki,C.,K.W.BedfordK.W.,andJ.F.Lynch(1989),Theinterpretationandevaluationofa3-MHzacousticbackscatterdeviceformeasuringbenthicboundarylayersedimentdynamics,J.Acoust.Soc.Am.,85(4),1501-1511. Liu,K.-F.,andC.-C.Mei(1989),ApproximateequationsfortheslowspreadingofathinsheetofBinghamplasticuid.,Phys.Fluids.,A2,30. Liu,K.-F.,andC.-C.Mei(1993),Longwavesinshallowwateroveralayerofbingham-plasticuidmud-I.Physicalaspects,Int.J.Eng.Sci.,31(1),125-144. Lynch,J.F.,T.F.Gross,B.H.Brumley,andR.A.Filyo(1991),SedimentconcentrationprolinginHEEBLEusinga1-MHzacousticbackscattersystem.Mar.Geo.,99,361-385. Maa,J.P.-Y.,andA.J.Mehta(1990),Softmudresponsetowaterwaves,J.Wtrwy.,Port,Coast.andOceanEngrg.,ASCE,116(5),634-650. Mehta,A.J.(2002),Mudshoredynamicsandcontrols.Muddycoastsoftheworld:Processes,depositsandfunction,T.Healy,Y.WangandJ.A.Healyeds.,Elsevier,Amsterdam,19-60. 87

PAGE 88

Mossa,J.(1996),SedimentdynamicsinthelowermostMississippiRiver,Eng.Geo.,45,457-479. Neill,C.F.,andM.A.Allison(2005),SubaqueousdeltaicformationontheAtchafalayaShelf,Louisiana,Mar.Geo.,214,411-430. Ng,C.-N.(2000),Waterwavesoveramuddybed:atwo-layerStokesboundarylayermodel,CoastalEng.,40,221-242. Robillard,D.J.(2009),ALaboratoryInvestigationOfMudSeabedThicknessContributingToWaveAttenuation,PhD.Dissertation,Online:http://purl.fcla.edu/fcla/etd/UFE0024823,UniversityofFlorida,Gainesville,FL. Safak,I.(2010),Interactionofbottomturbulenceandcohesivesedimentonthemuddyatchafalayashelf,Louisiana,USA,PhD.Dissertation,Online:http://purl.fcla.edu/fcla/etd/UFE0042017,UniversityofFlorida,Gainesville,FL. Safak,I.,A.Sheremet,M.A.Allison,andT.-J.Hsu(2010),BottomturbulenceonthemuddyAtchafalayaShelf,Louisiana,USA.J.Geophys.Res.,115,doi:10.1029/2010JC006157 Safak,I.,,M.A.Allison,andA.Sheremet(inpreparation),FlocbehaviorinhighturbiditycoastalsettingsasrecordedbyLISST:theAtchafalayaDeltainnershelf,Louisiana. Sahin,C.,I.Safak,A.SheremetandM.A.Allison(2011),Bed-SedimentResponsetoEnergeticWaves,AtchafalayaInnerShelf,Louisiana,TheProceedingsoftheCoastalSediments2011,Vol3,pp2415-2424 Sahin,C.,I.Safak,T.-J.Hsu,andA.Sheremet(2012a),ObservationsofsedimentstraticationonthemuddyAtchafalayaShelf,Louisiana,USA,underreviewinMar.Geo. Sahin,C.,I.Safak,A.SheremetandA.J.Mehta(2012b),ObservationsonCohesiveBedReworkingbyWaves:AtchafalayaShelf,Louisiana,underreviewinJ.Geophys.Res. Sheng,J.,andA.E.Hay(1988),Anexaminationofthesphericalscatterapproximationinaqueoussuspensionsofsand.,J.Acoust.Soc.Am.,,83,598-610. Sheremet,A.,andG.W.Stone(2003),Observationsofnearshorewavedissipationovermuddyseabeds,J.Geophys.Res.,108,C11,24,doi:10.1029/2003JC001885. Sheremet,A.,A.J.Mehta,B.Liu,andG.W.Stone(2005),Wave-sedimentinteractiononamuddyinnershelfduringHurricaneClaudette,Est.,Coas.ShelfSci.,63,225-233. 88

PAGE 89

SonM.,andT.-J.Hsu(2011),Theeffectsofocculationandbederodibilityonmodelingcohesivesedimentresuspension.J.Geophys.Res.,116,C03021,doi:10.1029/2010JC006352 Thorne,P.D.,P.J.Hardcastle,andR.L.Soulsby(1993),Analysisofacousticmeasurementsofsuspendedsediments.J.Geophys.Res.,98,C1,899-910. Thorne,P.D.,andR.Meral(2008),Formulationsforthescatteringpropertiesofsuspendedsandysedimentsforuseintheapplicationofacousticstosedimenttransportprocesses,Cont.ShelfRes.,28,309-317. Thosteson,E.D.,andD.M.Hanes(1998),Asimpliedmethodfordeterminingsedimentsizeandconcentrationfrommultiplefrequencyacousticbackscattermeasurements.J.Acoust.Soc.Am.,104(2),820-830. Traykovski,P.,W.R.Geyer,J.D.Irish,andJ.F.Lynch(2000),Theroleofwave-induceddensity-drivenuidmudowsforcross-shelftransportontheEelRivercontinentalshelf,Cont.ShelfRes.,20,2113-2140. Traykovski,P.,P.L.Wiberg,andW.R.Geyer(2007),Observationsandmodelingofwave-supportedsedimentgravityowsonthePoprodeltaandcomparisontopriorobservationsfromtheEelshelf,Cont.ShelfRes.,27,375-399. Urick,R.J.(1948),Theabsorptionofsoundinirregularparticles,J.Acoust.Soc.Am.,20(3),283-289. Walker,N.A.andA.B.Hammack(2000),Impactsofwinterstormsoncirculationandsedimenttransport:Atchafalaya-VermilionBayRegion,Louisiana,U.S.A.J.Coas.Res.,16,996-1010. Wells,T.J.andG.P.Kemp(1981),Atchafalayamudstreamandrecentmudatprogradation:Louisianachenierplain,GulfCoastAssoc.Geol.Soc.Trnas.31,409-416. Winterwerp,J.C.(2001),Straticationeffectsbycohesiveandnon-cohesivesediment,J.Geophys.Res.,106(C10),22,559-22,574,doi:10.1029/2000JC000435 89

PAGE 90

Winterwerp,J.C.andW.G.M.vanKesteren(2004),Introductiontothephysicsofcohesivesedimentinmarineenvironment,Elsevier. Winterwerp,J.C.,R.F.deGraaff,J.Groeneweg,andA.P.Luijendijk(2007),ModelingofwavedampingatGuyanamudcoast,Coas.Eng.,54,249-261. 90

PAGE 91

CihanSahinwasbornandgrewupinAnkara,Turkey.HeobtainedhisundergraduatedegreeinCivilEngineeringin2005,andmaster'sdegreeincoastalandharborengineeringin2007,bothfromtheYildizTechnicalUniversity,Istanbul.Heworkedonparametricwind-wavemodelingbasedonthedatainWesternBlackSea,duringhisM.Sc.studies.Since2008,hehasbeenworkingwithAlexSheremetontheinteractionbetweensurfacewavesandmuddyseaoorusingtheobservationsontheAtchafalayaShelf,Louisiana. 91