DOC

Spatio-temporal

By Veronica Murray,2014-07-19 00:24
11 views 0
Spatio-temporalSpatio

    Spatio-temporal

ActaOceanologicaSinica2005,Vo1.24,No.1,P.107(114

    ;http://www.oceanpress.com.cn

    ;E-mail:hyxbe@263.net

    ;Spatiotemporalvariabilityinseasurfacewindstress

    ;nearandofftheeastcoastofKorea

    ;NAMSungHyun,KIMYoungHo,PARKKyungAe,KIMKuh

    ;1.OceanLaboratory/RIO,SchoolofEarthandEnvironmentalSciences,SeoulNationalUniversity,Seoul,151742,Korea

    ;Received11March2004;accepted20October2004

    ;Abstract

    ;SeasurfacewindstressvariabilitiesnearandofftheeastcoastofKorea,areexaminedusing7kindsofwinddatasetsfrom

    ;measurementsat2coastalfland)stationsand2oceanbuoys,satellitescatterometer(QuikSCAT),andglobalreanalyzedproducts

    ;(ECMWF,NOGAPS,andNCEP/NCAR).Temporalvariabilitiesareanalyzedat3frequencybands;synoptic(2-20d),intraseasonal

    ;(20(

    90d),andseasonal(>90d).SynopticandintraseasonalvariationsarepredominantnearandofftheDo

    nghaeCityduetothe

    ;passageofthemesoscaleweathersystem.SeasonalvariationiscausedbysoutheastwardwindstressduringAsianwintermonsoon.The

    ;seasurfacewindstressfromreanalyzeddatasets,QuikSCATandKMABmeasurementsoffthecoasts

    howgoodagreementinthe

    ;magnitudeanddirection,whicharestronglyalignedwiththealongshoredirection.Atthelandbasedsit

    es,windstressesaremuch

    ;weakerbyfactorsof3-10duetothemountainouslandmassontheeastpartsofKoreaPeninsula.ThefirstEOFmodes(67%-70%)of

    ;windstressesfromreanalyzedandQuikSCATdatahavesimilarstructuresofthestrongsoutheastwardwindstressinwinteralongthe

    ;coastbutshowdifferentcurlstructuresatscaleslessthan200kmduetotheorographiceffects.ThesecondEOFmodes(23%-25%1

    ;showsouthwestwardwindstressineverySeptemberalongtheeastcoastoftheNorthKorea ;Keywords:seasurfacewindstress,spatiotemporalvariability,monsoon,seasonal,intraseasonal,s

    ynopticband,orographiceffect

    ;1Introduction

    ;Understandingsurfacewindstressfieldisessen

    ;tialtostudyoncoastaloceancirculationanditsvari.

    ;ability(Samelsoneta1..2002).Technicaladvancesin

    ;observingandreanalyzingwindsenableustostudy

    ;theirspatio-temporalvariabilityinhighresolution. ;Inthel990stheseasurfacewindfieldsoverthe ;seasaroundtheKoreaPeninsulahavebeeninvestigat

    ;edbyNaeta1.(1992)andNaandSeo(1998).They ;haveinvestigatedthesewindfieldsestimatedfrom ;18-yearsweatherchartsovertheEastAsianmarginal ;seas.Kangeta1.(1994)investigatedwindstressand ;curlfieldusing3-yearsdata.Morequantitatively,Na ;?Correspondinganthor:Email:namsh@ocean.SNU.ac.kr

    ;eta1.f1997)examinedspatiotemporalvariabilityand

    ;periodicityofseasurfacewindstressovertheEastSea ;byperformingEOFanalysis.

    ;Spatialandtemporalscaleshandledintheirstud

    ;iesarelimitedtomorethanamonthandafewhun- ;dredkilometersinspiteofstrongtemporalvariability ;ofseasurfacewindstressattimescaleslessthansea- ;sons.Thereareonlyfewstudiesonthehighfrequency ;windstressvariationsinthisarea.Cheneta1.(2002) ;reportedhighfrequencyatmosphericforcingassociat

    ;edwiththemesoscaleweathersystemusingnonhy-

    ;drostaticatmosphericmesoscalemodel(MM5)overthe ;EastSea.Choi(2002)simulatedseasurfacewindfield ;usingboththree--dimensionalhydrostaticandnon--hy-- ;drostaticmodelneartheeastcoastofKorea.However, ;suchseasurfacewindstressvariabilityhasnotbeen ;

    ;108NAMSungHyuneta1.ActaOceanologicaSinica2005,Vo1.24,No.1,P.107(114

    ;widelydealtwithmostlyduetolackofwindmeasure. ;mentsnearandoffthecoast.Andsatellitesca~erome

    ;ter(e.g.,QuikSCAT)hasprovidedhighspatial(25km) ;andtemporal(daily)resolutionwindfields(Parketa1., ;2003;Lee,1998).

    ;Thus,mainobjectivesofthisstudyare(1)toexam

    ;ineseasurfacewindstressvariabilityattemporalscales ;ofseasonal(>90d),intraseasonal(20-90d),andsyn

    ;opticband(2-20d);(2)toinvestigatespatiallycoherent ;windstressfieldanditscurlvariationoffthecoast;and ;(3)toprovideanevidencefororographicinfluenceusing ;windmeasurementsfromcoastal(1and)stationsando

    ;ceanbuoysnearandofftheeastcoastofKoreainclud

    ;inggloballyreanalyzedwind/windstressdatasets(e.g., ;ECMWF;EuropeanCentreforMediumRangeWeather

    ;Forecast,NOGAPS;NavyOperationalGlobalAtmo

    ;sphericPredictionSystem,andNCEP/NCAR;National

    ;CentersforEnvironmentalPrediction/NationalCenter ;forAtmosphericResearch).

    ;2Dataandmethod

    ;Thedatasetsusedinthisstudyare(1)windsmea

    ;l27.l29.

    ;.|ESROBandKMAB

    ;?KMA—DH(Donghae)

    ;ESOREC

    ;1(7OuikSCAT

    ;4l.

    ;N

    ;39.

    ;37.

    ;35.

    ;suredat2landbasedstationsand2oceanbuoys;(2) ;satellitescatterometerwind(QuikSCAT);and(3)glob

    ;allyreanalyzedwindstressproducts(NCEP/NCAR, ;NOGAPS,andECMWF).TheoceanbuoysaretheEast ;Searealtimeoceanbuoy(ESROB)nearthecoast ;(about10kmfromthecoast)andtheKoreaMeteoro- ;logicalAdministrationMiddleEastSeabuoyfKMA

    ;B1about100kmoffthecoast.Thelandbasedstations

    ;arelocatedattheDonghaeCity(KMADH)andthe

    ;EastSeaOceanResearchCenter(ESOREC)(Fig.11. ;Allthewindstressesarecomputedfromwinddataby ;simplequadraturelaw(Csanady,l984):

    ;PaCdV,

    ;whereP,Cd,andVaretheconstantairdensity(1.223 ;kg/m),constantdragcoefficient(0.0017),scalarwind ;speed,andvectorwindinm/s,respectively.Theinter

    ;valandrangeofthedataareinTable1.

    ;Inordertoseparatethetemporalvariationsintodi

    ;rectionalcomponents,alongshoreandcrossshoredirec

    ;tion,windstressvectorswererotatedtoacoordinate ;systembasedonthecoastlineorientation,withthea. ;1ongshorecoordinateypositivetoward30.anticlock

    ;wisefromthenorthandthecross.shorecoordinate ;i3i.El22.00l22.05l22.l0l22.15122.20E ;Fig.1.PositionofESROB(neartheDonghaeCity),KMA-B(betweentheDonghaeCityandUlleungIsla

    nd),ESOREC(atland).

    ;KMADH(atland),and7boxeswhereQuikSCATdataareaveragedevery12h.withcoastlineandbathy

    metryoflandmassonthe

    ;northemandeastempartsofKoreaPeninsula.Thecontourlabelsareinm.

    ;

;NAMSungHyuneta1.ActaOceanologicaSinica2005,Vo1.24,No?1,P?107(114

    ;Table1.Theintervalsandrangesofdatausedinthisstudy.Note ;thatspaceandlaneintervalsfortheEOFanalysis(numbersin ;brackets)aredifferentfromthoseforthetimeseriesanalysis ;1)Negativevaluedenotesthelandwarddistance;+datapointsare ;coincidentofKMABandNCEP;dataintheareaofNo.5(seeFig.11 ;areaveraged.wheredatapointsdistributedintherangesnoted ;positiveoffshore(seeFig.1,.Temporalvariationsare ;comparedforafixed1.×1.areaneartheDonghae ;City(No.5inFig.1,.Timeseriesdataarereproduced ;to3differenttimescales:seasonal,intraseasonal,and

    ;synopticband,bylowpassorbandpassfiltering.Spa

    ;tialvariationsofthewindstressfieldsareexamined ;comparingspatialmodesandtheircurls.

    ;3Temporalvariability

    ;Figure2showsexamplesoftemporalvariationof ;windstressinthealongshoredirectionnearandoffthe ;eastcoastofKorea.Allthereanalyzeddatashowgood ;agreementinthevariabilityofalongshorecomponent ;ofwindstress.TheNCEP/NCARdata,withcontinu- ;OUStimeseries,wereshownandcomparedwithob

    ;servedwindstressesinFig.2.Strongnorthwesterly ;wind,correspondingtotheAsianmonsoon,makes ;southeastward(negativealongshore)windstressevery ;winterfrom1999to2002(Figs2ag).Themagni

    ;tudesofECMWF,NOGAPS,NCEP/NCAQuikSCAT,

    ;andKMABreachto0.2N/moffthecoastinwinter. ;Bycontrast.themagnitudesofKMADHandES?

    ;ORECdecreasebyfactorsof3(10nearthecoastin

    ;winter(Figs2a,df).Thisdistinctionispossibly

    ;duetoorographicinfluenceontheeasternpartsofthe ;KoreaPeninsulaalthoughitisnotclearinsummer ;(seeFig.20.

    ;3.1Seasonal(>90d)

    ;Thetemporalvariabilityofthelowpassed(>90d)

    ;windstressindicatesthatthewindstressesoffthecoast ;showgoodagreementinmagnitude(0.20-4).25N/m, ;seeFig-3).However,themagnitudesofseasonalwind ;stressvariationsnearthecoast(i.e.,KMADHand

    ;ESOREC)aremuchweakenedbylessthanO.05N/m ;(seeFig-3,.Thoseseasonalvariationsmaybeprimari

    ;lycausedbytheAsianmonsoon.

    ;3.2Intraseasonal(20-90d1

    ;Bandpassed(20-90d1windstressshowsthatthe

;magnitudesofthealongshoreintraseasonalvariation

    ;areinrangeof0.02-0.04N/moffthecoast.whichis ;muchweakerby10%-20%ofseasonalvariation(see ;Fig.3).Similartotheseasonalvariation,thewindstress? ;esnearthecoastr<10km)appeartobemuchweak

    ;enedduetotheorographicinfluence.Relativelystrong ;fluctuationsf?0.2N/m)attheintra—seasonaltime

    ;scalefrequentlyoccurredinspringandfal1.Thisseems ;tOberelevanttosouthwestwardwindstressalongthe ;eastcoastoftheNorthKoreaineverySeptember. ;whichisdiscussedinnexttwosections.

    ;3.3Synopticband(2-20d1

    ;Synopticband(2-20d)variationsofalongshore ;windstresseshavewiderangeofstandarddeviations ;from0.01to0.18N/m.somewhatlargerthanthoseof ;theintraseasonalvariations.Thestandarddeviations ;nearandoffthecoastare0.01~0.04and0.04--0.06 ;

    ;

    ;03

    ;E

    ;03

    ;E

    ;Z

    ;

    ;

    ;

    ;?

    ;OO

    ;

    ;O3

    ;O3

    ;,

    ;Z

    ;0.0

    ;

    ;?aboveaxesrepresentJuliandaysfromJanuary1,1999.Here,thealongshore

    ;directionispositivetoward30.anticlockwisefromthenorthasdescribedinSection2.

    ;N/minsummerand0.02.03and0.1l(0.18N/min

    ;winter(seeFig-3,.Thisindicatesthattheorographicef- ;fectsinwinteraremorepronouncedthaninsummerat ;thesynopticband(Figs2d,e,seeFig.3).

    ;5O53O3

    ;OOOO

;_I

    ;EZ,s?2c?__EZ/s?c?

    ;O33O3

    ;O0O

    ;Z,s?an?pu??l?Z,s??pu?Z,s?Qpu?

    ;

    ;()6

    ;04

    ;()3

    ;)

    ;f()

    ;AMSungtt)LIHetalActaOceanologicaSimca2005.VOI.24.No ;nnopllcband

    ;?IlOplicband

    ;i{,

    ;?11[1aci,l2&

    ;SeaSOII{{j_——

    ;__lll

    ;lc

    ;!il:

    ;K\1-kDf[SORI1NROBK\lf{fK\ll\(X,f\(f()(

    ;?mid1(1and)?ci(off)folnft1,fL,miL,m

    ;Fig.3.Typicalmagnitudesfstandarddeviations1ofthe ;seasona1.intraseasona1.andsynopticband(insummerand ;winter)foralongshorewindstress.Notethatthevaluesare ;invariance(N/m1.Themagnitudesofseasonalandsynoptic ;bandinwinterfortheESROBdataarenotavailane. ;Thesynopticbandvariationsofwindstresscould ;becausedbythepassagesofmesoscaleweathersys

    ;tems(Cheneta1..2002).Windstressatthisbandshows ;strongseasonality:thewindstressfluctuationsinwin

    ;terreachupto0.5-0.6N/m!offthecoastwhereasthey ;aresinailto0.1-0.2N/m:insummer(seeFjg.2)with ;exceptionduringafewevents(forexample,about0.4 ;N/mduringthepassagesoftyphoon).

    ;4WindstressE0Fsandtheircurls

    ;TheECMWF,NOGAPSandOuikSCATdatafor

    ;theperiodsshowninTable1arelowpassedtoex

    ;cludehighfrequencywindstressvariationswithape

    ;riodlessthan20d.Theempiricalorthogonalfunction ;(EOF)analyseswerethenappliedtodecomposethem ;intodominantpatterns.Inordertoexaminefinescale

    ;spatialstructureweuseddifferentproductsof ;ECMWF,NOGAPS,andOuikSCATdata.Notethat

    ;thegridsizesofECMWF,NOGAPS,andOuikSCAT ;are0.51.00,and0.5respectively.whicharedifferent ;fromthosef0rthetimeseriesanalyses(seeTable1,. ;ThetimecoefficientsfortheEOFsarenormalizedbv ;eachstandarddeviation.

    ;andtheEOFsaremultiplied

    ;bythestandarddeviations.Thenwindstresscurlsare ;calculatedfromthedecomposedEOFfields.

    ;Figure4showt]rsiEOFmodesfvectors)andtheir ;cur1s(contours)ofFCMWFN(GAPS.andOuikSCAT ;windstress.Thesefirstmodes.accountingfor67%(

    ;70Oftota1variancefseeTable21havesimilarpat

    ;ternsthatthealongshorecomponentisdorainantand ;thedirectionissoutheastwardinwinter.whichiscon

    ;sistentwiththeresultsintimeseriesanalyses.Thefirst ;modeEOFofECMWFandOuikSCATwindstresses

    ;hasstrong(>25×10N/mattheCO1_c1positivecurlin ;winterbetween39.and40.NfFigs4aandc1whj1ethat ;ofNOGAPShasweakerandbroadcurlsf15×10

    ;N/matthecore)around38.N(Fig.4b,.Thepositive ;windstresscur1offtheWonsanCits,canbecausedbv ;strongwindstressthroughthegapbetweenthemoun

    ;tainous1andmassesinwinter.whichwerealsoshown ;fromNSCATanalysis(Parketal,.2003).Thecompar

    ;isonsofpositivecurlofftheWonsanCitysuggestthat ;theECMreanalyzeddatacorrectlyaccountfolthis

    ;orographicef-fect.

    ;Table2.TheresuhsofEOFanalvsis”)rECM\\F.

    ;NOGAPS.

    ;andQuikSCATwindstresses.wheretheunitsareinpercentage ;bywhicheachmodeexplainstotal,ariance.Notethatthefirst ;twoEOFmodesexplainover90oIoftotal,ariance

    ;ThesecondEOFmodeofHrindstressexplain

    ;23%(25%oftota1variances.Thefirsttwomodesac

    ;countformost(>90%)oftota1variancesfseeTable21. ;Thestructuresalsohavesimilarpredominantpattema. ;mongthedatasetsandtheirdirectionissouthwestward ;alongtheeastcoastoftheNorthKoreainwinterfFig.5). ;Thismodeshowsweak(?15×10N/in1negatix,e

    ;curlsnearthenorthern/northwesternpartsofthestudy ;area.Theweaknegatix,ecurlsnearthecoastcanbe ;causedbythehorizonta1difierencesofsouthwestward ;windstressesduetotheorographicinfluenceofthe ;highmountains.asshownbythcstrongwindsthrough

;zJ1?

    ;

    ;

    ;Ih~vallwofflh~Vladivostokl/.ussla(Parkctat ;21103)

    ;TimeCOefficientsofnrttwoEOFmodesdlc ;“highlyeorvehlledamongthedatasetsandl?estrong

    ;„„seasonalvariationswithposirwevaluesinevelTDe‟

    ;

    ;_?cembtoFebruarycon‟espondingtotheperiodofthe

    ;20AsianwhiterIi1OnSOOHfsecFl61Thesecondnlode ;31)11assignificantIiltraseasona1~aliationswithpeaks1i1 ;0Septemberindie,~tingthc1elevanceVdThinml-seaslmal ;windstressvariationsnoarandoiltheeagercoagtol ;Koren[isFerI_edinpreviou~~ecgonIns~tble…e

    ;rtilcrclcvantphcilo[1lcnOlldiscussedinthel

    ;IowiIsection

    ;

    ;JIN]m‟

    ;l;-

    ;2

    ;

    ;30

    ;0INm

    ;FigSamen…nFig4butm.uTEt…l,d

    ;5Concludingremarksanddiscussion

    ;Sc~cnkindsofobsen,edandreanah‟zcdwind

    ;s[r~ssdatasetsJn1nI999to201)2.Iramdvzedner

    ;and(ntheeast~oastofKoreaT~picalmamlitudesol ;theseasona1.1nfla-seasonalandsynopticbandwind ;rcssvariationsoffthe~ilstcr)nslofKoreaalc020 ;u03and0040IHN/m_1respectivelywhereasthose ;lcrIh‟(I‟IccracbyIcurof3-10inwinter

    ;TheAsiani]~t)llNOOllissugpested”?Ihenl?hieUHtlNP

    ;l0rtheseasonalvariationsandtheestuardpassages ;_Ithen,es~lscaleeathmsvstonlsI_thesD~optie ;band~ariabjIitvThcsynopticbandwdstl-~s$vari-

    ;abIityl1…rnng.1l_nhIvht-l-i?ningI”1nrln

    ;winter105(06N/n11thanj11sl1mmcr”1I—u2N/m,l

    ;H0wreFitreachedapeakolIJ,cr1,1I:duringthe

    ;I,Phoo11cVcnIsc,o11i11SU111nlel

    ;ThclhsttwoEOFsofECMWrlNfJC~APS.1jlLI ;QuikSCA1windS[FPSRe~explainl1,el-q(1n?of”total

    ;Valianceli1eachdatasetThefirstmodeshowecit1nng

;~eaSOilatvnf1.1lIOrlStllong[hecncogistTheircurls

    ;denlonstratethattheSllongwindStl‟~SscscaLJcpusi ;tivcurl”ITtheWansan‟ilyducit1thegapbetween ;IhcmountainousIandmasse~TheECMWFdalarela ;tl,eI3wclIdesclibedthet)rl~graphlcetleetsamongthe

    ;folln~llyzeddatasetsAIIthesecondEOFmodessl_0wed

    ;strongsouth~estwardrindsIicf0ll_I):~mbelIo

    ;._”__

    ;

    ;

    ;4O

    ;2O

    ;00

    ;

    ;20

    ;

    ;40

    ;4O

    ;2O

    ;00

    ;

    ;20

    ;

    ;40

    ;4O

    ;2.0

    ;..

    ;三一2o

    ;

    ;4.0

    ;NAMSungHyuneta1.ActaOceanologicaSinica20{}5Vo1.24-No?1,P?107(114

    ;99920002002O)2

    ;,

    ;i

    ;1:

    ;365902()2

    ;{,,2‟

    ;36590:O2

    ;/

    ;,

    ;?

    ;I,l,

    ;?

    ;365902O:

;Perfod/d

    ;O5

    ;04

    ;O3

    ;02

    ;1)j

    ;00

    ;05

    ;O4

    ;03

    ;02

    ;O1

    ;00

    ;05

    ;O4

    ;03

    ;O2

    ;Ol

    ;OO

    ;TImr2

    ;Fig.6.Timecoefficients(1eft)andtheirspectralvariances(fight)forthefirsttwoEOFmodesofECMWF

    (a),NOGAPS(b),andQuikSCAT

    ;(c)windstresses,wheresolidanddashedcurvesdenotethe1stand2ndmodes,respectively.The1stmode

    shavepositivevaluesin

    ;veryDecembertoFebruarywhile2nd

Report this document

For any questions or suggestions please email
cust-service@docsford.com