Julyseasonalwinds

1PastsummerupwellingeventsintheGulfofOmanderivedfromacoralgeochemicalrecordTakaakiK.
Watanabe1,TsuyoshiWatanabe1,AtsukoYamazaki1,2,MiriamPfeiffer3,DieterGarbe-Schnberg4&MichelR.
Claereboudt5Weusedahigh-resolutionoxygenisotope(δ18Ocoral),carbonisotope(δ13Ccoral)andSr/Caratiosmeasuredintheskeletonofareef-buildingcoral,Poritessp.
,torevealseasonal-scaleupwellingeventsandtheirinterannualvariabilityintheGulfofOman.
Ourδ13Ccoralrecordshowssharpnegativeexcursionsinthesummer,whichcorrelatewithknownupwellingevents.
Usingδ13Ccoralanomaliesasaproxyforupwelling,wefound17summerupwellingeventsoccurredinthelast26years.
Theseanomalousnegativeexcursionsofδ13Ccoralresultfromupwelledwaterdepletedin13C(dissolvedinorganiccarbon)anddecreasedwater-columntransparency.
WereconstructedbiweeklySSTsfromcoralSr/Caratiosandtheoxygenisotopiccompositionofseawater(δ18OSW)bysubtractingthereconstructedSr/Ca-SSTfromδ18Ocoral.
Significantδ18OSWanomaliesoccurduringmajorupwellingevents.
Ourresultssuggestδ13CcoralanomaliescanbeusedasaproxyforseasonalupwellingintensityintheGulfofOman,which,drivenbytheIndian/ArabianSummerMonsoon,issubjecttointerannualvariability.
TheGulfofOmanislocatedonthenortheasterncoastoftheArabianPeninsulaandboththeArabianSeaandtheGulfofOmanarelocatedinaridenvironments.
TheclimateisdominatedbytheseasonalreversaloftheIndian/ArabianMonsoon,whichinturngovernsthesurfacewindfieldoftheIndianOceannorthof10°S.
Theintensityanddirectionofthemonsoonwindsvaryseasonally.
Duringthesouthwest(SW)Monsoondevelopsduringtheborealsummer(fromJunetomid-September)andischaracterizedbystrongairflowacrosstheArabianSeathatfeedsmoistureandrainfalltotheIndiansubcontinent.
TheIndian/ArabianSummerMonsooncausescoastalupwellingbringingcoolertemperatures,nitrifiedandsalinewatertotheseasurfacealongthesoutherncoastoftheArabianPeninsula.
Upwelledwaterflowsnorth-wardandaffectstheoceanicstratificationoftheGulfofOmanthroughgyresandeddysystemsthatsweepintotheOmanSea1.
TheNorthernArabianSeaisthereforeoneofthemostproductiveareasintheworld2.
TheSWMonsoonisalsothemajorclimaticfactoraffectingthenear-shoreenvironmentandareasofcoralgrowthinOmanduringthesummermonths3.
Thehighnutrientcontentofthiswaterinducesphytoplanktonblooms.
Satellite-basedoceancolormeasure-mentsshowthetemporalandspatialvariabilityofthesurfacechlorophyll-adistributionalongthecoastoftheSouthernArabianPeninsula4.
IntheGulfofOman,upwellingdoesnotnecessarilyoccureverysummer5–7.
Inaddition,observationalrecordsthatallowustounderstandthedynamicsofupwellingeventsintheGulfofOmanarescarce.
Satellitebasedseasurfacetemperature(SST)intheGulfofOmandidnotreflectlowSSTexcursionsinsummermeasuredbyCTDs.
(Fig.
S1).
Long-termandinsiturecordsofprimaryproduction,salinityandtemperaturearenecessaryinordertounderstandupwellingevents7.
Inthisstudy,weusedpaleo-climaticrecon-structionsfromcoralgeochemicalrecordstoprovideahistoryofsummermonsoon-drivenupwellingvariabilityintheGulfofOman.
Thegeochemicalproxiesincoralskeletalcarbonateprovidealong-termhistoryofenvironmentalvariation,withhightimeresolution(2weekstoamonth)8,9.
Coralskeletaloxygenisotopes(δ18Ocoral)reflectSSTandoxy-genisotopesinseawater(δ18OSW)10,11.
CoralskeletalSr/CaratiosareusedasSSTproxies12.
Sea-surfacesalinity1DepartmentofNaturalHistorySciences,FacultyofScience,HokkaidoUniversity,Sapporo,060-0810,Japan.
2AtmosphereandOceanResearchInstitaute,TheUniversityofTokyo,Kashiwa,277-5564,Japan.
3RWTHAachenUniversity,GeologicalInstitute,Wuellnerstrasse2,52056,Aachen,Germany.
4InstituteofGeosciences,UniversityofKiel,Ludewig-MeynStrasse10,24118,Kiel,Germany.
5DepartmentofMarineScienceandFisheries,CollegeofAgriculturalandMarineSciences,SultanQaboosUniversity,Box34,Al-Khod,123,SultanateofOman.
CorrespondenceandrequestsformaterialsshouldbeaddressedtoT.
W.
(email:nabe@mail.
sci.
hokudai.
ac.
jp)Received:4November2016Accepted:22May2017Published:xxxxxxxxOPEN2(SSS)isderivedfromδ18OSW,whichisgeneratedbysubtractingthetemperaturecomponent(obtainedfromcoralSr/Ca)fromδ18Ocoral12.
Stableisotopesofcarbonincoralskeletons(δ13Ccoral)areinfluencedbykineticiso-topicfractionation,vitaleffects(photosynthesisandrespiration)andthecarbonisotopiccompositionofdis-solvedinorganiccarboninseawater(δ13CDIC-SW)13–16.
Coralskeletonsareprecipitatedinisotopicdisequilibriumwithambientseawaterasaresultofkineticandvitaleffects.
Thekineticeffectselectivelydepletes12Cand16Oincoralskeletonsandisparticularlyimportantwhencoralgrowthratesareverylow(0.
05:Fig.
S2a).
Winterδ13Ccoralhadnosignificantcorrelationwithwinterδ18Ocoral(r=0.
04,P>0.
05,n=159:Fig.
S2b).
TheextensionratesshowthattheOmancoralgrewveryquickly,onaverage25.
1mm/yearwitharangebetween19to31.
5mm.
Thesevalueswereconsiderablyhigherthanthecriticalvalueestimatedforkineticisotopicfractionationeffects(4mm/year)(Fig.
1f)14.
Therefore,thecoralgrowthhistoryandthelackofcorrelationbetweenδ13Ccoralandδ18Ocoralsuggestthatthekineticisotopiceffectdidnotsignificantlyaffectthiscoralrecord.
Previousstudiesreportedδ13Ccoralonseasonalandinter-annualvariationsareattributabletosolarradiation17,26.
Toinvestigatetheprocessesdrivingtheseδ13Ccoralfluctuations,wecomparedδ13Ccoralwithsatellite-basedoutgoinglongwaveradiation(OLR)(Fig.
S3a)whichreflectcloudcover.
Foracomparisonofδ13Ccoralwithmonthly-resolvedOLRdata,biweeklyresolvedδ13CcoraldatawereresampledatamonthlyresolutionusingthesoftwareAnalySeries(version2.
0.
8)27.
Theδ13CcoralwerecomparedwithOLR,andwecalculatedthecorrela-tioncoefficientsbetweenthesetimeseries.
δ13Ccoralwithoutanomalousδ13CcoralexcursionspositivelycorrelatedwithOLRatasignificantlevel(r=0.
411,P0.
3Fig.
S4candS4d),suggestingthatanomalousnegativeexcursionsofδ13Canomalyinthesummer(AN-δ13C)wouldnotbegeneratedfromOLRvariations.
WeexaminedthetimingoftheAN-δ13Cwiththecompiledevidenceofeachpastupwellingeventdocumentedfrominsituandsatelliteobservations(Fig.
1fandg).
AbruptSSTdecreasingeventsinsummerwererevealedin1987–198929and200030fromsatelliteSSTdata,in19926,19944,200130,200230(Fig.
S5)and199031basedoninsituSSTdata,and2010basedonourverticalseawatertemperatureprofile(Fig.
S6).
Theverticalprofileofseawatertemperaturededucedbytemperaturesensorsattachedtothedivinggearoflocalvolunteerdiversin2010,alsosuggestthatthethermoclinewasclosertothesurfaceduringsummerupwellingevents(Fig.
S6).
Inaddition,Al-Azrietal.
1hadmeasuredchlorophyll-aconcentrations,nutrients,phytoplanktondensityandSSTinFahalIsland(23.
67°N,58.
5°E)andBandarAlKhayran(23.
51°N,58.
72°E:neartoourcoralsamplesite).
FromJulytoSeptember2004,upwellingwasobservedasincreasingchlorophyll-aconcentrationsandphytoplanktondensityaswellasdecreasingSST1.
InAugust2005,SSTdecreasedfor1month,whileotherparametersdidnotchange1.
Al-Azrietal.
,2012reportedthatinsituchlorophyll-aandsatellitebasedSSTsuggestedthatupwellingalsooccurredinJuly,20087.
Thesatelliteobservations(SeaWiFSandMODISat24°N,58°EfromAsia-PacificDataResearchCenter32)from1997to2013suggestedthatchlorophyll-aconcentrationsintheGulfofOmanincreasedinAugust2000,September2004andAugust2008(Fig.
S7).
Inotherupwellingyears,chlorophyll-a4concentrationsinsatellitedatawerenotavailabletocomparewithAN-δ13Cduetothelackofsatellitedatainsummer.
Basedontheseinsituandsatellitedatasets,pastupwellingeventsoccurredin198729,198829,198929,199031,19926,19946,2000–200230,20041,20087and2010(Fig.
1g).
TheAN-δ13Ccorrespondswiththesepastupwellingevents.
ThepossiblecontrollingfactorsoftheAN-δ13Cwithupwellingeventsare:(1)decreasingwater-columntrans-parency33,(2)variationsofδ13CDIC-SW16,34,and(3)changetoheterotrophfeeding10.
Itisknownthatincreas-ingchlorophyll-aconcentrationscorrespondwithupwellingevents1,7inducingphytoplanktonblooms,therebydecreasingwater-columntransparencyanddepleting13Ccoralwithlowphotosyntheticactivitiesofzooxanthellae33.
Moreover,lowerδ13CDIC-SWsupplyfromgreaterdepthsdecreasesδ13CDIC-SWattheseasurface24,25,34.
UpwellingeventsmayproduceanAN-δ13Cduetosuddendecreasesinwater-columntransparencyandδ13CDIC-SW.
Heterotrophicfeedingwouldalsobethecontrollingfactorofnegativeδ13Ccoralwithupwellingevents.
Astudy35reportedthatcoralsfeeding13C-depletedzooplanktondecreasedtheirδ13Ccoral.
ThecoralrecordsfromtheGulfofAqaba,RedSeasuggestedthatincreasingheterotrophywithupwellingdecreasedδ13Ccoralforanapproximatelyhalfayear10.
Afterwards,δ13Ccoralcouldbeincreasedbythepreferentialuptakeof12Cbyphytoplanktonattheseasurface16.
InthewesternIndonesiancoast,itwasreportedthatδ13Ccoralincreasedbyapproximately2.
2‰VPDBafterlargephytoplanktonbloomsduetoupwelling16.
Weproposethefollowingmechanismtoexplaintheshort-termnegativepeaksintheδ13Ccoral:1.
Upwellingeventsbringdeep,coldandnutrient-richwaterwithlowδ13CDIC-SWtothesurfaceinsummer.
UpwellingeventscauseunusuallyhighnutrientconditionsintheGulfofOman.
Photosynthesisactivitiesinzooxanthellawouldbeemphasizedineutrophicconditionsandtemporarilyincreasedδ13Ccoral.
2.
Lowerδ13CDIC-SWfromthedeepseadecreasesδ13Ccoral.
3.
Phytoplanktonbloomsarisefromanutrientsupplytotheseasurface.
4.
Phytoplanktonprimarilydepletes12CO2-SW.
Activephytoplanktonphotosynthesisincreases13CO2-SW.
5.
δ13Ccoralincreaseswiththerestorationofδ13CDIC-SW.
WecomparedtheAN-δ13Cminimawiththeupwellingperiods(numberofthedays)insummer(Figs2a,S5andS6).
InsitudailytoweeklySSTdatain1992,1994,2001,2002and2010revealedthatSSTduringupwellingeventswasassameaswinterSST(23.
5°C),anddailyfluctuationsofSSTinupwellingperiodsrangedwithin3°C.
Therefore,thenumbersofthedaysforupwellingperiodsweredefinedasthedurationofSSTlowerthan26.
5°Cinsummer.
δ13Canomalyvaluesofnoupwellingyears(0days)wasestimatedfrominsituδ13CDIC-SWinArabianSea(+1.
325‰VPDBat0–10mdepthinnon-upwellingseasons36)andthevalueofδ13Cinisotopicequilibriumbetweencoralcarbonateandseawater37.
TheAN-δ13Cminimawerecorrelatedtotheupwellingperiodsasbelow.
Upwellingperiods(days)=87.
16±16.
40*AN-δ13Cminima(‰VPDB)4.
92±9.
45(r=0.
937,P<0.
05;Fig.
2a).
Then,pastupwellingperiodsintheyearwithnoinsituSSTdatawerereconstructedfromeachAN-δ13Cusingthisequation(Fig.
2b).
Theestimateduncertaintyforreconstructedupwelling-periodswas12.
66days(1σ)includingtheanalyticalprecisionsofδ13Ccoral,theinterceptandtheslopeofthisequation.
In1987,2006,2008,2009,eachupwellingperiodwasextremelylong,over120days(Fig.
2b).
Inthoseyears,coralextensionratesdecreasedto23mm/year(Fig.
1e).
Thelongupwellingeventswouldthereforehaveanegativeeffectoncoralextensionrateduetoeutrophicconditionsanddecreasedwater-columntransparency.
WecomparedthereconstructedupwellingeventsfromAN-δ13C(Fig.
2b)withSr/Caratiosandδ18OSW-anomaly(Fig.
1bandc).
Sr/Caratiosshowed1-monthincreasing(cooling)insummerexceptin1994,2001,2002,2006,and2009,however,thesedidnotcorrespondtoreconstructedupwellingevents.
Innon-AN-δ13C(upwelling)years(1989,1991,1997–1998,2003,2007,2011–2012),theδ18OSW-anomalywaslowinsummer.
UpwellingeventsintheGulfofOmanaredrivenbytheSWMonsoon,whichcausesstrongseasonalwindsparalleltothecoastofSouthernOmanintheArabianSea,whiletheassociatedEkmantransportcreatesstrongupwellingalongthecoastalmargins,bringingcold,nutrient-richwatertothesurface1,4.
ThisupwelledwaterhasindirectimpactsonFigure2.
(a)Cross-plotofinsitulow-SSTperiodsandAN-δ13Cminima.
Thedottedlineindicatestheregressionbetweeninsitulow-SSTperiodsandAN-δ13Cminima(filledcircles)andestimatedδ13Canomalyvaluefornoupwellingperiods(r=0.
937,P<0.
05,n=6).
Theδ13Canomalyvaluefornoupwelling(opencircle)wasestimatedfrominsituδ13CDIC-SWfromtheArabianSea36andthevalueofδ13Cinisotopicequilibriumbetweencoralcarbonateandseawater37(b)EstimatedupwellingperiodsfromAN-δ13Cminima.
Blackbarsindicatetheyearswhichwereusedfortheregressionbetweeninsitulow-SSTperiodsandδ13Canomalyminima.
5coralsandreefareasfarthernorththroughgyresandeddysystemsthatsweepintotheOmanSea1,4.
Inaddition,upwellingmaybeinfluencedbyverticalseawaterdensity,dependingonSSTandSSS38.
Theδ18OSW-anomalyrecordsuggestedthatdeepseawaterdidnotreachtheseasurfaceaslow-densitywatermassesmightformacapontheseasurfaceintheGulfofOman.
ObservationssuggestthattheprimaryproductivityoftheGulfofOmanissubjecttointer-annualvariability1,butlong-termobservationalrecordsarelacking.
Ournewδ13CcoralrecordcapturedpastupwellingeventsandtheirperiodsintheGulfofOmanfor26years.
Thus,coralskeletalarchivesfillanimportantgapintheobser-vationalrecordandhavegreatpotentialforincreasingourunderstandingoftheupwellingmechanismsintheGulfofOman.
Moreover,itispossibletoreconstructpastSST,SSSandupwellingfrequency/intensityduringtheHolocene(from0to10ka)byapplyingthesamemethodstofossilcoralsfromtheArabianPeninsula.
MethodsCoralsampling.
OnFebruary23,2013,wedrilledaPoritessp.
coralcolonyintheGulfofOman(23°30′N,58°45′E:Fig.
3aandb).
ThisPoritescolonywaslivingata2mwaterdepthinasmallbay(BandarKhayran)southofMuscat.
Therewasnodry-riverbed(locallyname:wadi)nearby;thus,weexcludedtheinfluenceofoccasionalplumesoffreshwaterfromcoastalrunoffatthesite.
Intotal,thecoralcorewas71cmlong.
Onthesamplingdate,wemeasuredinsituSSTandSSSat24.
3°Cand38.
2PSU(practicalsalinityunit).
MeteorologicalrecordsfromtheweatherstationatSeebAirport(23.
60°N,58.
30°E)showedlowprecipitationrates,withlessthan14.
0mm/month(themonthlyaverageprecipitationclimatologyforthepast23yearswas0.
28–14.
0mm/month;GHCN-Monthlyver.
2).
Forcoralproxycalibration,weusedAdvancedVeryHighResolutionRadiometer(AVHRR)satelliteSSTdata,SODAsatelliteSSSdata(http://iri.
columbia.
edu:Fig.
3c)andOLRdata(https://climexp.
knmi.
nl/:Fig.
3c)39–41.
Salinityrecordsdecreaseinsummersuggestingapossibleoccurrenceofupwellingevents.
Subsampling.
Thecoralcorewasslicedinto5-mm-thickslabs.
WetookX-radiographsofthecoralslabstoidentifythecoralgrowthaxis(Fig.
4).
Wepreparedledgesof1.
5mminthicknessalongthemaximumgrowthaxisandobtainedcoralpowderataresolutionof0.
5mmforgeochemicalanalysis.
Oxygenandcarbonisotopemeasurements.
Thecoralpowderwasweighed,and100μg(±20μg)weretakenforoxygenandcarbonstableisotopeanalysis.
Thesamplepowderwasreactedwith100%H3PO4at70°Cinanautomatedcarbonatepreparationdevice(KielII).
Theδ13Ccoralandδ18OcoralwereanalyzedwithaFinniganMAT251stableisotoperatiomassspectrometersysteminstalledatHokkaidoUniversity.
Analyticalerrorsforδ13Ccoralandδ18Ocoralweredeterminedtobe0.
08and0.
07‰,respectively,basedonreplicatemeasurementsoftheNBS-19standard(1σ,n=40).
Figure3.
(a),(b)MapofthesamplingsiteintheGulfofOman.
ThefiguresweregeneratedusingGenericMappingTools(GMTver.
4.
5.
12)44.
(c)ClimatologicaldataestimatedfrombiweeklySSTfrom1987to2013(datafromAVHRR39),monthlySSSfrom1987to2008(datafromSODA40)andOLR41duringpast26years(datafromhttps://climexp.
knmi.
nl/).
Errorbarsindicateclimatologydeviation(1σ).
6Traceelementmeasurements.
WemeasuredSr/CaratioswithaSPECTROCIROSCCDSOPinduc-tivelycoupledplasmaopticalemissionspectrophotometerinstalledatKielUniversityfollowingacombinationofmethodsdescribedbySchrag42anddeVilliersetal.
43.
Approximately250μgofcoralpowderwasdissolvedin4mLofHNO3.
Thesamplesolutionforthemeasurementoftraceelementswaspreparedviaserialdilutionwith2%HNO3foraCaconcentrationofca.
8ppm.
AnalyticalprecisionoftheSr/Cadeterminationswas0.
07%RSDor0.
01mmol*mol1(1σ).
Dataanalysis.
WeusedthecoralSr/Caratiostodevelopanagemodelforallproxies.
MinimaandmaximaofthecoralSr/CaratioswerechosenasanchorpointsandtiedtothemaximaandminimaofSST,respectively.
Toobtainatimeserieswithequidistanttimesteps,weresampledtheproxydataatabiweeklyresolutionusingtheAnalySeriessoftware,version2.
0.
827.
Annualextensionrateswereestimatedfromthedistance(inmm)betweenthewinteranchorpointsineachsclerochronologicalyear.
References1.
Al-Azri,A.
R.
,Piontkovski,S.
A.
,Al-Hashmi,K.
A.
,Goes,J.
I.
&DoGomes,H.
R.
ChlorophyllaasameasureofseasonalcouplingbetweenphytoplanktonandthemonsoonperiodsintheGulfofOman.
Aquat.
Ecol.
44,449–461(2010).
2.
Qasim,S.
Z.
OceanographyofthenorthernArabianSea.
DeepSeaRes.
PartA.
Oceanogr.
Res.
Pap.
29,1041–1068(1982).
3.
Burt,J.
A.
etal.
Oman'scoralreefs:Auniqueecosystemchallengedbynaturalandman-relatedstressesandinneedofconservation.
Mar.
Pollut.
Bull.
105,498–506(2016).
4.
Wiggert,J.
D.
,Hood,R.
R.
,Banse,K.
&Kindle,J.
C.
Monsoon-drivenbiogeochemicalprocessesintheArabianSea.
Prog.
Oceanogr.
65,176–213(2005).
5.
Elliott,A.
J.
&Savidge,G.
SomefeaturesoftheupwellingoffOman.
J.
Mar.
Res.
48,319–333(1990).
6.
Coles,S.
L.
ReefcoralsoccurringinahighlyfluctuatingtemperatureenvironmentatFahalIsland,GulfofOman(IndianOcean).
CoralReefs16,269–272(1997).
7.
Al-Azri,A.
R.
etal.
MesoscaleandNutrientConditionsAssociatedwiththeMassive2008CochlodiniumpolykrikoidesBloomintheSeaofOman/ArabianGulf.
EstuariesandCoasts37,325–338(2013).
8.
Cobb,K.
M.
,Charles,C.
D.
,Cheng,H.
&Edwards,R.
L.
ElNio/SouthernOscillationandtropicalPacificclimateduringthelastmillennium.
Nature424,271–276(2003).
9.
Watanabe,T.
etal.
PermanentElNioduringthePliocenewarmperiodnotsupportedbycoralevidence.
Nature471,209–211(2011).
10.
Felis,T.
,Ptzold,J.
,Loya,Y.
&Wefer,G.
VerticalwatermassmixingandplanktonbloomsrecordedinskeletalstablecarbonisotopesofaRedSeacoral.
J.
Geophys.
Res.
103,30731(1998).
11.
Tudhope,A.
W.
,Lea,D.
W.
,Shimmield,G.
B.
,Chilcott,C.
P.
&Head,S.
MonsoonClimateandArabianSeaCoastalUpwellingRecordedinMassiveCoralsfromSouthernOman.
Palaios11,347(1996).
12.
Gagan,M.
K.
etal.
TemperatureandSurface-OceanWaterBalanceoftheMid-HoloceneTropicalWestern.
Pacific.
Science279,1014–1018(1998).
13.
McConnaughey,T.
13Cand18Oisotopicdisequilibriuminbiologicalcarbonates:I.
Patterns.
Geochim.
Cosmochim.
Acta53,151–162(1988).
14.
McConnaughey,T.
13Cand18Oisotopicdisequilibriuminbiologicalcarbonates:II.
Invitrosimulationofkineticisotopeeffects.
Geochim.
Cosmochim.
Acta53,163–171(1989).
15.
Swart,P.
K.
,Leder,J.
J.
,Szmant,A.
M.
&Dodge,R.
E.
Theoriginofvariationsintheisotopicrecordofscleractiniancorals:II.
Carbon.
Geochim.
Cosmochim.
Acta60,2871–2885(1996).
16.
Abram,N.
J.
,Gagan,M.
K.
,McCulloch,M.
T.
,Chappell,J.
&Hantoro,W.
S.
Coralreefdeathduringthe1997IndianOceanDipolelinkedtoIndonesianwildfires.
Science301,952–955(2003).
17.
Fairbanks,R.
G.
&Dodge,R.
E.
Annualperiodicityofthe18O/16Oand13C/12CratiosinthecoralMontastreaannularis.
Geochim.
Cosmochim.
Acta43,1009–1020(1979).
18.
Grottoli,A.
Effectoflightandbrineshrimponskeletalδ13CintheHawaiiancoralPoritescompressa:atankexperiment.
Geochim.
Cosmochim.
Acta66,1955–1967(2002).
19.
Ptzold,J.
GrowthrhythmsrecordedinstableisotopesanddensitybandsinthereefcoralPoriteslobata(Cebu,Philippines).
CoralReefs3,87–90(1984).
20.
Wellington,G.
M.
&Dunbar,R.
B.
StableisotopicsignatureofElNioSouthernOscillationeventsineasterntropicalpacificreefcorals.
CoralReefs14,5–25(1995).
21.
Nurhati,I.
S.
,Cobb,K.
M.
&DiLorenzo,E.
Decadal-scaleSSTandsalinityvariationsinthecentraltropicalpacific:Signaturesofnaturalandanthropogenicclimatechange.
J.
Clim.
24,3294–3308(2011).
22.
Juillet-Leclerc,A.
&Schmidt,G.
Acalibrationoftheoxygenisotopepaleothermometerofcoralaragonitefromporites.
Geophys.
Res.
Lett.
28,4135–4138(2001).
23.
Hartmann,A.
C.
,Carilli,J.
E.
,Norris,R.
D.
,Charles,C.
D.
&Deheyn,D.
D.
StableisotopicrecordsofbleachingandendolithicalgaebloomsintheskeletonoftheboulderformingcoralMontastraeafaveolata.
CoralReefs29,1079–1089(2010).
24.
Nozaki,Y.
,Rye,D.
M.
,Turekian,K.
K.
&Dodge,R.
E.
A200yearrecordofcarbon-13andcarbon-14variationsinaBermudacoral.
Geophys.
Res.
Lett.
5,825–828(1978).
Figure4.
X-radiographofcoralcoreOMN130221(Poritessp.
).
Thewhitelineindicatesthemeasurementlines.
Severaloverlappingmeasurementlinesweresampledtoensuringreproducibility.
725.
Al-Rousan,S.
&Felis,T.
Long-termvariabilityinthestablecarbonisotopiccompositionofPoritescoralsatthenorthernGulfofAqaba,RedSea.
Palaeogeogr.
Palaeoclimatol.
Palaeoecol381–382,1–14(2013).
26.
Klein,R.
,Ptzold,J.
,Wefer,G.
&Loya,Y.
SeasonalvariationsinthestableisotopiccompositionandtheskeletaldensitypatternofthecoralPoriteslobata(GulfofEilat,RedSea).
Mar.
Biol.
112,259–263(1992).
27.
Paillard,D.
,Labeyrie,L.
&Yiou,P.
MacintoshProgramperformstime-seriesanalysis.
Eos,Trans.
Am.
Geophys.
Union77,379(1996).
28.
Trenberth,K.
E.
&Dai,A.
EffectsofMountPinatubovolcaniceruptiononthehydrologicalcycleasananalogofgeoengineering.
Geophys.
Res.
Lett.
34,1–5(2007).
29.
Glynn,P.
W.
MonsoonalupwellingandepisodicAcanthasterpredationasprobablecontrolsofcoralreefdistributionandcommunitystructureinOman,IndianOcean.
AtollResearchBulletin379,379(1993).
30.
Claereboudt,M.
R.
ReefcoralsandcoralreefsoftheGulfofOman.
HistoricalAssociationofOman,322(2006).
31.
Quinn,N.
J.
&Johnson,D.
W.
ColdwaterupwellingscoverGulfofOmancoralreefs.
CoralReefs-JournaloftheInternationalSocietyforReefStudies15.
4,214(1996).
32.
Asia-PacificData-ResearchCenter.
Availableat:http://apdrc.
soest.
hawaii.
edu(Accessed:March2017).
33.
Yamazaki,A.
etal.
ReconstructingpalaeoenvironmentsoftemperateregionsbasedonhighlatitudecoralsatTatsukushiBayinJapan.
J.
JapaneseCoralReefSoc11,91–107(2009).
34.
Kroopnick,P.
M.
Thedistributionof13CofΣCO2intheworldoceans.
DeepSeaRes.
PartA.
Oceanogr.
Res.
Pap.
32,57–84(1985).
35.
Grottoli,A.
G.
&Wellington,G.
M.
Effectoflightandzooplanktononskeletalδ13CvaluesintheeasternPacificcoralsPavonaclavusandPavonagigantea.
CoralReefs18,29–41(1999).
36.
Peeters,F.
TheeffectofupwellingonthedistributionandstableisotopecompositionofGlobigerinabulloidesandGlobigerinoidesruber(plankticforaminifera)inmodernsurfacewatersoftheNWArabianSea.
Glob.
Planet.
Change34,269–291(2002).
37.
Romanek,C.
S.
,Grossman,E.
L.
&Morse,J.
W.
Carbonisotopicfractionationinsyntheticaragoniteandcalcite:Effectsoftemperatureandprecipitationrate.
Geochim.
Cosmochim.
Acta56,419–430(1992).
38.
Kumar,S.
P.
&Prasad,T.
G.
FormationandspreadingofArabianSeahigh-salinitywatermass.
J.
Geophys.
Res.
104,1455(1999).
39.
Reynolds,R.
W.
etal.
Dailyhigh-resolution-blendedanalysesforseasurfacetemperature.
J.
Clim.
20,5473–5496(2007).
40.
Giese,B.
S.
&Carton,J.
A.
AReanalysisofOceanClimateUsingSimpleOceanDataAssimilation(SODA).
Mon.
WeatherRev.
136,2999–3017(2008).
41.
Liebmann,B.
&Smith,C.
A.
Descriptionofacomplete(interpolated)outgoinglongwaveradiationdataset.
Bull.
Am.
Meteorol.
Soc.
77,1275–1277(1996).
42.
Schrag,D.
P.
Rapidanalysisofhigh-precisionSr/Caratiosincoralsandothermarinecarbonates.
Paleoceanography14,97–102(1999).
43.
deVilliers,S.
AnintensityratiocalibrationmethodfortheaccuratedeterminationofMg/CaandSr/CaofmarinecarbonatesbyICP-AES.
GeochemistryGeophys.
Geosystems3(2002).
44.
Wessel,P.
&Smith,W.
H.
F.
FreeSoftwareHelpsMapandDisplayData.
EosTrans.
AGU.
72(41),441–446(1991).
AcknowledgementsWeacknowledgeC.
A.
Grove,H.
TakayanagiandK.
Ohmorifortheirhelpwithcoralcore-drillingandfieldworkattheSultanateofOman.
K.
BremersupportedICP-OESanalysis.
CREESmembersatHokkaidoUniversityprovidedassistancewithslicingthecoralcore.
ThisworkwassupportedbyJSPSKAKENHIGrantNumberJP25257207.
AuthorContributionsT.
W.
andM.
P.
designedtheproject.
T.
W.
,A.
Y.
,M.
P.
M.
R.
C.
collectedsamples.
T.
K.
W.
,A.
Y.
andD.
G.
-S.
analyzedthesamples.
T.
K.
W.
wrotethemanuscript.
Allauthorshelpedwiththeinterpretationofthedataandwritingthemanuscript.
AdditionalInformationSupplementaryinformationaccompaniesthispaperatdoi:10.
1038/s41598-017-04865-5CompetingInterests:Theauthorsdeclarethattheyhavenocompetinginterests.
Publisher'snote:SpringerNatureremainsneutralwithregardtojurisdictionalclaimsinpublishedmapsandinstitutionalaffiliations.
OpenAccessThisarticleislicensedunderaCreativeCommonsAttribution4.
0InternationalLicense,whichpermitsuse,sharing,adaptation,distributionandreproductioninanymediumorformat,aslongasyougiveappropriatecredittotheoriginalauthor(s)andthesource,providealinktotheCre-ativeCommonslicense,andindicateifchangesweremade.
Theimagesorotherthirdpartymaterialinthisarticleareincludedinthearticle'sCreativeCommonslicense,unlessindicatedotherwiseinacreditlinetothematerial.
Ifmaterialisnotincludedinthearticle'sCreativeCommonslicenseandyourintendeduseisnotper-mittedbystatutoryregulationorexceedsthepermitteduse,youwillneedtoobtainpermissiondirectlyfromthecopyrightholder.
Toviewacopyofthislicense,visithttp://creativecommons.
org/licenses/by/4.
0/.
TheAuthor(s)2017