bachelor thesis · 2018-06-08 · fanny axelsson bachelor thesis geochemistry 30 hp 5 the...

Degree Project in Geochemistry 30 hp

Bachelor Thesis

Stockholm 2018

Department of Geological SciencesStockholm UniversitySE-106 91 Stockholm

An investigation of methane production in oxygenated surface water during an algal bloom in the Askö area,

northern Baltic proper

Fanny Axelsson

FannyAxelsson Bachelorthesisgeochemistry30hp

AbstractMethane concentration maxima in surface waters have been observed in several studies, whichcontradictstheunderstandingofmethaneproductionoccurringstrictlyinanoxicenvironments.Varioushypotheses of production pathways of the gas in that zone have been presented, butwith furtherresearchneededtobeconductedinthesubject.Inthisbachelorthesis,itisinvestigatedifindicationsareseenofmethaneproductionintheoxygenatedsurfacewaterintheareaofAskö,northernBalticproper,andifthepossibleproductioniseffectedbythesummeralgalbloom.Contributiontowater-air-fluxes fromthepossibleproductionand indicationsofproductionpathway(s)arealsoexamined.Watersamples(profilesandsurfacewaters),incubationexperiments,water-air-fluxmeasurementsandqualitativeanalysesofphytoplanktongenerawereconductedbefore,duringandafterthealgalbloomintheAsköarea.Concentrationsofmethane,carbondioxideandnitrousoxidewereanalysedaswellastheisotopecompositionofthemethane.Thecollecteddatashowsofhighconcentrationsofmethaneinsurfacewaterswithrespecttotheatmosphereandwithincreasingmethaneduringthealgalbloom:rangebetween~20-30nMatdeeperareasand~140-150nMatshallowerareas.Aconcentrationmaxima within the thermocline was found in one of the sites in June, otherwise, increasingconcentrations towards theseafloorwereseen.The isotopecomposition insurfacewaterswas 13C-depletedwhen compared tomethane in the atmosphere, with a range between ~ -60‰ to -55‰throughoutthesummer.Thewaterprofilesalsoshowedofmoredepletedmethaneinsurfacewatersthaninbottomwatersinmostofthemonths.Themethanewater-air-fluxeswerehighestinAugustatthesouthernpartofAskö,withfluxesupto5.0±0.4mgm-2day-1inshallowwaterareas.Darkincubationexperiments (5weeks long)didnotprovideanysignsofmethaneproduction.Conclusions fromthisinvestigationisthatit isbelievedthataverticalupwardtransportofdiffusinggasfromthesedimentwasnottheonlysourceofmethanetothesurfacewater.Twoothermethanesourcesaresuggested,i)ebullitionof13C-depletedmethanefromthesedimentwithbubblesreachinguptothemixedlayer:insituatthenorthernsideofAsköaswellasatshallowerareasatthesouthernsidewithfurtherlateraltransporttoopencoastalwaters,ii)methaneproductioninthephoticzoneofthesurfacewateratthesouthernsideofAskö.FurtherresearchisneededtobeconductedintheAsköareatoprovidemoreinsightsofthesehypotheses:lightincubationexperimentsofsurfacewater,dataofwatercurrentsinthearea,isotopecompositionofmethanefromdiffusionandebullitionfromthesedimentsaswellasgeophysicalacousticdatafromebullitionsites.Bachelorthesisingeochemistry30hpFannyAxelssonSupervisorsVolkerBrüchertattheDepartmentofGeologicalSciencesatStockholmUniversity&HellePlougattheDepartmentofMarineSciencesattheUniversityofGothenburg


2

Tableofcontent1.Introduction...............................................................................................................................41.1Aim&hypothesis...................................................................................................................................41.2Background............................................................................................................................................4

1.2.1Watercolumnprofiles....................................................................................................................41.2.2Methaneproductionpathways......................................................................................................51.2.3Methanewater-air-fluxes..............................................................................................................51.2.4AlgalbloomintheBalticSea:phytoplanktongenera...................................................................61.2.5Researchofmethaneproductioninoxygenatedsurfacewaters..................................................6

2.Method......................................................................................................................................72.1Studysite...............................................................................................................................................72.2Fieldmethods........................................................................................................................................8

2.2.1Watersamples:waterprofiles,surfacewater,incubationexperimentandmicroscopesamples.................................................................................................................................................................82.2.2CTD:oxygen,salinity,temperature,photosyntheticallyactiveradiationandturbidity...............92.2.3Manualfluxchambers.................................................................................................................102.2.4LosGatosResearchInfraredSpectrometer:fluxmeasurementsandatmospherictransects....10

2.3Laboratory&dataanalysis..................................................................................................................112.3.1Concentrationanalysis:watersamples,incubationexperiments&airsamples........................112.3.2Isotopeanalysis:watersamples,incubationexperiments&airsamples...................................132.3.3Dataanalysisofwater-air-fluxes:manualfluxchambers&LGR................................................132.3.4Qualitativeanalysisofphytoplanktongenera.............................................................................15

3.Results......................................................................................................................................163.1Waterconcentrationsandisotopecomposition:waterprofilesandsurfacewaters.......................16

3.1.1Waterprofile:B1..........................................................................................................................163.1.2Waterprofile:Fifångsdjupet........................................................................................................193.1.3Surfacewater:JulaftonsFyr&chambersites.............................................................................21

3.2Incubationexperiment........................................................................................................................233.3Water-air-fluxes...................................................................................................................................24

3.3.1Methanewater-air-fluxes&isotopecomposition.......................................................................243.3.2Carbondioxidewater-air-fluxes...................................................................................................25

3.4Atmospherictransects.........................................................................................................................263.5Qualitativeanalysisofphytoplanktongenerainthewater...............................................................27

4.Discussion................................................................................................................................304.1Hypothesis1:lateralinputofisotopicallyheavybottomwaters.......................................................314.2Hypothesis2:methaneebullition.......................................................................................................32

4.2.1EbullitioninsituatFifångsdjupet................................................................................................324.2.2EbullitionatshallowerareaswithlateralinputofsurfacewaterstoB1....................................33

4.3Hypothesis3:methaneproductionintheoxygenatedsurfacewater...............................................364.4Finalcomments...................................................................................................................................37

5.Conclusions..............................................................................................................................37


3

Acknowledgements.....................................................................................................................37

References...................................................................................................................................38

Appendix1...................................................................................................................................41A1.1.Resultstwo-sidedpairedt-testoftheincubationexperimentsamples.........................................41....................................................................................................................................................................41A1.2Incubationexperiment:method.......................................................................................................41A1.3Incubationexperiment:results.........................................................................................................42

Appendix2...................................................................................................................................45

Appendix3...................................................................................................................................46

Appendix4...................................................................................................................................47

Appendix5...................................................................................................................................51A5.1Hypothesis2.1:ebullitioninsituatFifångsdjupet...........................................................................51A5.2Calculationinhypothesis2.2:ebullitionatshallowerareaswithlateralinputofsurfacewaterstoB1...............................................................................................................................................................52A5.3Fluxvs.daytime/nighttimeinHypothesis2.2:ebullitionatshallowerareaswithlateralinputofsurfacewaterstoB1..................................................................................................................................52A5.4Leakagesuspicion..............................................................................................................................53

Appendix6...................................................................................................................................55


4

1.Introduction Methane (CH4)asanatmosphericgreenhousegashasacontributingroleintheresearchaboutclimatechange.Theglobalannualmeanoftheatmospheric concentration of CH4 hasincreased since the beginning of the pre-industrial time from 0.722 ppm in year 1750(IPCC,2013)to1.89ppmin2011(Dlugokencky,2017). Sources of CH4 to the atmosphere arefrom either anthropogenic sources or naturalsources, where natural sources includeemissionsfromwetlands,termites,plants,lakesand oceans (Bousquet et al., 2006). Eventhough there has been extensive researchaboutmethaneanditsemissionsourcestotheatmosphere,theannualglobalCH4budgetwithsources and sinks is not complete whereobserved long-term variations in atmosphericconcentrationisnotyetexplained(IPCC,2013).An on-going debate that could be of use tocreate further insights of these observedvariations regards CH4 production inoxygenated surface waters. Dissolved CH4 inoceans and freshwater lakes is producedthroughbiologicalactivity.Theproductionhaslong been believed to strictly occur in anoxicenvironments by methanogenic archaea(methanogens) during anaerobic respiration(Ferry, 1993a). Although, concentrationsmaxima of CH4 in oxygenated surface watershas regularly been observed. ProductionprocessesforCH4intheoxygenatedzoneisnotyet understood. Different theories have beenpresented,butnoagreementhasbeenmade.One suggestion is interaction betweenmethanogenswitheitherzooplankton(Angelisand Lee, 1995; Karl and Tilbrook, 1994) orphytoplankton (Grossart et al., 2011; Tang etal.,2014),whileothertheoriesregardstheroleofnutrient limitationsinthewater(Karletal.,2008; Damm et al., 2010). Several of theobservations of high CH4 concentration insurfacewatershavebeenmade in freshwaterlakes and in high latitude oceanswaters. It istherefore of interest to investigate if similarobservationscanbemadeinthebrackishwater

of the Baltic Sea. For this bachelor thesis,furtherexaminationsaremadeconcerningthetheory of CH4 production due to interactionbetweenmethanogensandphytoplankton.

1.1Aim&hypothesis

Theaimofthisbachelorthesisistoinvestigateifthereareindicationsofmethaneproductionin oxygenated surface waters due tomethanogens-phytoplanktoninteractionandtoanalyse if this production of CH4 significantlycontributes to a water-air-flux. It is also ofinteresttosearchforindicationsofproductionpathway(s) of CH4 in the waters of the studyarea. To meet the aims, concentration andisotope composition of CH4 in water columnswereanalysedandamountofwater-air-fluxeswere measured before, during and after thesummeralgalbloomintheareaofAskö,southof Stockholm. Light and dark incubationexperiments were conducted to investigate iftheCH4concentrationandisotopecompositionchangedduringincubation.Dominatinggeneraof phytoplankton in the surface waters werealsoexamined. The working hypothesis is that there exists aCH4 production in the surface waters in thestudyareathatsignificantlycontributestoaCH4water-air-flux. The production pathway isbelieved to be through hydrogenotrophicmethanogenesis,wheremethanogensusesH2produced by cyanobacteria. Therefore, thehypothesis is that themethane concentrationshould increase during the summer algalblooms.

1.2Background

1.2.1Watercolumnprofiles

Profiles of CH4 concentrations throughout awater column usually show lowerconcentrations in the oxygenated surfacewaters,withincreasingconcentrationstowardsbottomwaters(Whiticar,1999).Whatgoverns


5

thedistributionofCH4 in thewater column isthe CH4 production by methanogens in theanoxic sediments, which provides a CH4 fluxfrom the sediment to the water either bydiffusionorbyebullition.WhenCH4entersthewater column, a vertical transport of CH4diffuses the gas upwards. Although, thistransport is limited in highly stratified watersand due to aerobic and anaerobic methaneoxidation by organisms in the water, theconcentration decreases towards the surface(Reeburgh and Heggie, 1977). Similar waterprofileshavebeenobserved in theBaltic Sea,withsurfaceconcentrationsofCH4<8nMandbottomwaters>80nM(Gülzowetal.,2013).Ithas also been observed surface waterssupersaturatedwithCH4,leadingtoconclusionsthat the Baltic Sea is a source of CH4 to theatmosphere(Bangeetal.,1994;Gülzowetal.,2013;Schmaleetal.,2010).RegardingtheisotopecompositionofCH4 inawatercolumn,usuallyheaviercompositionsarefoundinthesurfacewatercomparedtointhebottomwaters.Thisisduetothatproductioninthe sediment provides 13C-depleted CH4 andwhendiffusingintothewater,oxidationofthegasenrichestheCH4withδ13C(Whiticar,1999).Dissolved CH4 in the surface water inequilibrium with the atmosphere providessimilarvaluesasCH4inthegasphase(Knoxet.al.,1992).

1.2.2Methaneproductionpathways

Methanogenesis is the production of CH4 bymethanogens during oxidation of carbon-containing components. There are differentpathwaysfortheCH4production,meaningthatmethanogens use different compounds assource of carbon and source of energy toproduce methane. The production pathwaysare hydrogenotrophicmethanogenesis (eq. 1)(Mahetal.,1977),acetoclasticmethanogenesis(eq. 2) (Gelwicks et al., 1994) andmethylotrophic methanogenesis (eq. 3)(Deppenmeieretal.,1996;Pengeretal.,2012):4H2(aq)+CO2→CH4(aq)+2H2O

(eq.1)

CH3COOH→CH4(aq)+CO2(aq) (eq.2)

4CH3OH→3CH4(aq)+CO2 aq +2H2O (eq.3)

In the hydrogenotrophic methanogenesis,hydrogen (H2) and carbon dioxide (CO2) areusedintheredoxreactiontoproduceCH4andwater (H2O) where H2 is oxidised and CO2reduced.ForthecaseofCH4productioninthesurfacewaters,H2isproducedduringfixationofnitrogen (N2) by heterocystous cyanobacteria(Bothe et al., 2010). Incubation experimentsconductedbyBerget.al(2014)givessupporttothe hypothesis thatmethanogens are able touse the produced H2 from cyanobacteria tofurtherproduceCH4.In acetoclasticmethanogenesis, methanogensuse acetate (CH3COOH) to produce CH4 andCO2,wheretheacetateisproducedbybacteriaduring fermentation of organic matter. Thereaction isa redox reactionwhere themethylgroup(CH3)isbeingreduced,whilethecarbonylgroup(COOH)isbeingoxidised(Ferry,1993b).In the CH4 production done throughmethylotrophicmethanogenesis,themethano-gens use methyl compounds, as for examplemethanol (CH3OH). The carbon in the methylcompoundformingCH4isbeingreduced,whilethe carbon forming CO2 is being oxidised(Deppenmeieretal.,1996).The isotopic signature from each pathway isdifficult to determine accurately (Conrad,2005). However, studies have shown thatmethylotrophicmethanogenesisisthepathwaythat fractionates the most, while theacetoclasticmethanogenesistheleast(Conrad,2005;Whiticaretal.,1986).

1.2.3Methanewater-air-fluxes

The oceans contain dissolved gases whosesolubilitydependsontemperature,salinityandpressure(Yamamotoetal.,1976).Therefore,ina specific environmental condition a specificamountofgascanbedissolved.Thedissolvedgasinthewaterstrivestoreachanequilibriumwiththegasintheatmosphere.Thiscreatesa


6

water-air-flux,wheregasiseitheremittedfromthewater into the air, and vice versa.With apositivewater-air-flux, gas is emitted into theair while with a negative flux, the opposite ishappening. Other factors that affects the fluxarewind speed,winddirection, surfactants inthewatersurface,amountofrainandthedropsize of the rain, breaking waves and bubbles.What is incommonfor these factors is that itprovides kinetic energy to the water surfaceand with higher kinetic energy, the fluxincreases (Ho et al., 1997;Wanninkhof et al.,2009).

1.2.4 Algal bloom in the Baltic Sea:phytoplanktongenera

TheyearlyalgalbloomintheBalticSeaoccursusuallyinJuly-August,withwarmtemperaturesand low amount of wind promoting a richerbloom. Therefore, the intensity of the algalbloomvariesfromyeartoyearwithspatialandtimelyvariation (Hansson,2006).Thesummeralgal bloom consists of phytoplanktoncyanobacteria. The capability of thecyanobacteria for nitrogen fixation provides agreat advantage when organic nitrogen isdepleted during parts of the summer,comparedtootherphytoplanktonthatcannotachieve this (Klawonn et al., 2016). Themostcommoncyanobacteriageneraduringthealgalbloom in the area around Askö areAphanizomenon spp., Nodularia spp.,Dolichospermumspp.andLyngbyaspp(Staletal., 2003; Wasmund and Uhlig, 2003). Asmentionedinsection1.2.3Methaneproductionpathways,nitrogenfixationcanbecoupledwithhydrogenotrophicmethanogenesis. 1.2.5 Research of methane production inoxygenatedsurfacewaters

Several research studies have presentedobservations of high CH4 concentration inoxygenated surface waters. The observationshavebeenmadeinfreshwaterlakesorinhighlatitudeoceanwaters,ofteninconjunctionwiththe thermocline or with an oxygenconcentration maxima. High methane water-

air-fluxes have in also been observed in thesame studies. Different theories of theproduction processes of CH4 have beenpresented. Already in 1977, Scranton et. al.presentedobservationsofthisphenomenaandsuggestedaninsitubiologicalpathwayforthemethaneproduction. Lateron, in1994Karl&Tillbrookobservedmethanesupersaturationinthe surface waters with respect to theatmosphere in the North Pacific Ocean. Thestudypresentedahypothesisthatthemethaneproductionoccurredinthegutsofzooplankton,which later was released out to the surfacewatersbythefaecalpelletsofthezooplankton.The zone of where the CH4 concentrationmaxima was found in that study was in thezooplankton down- and upward migrationzone. Another hypothesis that has beensuggested is methylotrophic methanogenesiswithmethylatedcompoundsfromdegradationofMPn(methylphosphonate) inphosphorous-limited waters (Karl et al., 2008) or fromdegradation of DMSP (dimethylsulfonio-propionate) innitrogen-limitedwaters (Dammet. al., 2010). Although,Grossart et al. (2011)andTanget.al.(2014)arguedagainstthisandsuggested that the methane production isperformed by oxygen tolerant methanogensthrough either hydrogenotrophic or aceto-clasticmethanogenesis.Besides CH4 production, another source ofmethaneintotheupperwatercolumnisfromebullition fromthesediment.Thegasbubbleseitherdissolveinthewatercolumnduringtheascent or reach the water surface withoutdissolving, leadingtodirectemissions intotheatmosphere (Schmale et. al., 2005: SchneidervonDeimlingetal.,2011).


7

2.Method Measurements and sampling methodsconducted in the fieldatAskö in June,Augustand September were i) water sampling, ii)water-air-flux measurements with manualchambers andwith an infrared spectrometer,iii) CTD measurements and iv) transectmeasurements of atmospheric concentrationof gases. Water sampling were madethroughout water profiles and in surfacewaters,where somesampleswereused inanincubation experiment. All samples wereanalysedforconcentrationofCH4,CO2andN2Oas well as for isotope composition of δ13CCH4.Some water samples were collected with aplanktonnettobeusedinmicroscopeanalysis.To provide a schematic overview of themethods, figure1demonstrateseachmethodconducted.

2.1Studysite

ThestudyforthisthesishasbeenconductedintheareaoftheislandAsköatAsköLaboratory(later on referred to as the station). Askö islocated south of Stockholm in the northern

Baltic proper in the Baltic Sea (figure 2). Thesurface water in the area has a salinity ofapproximately 6 - 8‰ (Björck, 1995) and ischaracterisedbystratifiedwatercolumnswithdistinctboundary layers fromthe thermoclineand the halocline. There are seasonaldifferences in the stratification, wheremixingwithinthewatercolumnislowerduringspringandsummerthanduringwinterandfall(Axell,1998). Stratified waters together with a longresidence time of the water in the Baltic seacauses seasonally oxygen deficiency in thebottomwaters(Carstensenetal.,2014).For this study, four sites around the area ofAskö were chosen for sampling andmeasurements:B1,Fifångsdjupet,JulaftonsFyrandAskö.ThemainfocushasbeenonB1andFifångsdjupet where water profiles sampleswere collected and water-air-fluxmeasurementsconducted in June,AugustandSeptember.Anextrasiteforfluxmeasurementwith the infrared spectrometerwas chosen inJune as well: Tallholmen (58°51'09.66"N,17°39'25.51"E). No further measurements orsampling were made in that site, though.Overviewofwhichmethods used in each sitearegivenintable1.The sedimentatall sitesexceptat chamber1consists of postglacial clay, gyttja clay and/orclaygyttja,whilethesedimentatchamber1is

Figure1.Aschematicoverviewofthemethodsusedtomeettheaimsofthisbachelorthesis.


8

composed of postglacial fine-grained sand(SGU, 2015). In the vicinity of Tallholmen,unpublished data shows that methaneebullitionfromthesedimentoccurs,leadingtomethane release into the water column (E.Weidner, personal communication). It wasthereforeofinteresttochooseFifångsdjupetasamainsite,asitewith24mwaterdepthstotheseafloor. The site at B1 (40 m depth to theseafloor)waschosenbecauseitwaslocatedinopencoastalwateroftheBalticSea,comparedto Fifångsdjupet that is located in a moreenclosed basin north of the island of Askö(figure2).Itwasofinteresttoseeifdifferenceswouldbeseenbetween thesitesdue to theirlocations.JulaftonsFyr(waterdepth10m)waschosenasasamplingsitebecauseduringfieldsamplinginJuneanomalouslyhighatmosphericconcentrationofCH4weremeasured.Thesiteof Askö was chosen for manual fluxmeasurements (depth chamber 1 ~10 m,chamber2~15m)forexaminationoffluxesinshallowareas.

2.2Fieldmethods

2.2.1 Water samples: water profiles,surfacewater, incubationexperimentandmicroscopesamples

WatersamplesthroughoutthewatercolumnatB1 and Fifångsdjupet and from the watersurfaceatJulaftonsFyrwerecollectedusinga5LNiskinbottle.Thewaterwaspouredthrougha pvc tube into 12 mL exetainers forconcentration analysis and into 110 mL glassbottles for isotope analysis. Duplicate ortriplicate samples were collected for theconcentration analysis and single samples forthe isotope analysis. The glass bottles werefitted with a butyl stopper and an aluminiumseal. Surfacewater samples for concentrationanalysis at the sites of the manual fluxchambers were collected with a syringeconnectedtoaPVCtubewiththewaterputintoexetainers. Samples for isotope analysis werecollected by manually lowering down thesample bottles in the water and filling themavoidingbubbleformation.Toprevent

Figure2.Right:overviewmapofthelocationofAskö,southofStockholm.Left:mapofAsköandthesiteswheresamplingandmeasurementswereconducted.


9

contamination from microbial activity, theconcentrationsampleswerepoisonedwith100µL of a zinc chloride solution (ZnCl2,concentration50%)andisotopesampleswith1mLof50%ZnCl2.The samples for the incubation experimentwerecollectedwiththeNiskinbottleinsimilarglass bottles as for the isotope analysis andsealedinthesamemanner.Thesampleswerestoredinacoolingroomat15°Cfora5-week-long incubation period. This was made tosimulatesimilarenvironmentalconditionsasitwasduringthesampling.Toendtheincubation,1 mL of 50% ZnCl2 was added to the bottlesthroughthestopperwithaneedle.Furtherexperiments were conducted in June, Augustand September using a plankton net instead(Appendix1).To collect water samples for microscopeanalysis of phytoplankton genera, a planktonnet(4.1µm)wasused.Thiswaslowereddowntoadepthof5mand thenpoured intoa2Lglass bottle directly on board the researchvessel.Inthelaboratory,thewaterwaspoured

into glass bottles with 2 mL of lugol (iodinesolution) added for separation of thephytoplankton.

2.2.2 CTD: oxygen, salinity, temperature,photosynthetically active radiation andturbidity

The oxygen (O2) concentration, salinity,temperature, photosynthetically activeradiation (PAR= solar radiationbetween400-700 nmused in photosynthesis) and turbidityweremeasuredinthewaterprofilesofB1andFifångsdjupet with a CTD (Sea & SunTechnology, frequency4Hz).BeforeusingtheCTDinthefield,theO2sensorwascalibratedbyputtingtheCTDintoabucketofwatertogetherwithanaquariumpumpfor>20min.ThepumpcausedthedissolvedO2toequilibratewiththeatmosphere.CalibrationwasmadewhentheO2levelwasstabilised.WhenusingtheCTDinthefield,the instrumentwasfirstly lowereddown~0.5m below thewater surface for 1min toadjust. Thereafter, it was sent down with aspeedof25-30cm/sec.TheCTDwaslettorest

Site Coordinates(WGS84)

Conc.&iso.sampledepth(m)

Depthplanktonnet(m)

Incubationexperiment

Airsamples LGR CTD

B1 58°48'18.32"N17°37'48.32"E

0,5,10,15,20,25,30,35,40

5 June,Sep

No June,Aug,Sep

Yes

Fifångsdjupet 58°51'27.90"N17°39'25.57"E

0,9,18,22 5 June,Sep No June,Aug,Sep

Yes

JulaftonsFyr 58°51ʹ17.35ʺN17°36ʹ7.764ʺE

0 None No No June,Sep No

Chamber1 58°49'4.303"N17°38'12.292"E

0 None No Yes No No

Chamber2 58°49'16.032"N17°38'4.275"E

0 None No Yes June No

JuneAsköexp.area

58°48ʹ46.148ʺN17°37ʹ57.723ʺE

None 5 June No No No

Aug.Asköexp.area

58°48'19.519"N17°37'42.589"E

None 5 Aug No No No

Table1.Thetablepresentsthesitesandthesamplingmethodsandmeasurementsconductedineachsite(conc.=concentrations,iso.=isotope,LGR=LosGatosResearchInfraredSpectrometer,exp.area=experimentarea).TheredmarkeddepthforB1forconcentrationandisotopesamplingwerethesampledepthsinJune,whileinAugustandSeptembersampleswerecollectedfromallthedepthsdescribedinthetable.


10

1minatthebottombeforeitwaspulledbackup again. No further analysis was needed forthe provided data. However, the results fromPAR was used to obtain the euphotic depth,whichisthedepthsofwhichthePARvalueis1%ofthePARvalueatthesurface(Leeetal.,2007). 2.2.3Manualfluxchambers

Manualfluxchamberswereplacedattwosites,bothnearthecoastofthestation(figure2and3: chamber 1 and chamber 2). The samelocations were chosen for each occasion inJune, August and September. Chamber 1wasplaced towards open waters, though stillprotected from high waves, while chamber 2wasplaced in amore enclosed area. 1 - 2 airsample of 50mL volumewere collected eachdayfromthechambersforfurtheranalysesoftheconcentrationsofCH4,CO2andN2Oaswellas the isotope composition of δ13CCH4. Airsamples collectedon the first daywere takenwithasyringedirectlyfromtheairtoobtainthestart values of atmospheric concentration ofthegases.Thefollowingdays,thesamplesweretakenwithasyringeconnectedtoaPVCtube.Thistubewasinreturnconnectedviaathree-waystopcocktoanotherPVCtubefixedtothechamber. The air sampleswere injected fromthesyringeintoanupsidedown-turned20mLWheaton vial (filled with saturated saltwater)through a needle inserted in the lid (butylstopperwithaluminiumseal).Whendoingso,

parts of the containing saltwater flowed outthrough a second needle leaving anoverpressureinthevial.

2.2.4 Los Gatos Research InfraredSpectrometer: flux measurements andatmospherictransects

Measurements of the water-air-fluxes madewith an infrared spectrometer (LGR) from LosGatos Research (model 908-0011-0002, serial12-0022) were conducted at the sites B1,Fifångsdjupet,JulaftonsFyrandatAskö(inthevicinityofchamber2).Afloatingchamberwasconnectedtotheinstrumentviaa26mlongpvctube (figure 4) (ID 4 mm, OD 7 mm). Theresearch vessel was put at drift and thechamberwasplacedatthewatersurfaceontheside of the boat that allowed the chamber todriftawayfromthevessel’swindshadow.Thiswas done so that themovement of the boatwouldnotcauseanydisturbance in thewaterbelow thechamberand tominimize thewindfield disturbance around the chamber.ConcentrationsofCH4,CO2andH2O in theairinside of the chamber were measured withfrequenciesofeither1Hzorevery20thsecond.The time of measurements varied betweenapproximately5-30minutes,dependingontherateofthechangeinconcentrationinsideofthechamber. For faster concentration changes,shorter measurements were done. Somemeasurementswerealsoabortedafterashortperiod of time due to high input of watervapour into the LGR instrument. During themeasurements,informationoftheairpressure(Pa), the air temperature (K) and wind speed(m/s)weregathered(Appendix3).TransectsofatmosphericconcentrationofCH4were measured between the sites. The tubewas disconnected from the chamber and theendofthetubewasplacedatthefrontoftheboat to not contaminate the airwith exhaustgas from the boat. Coordinates were notedduringthetransectalongwiththetimeandgasconcentration. The speed of the boat was 10knotstomakelateranalyseeasier.

Figure3.Themanualfluxchambersfloatingatthechosensites:chamber1towardsopenwatersandchamber2inamoreenclosedarea.


11

2.3Laboratory&dataanalysis

2.3.1 Concentration analysis: watersamples, incubation experiments & airsamples

Analysis were made of the concentrations ofCH4, CO2 and N2O in the water samples,incubation experiments and air samples. CO2and N2O were analysed to provide furtherinformationregardingbiologicalactivity in thewater and of greenhouse gas fluxes. Theanalysis was performed with a gaschromatograph(GC)fromSRI8610C(figure5:top).Thewatersampleswerepreparedwithahelium headspace > 24 hours before analysis(exetainers = 4 mL, incubation experimentbottles=16mL)(figure5:bottom).Thesamples

were shaken for ~ 22 hours to achieveequilibriumbetweenthegaseousandaqueousphases. 3 mL headspace gas from theexetainersandair sampleswereextracted foranalysis and 2.5 mL from the incubationexperiment bottles (the rest left for isotopeanalysis).When extracting the gas, sea waterwassimultaneouslyinjectedintotheexetainersand water saturated with sodium chloride(NaCl) into the incubationexperimentbottles.ThesampleswereinjectedintotheGCviatwosequential 1 ml loops onto a parallel pair ofprecolumns (3 feet, PorapakQ) andanalyticalcolumns(9feet,HayesepD)withcarriergasofnitrogen 5.0 (99.995% purity). A FID (flameionisation detector) was used to analyse theCH4,andCO2,withpriorreductionofCO2 ina

Figure 4. Top: The LGR at the boat during ameasurement. The graphs display the real-timeconcentration of the gases. Bottom: The floatingchamberconnectedtotheLGRwitha26mlongtube.

Figure5.Top:Thegaschromatograph(SRI8610C)usedforconcentrationanalysisofCH4,CO2andN2O.Sampleswereinjectedwithasyringe.Bottom:Preparationoftheheliumheadspace in an incubationexperimentbottle.Water is extracted with a syringe simultaneously asheliumisinjected.


12

methaniser.The initial temperatureoftheFIDwas30°Candfinaltemperatureof60°C.AnECD(electron capture detector) determined theN2O.Calibrationcurvesweremadeeachdaybyalinearregressionfromtheaveragepeakareas(n=3-10)ofastandardwith11.2±0.22ppmCH4,1076±22ppmCO2and1.07±0.11ppmN2O.Airsamplesobtainedfromtheairoutsideof the laboratory were used as referencesamples to control the quality of the results(concentrations of CH4 ~ 1,9 ppm, CO2 ~ 395ppmandN2Oof~342ppbpreviousmeasuredby2otherinstruments).Thestandarddeviationofthetotalnumberofmeasuredreferenceairsamples(n=32)was7.5%.The concentration in ppm of the gases werecalculated from the calibration curves. Tocalculatetheresultsintomolarconcentrations,thefollowingequationwasused(Johnsonetal.,1990):

Cw°=ng+nwVw

(eq.4)Cw° = concentration of the dissolved gas in theoriginalsample(molesliter-1)ng=molesinheadspaceafterequilibrationnW=molesinaqueousphaseafterequilibrationVW=volumeofthewatersample(liter)Tocalculatengtheidealgaslawwasused:

ng=P*P'*Vg

RT

(eq.5)P=airpressureintheheadspace(1atm)P’=partialpressureofthegas(ppmvaluecalculatedfromthepeakareamultipliedwith10-6)Vg=volumeofheadspace(liter)R=thegasconstant(0.08205atm*Lmoles*K-1)T=airtemperatureduringanalysis(293K)To acquire nW, firstly the Bunsen solubilitycoefficient of the dry mole fraction wascalculated for each gas (Wiesenburg andGuinasso,1979):

lnβ=A1+A2100

T+A3ln

T

100+S B1+

B2T

100+B3

T

100

2 (eq.6)

β=Bunsensolubilitycoefficient(molesliter-1atm-1)S = salinity (set to 6.0‰ for all samples. Thedifferenceinthefinalgasconcentrationwith±1‰isnegligible)A and B = coefficients for each gas provided fromliterature(Appendix2)Furtheron,nWcanbecalculatedusingHenry’sLawand thevolumeof thewater (Johnsonetal.,1990): nw=P*P'*KH*Vw

(eq.7)KH=Henry’sconstantforaspecificgasTherelationshipbetweenKHandβisKH=β/RT(Clark,2015).Therefore,intheequationusedinthisthesistheKHwassubstitutedbyβ:

nw=P*P'*β*Vw

RT

(eq.8)From the calculated concentrations (Cw°) foreach sample, the average concentration andthe total standard deviation were calculatedfromtheduplicateandtriplicatesamples: STOT= ss+sGC

(eq.9)STOT=totalstandarddeviationsS=standarddeviationfromtheduplicate/triplicatesamplessGC = standard deviation of the GC (7.5%) for theaverage of the samples (average concentration*0.075)Regarding the concentration results of theexperimentsamplesinJuneobtainedfromtheNiskin bottle at B1 and Fifångsdjupet, thesewerecomparedtotheresultsobtainedfrom0m depth at the same sites using a two-sided


13

pairedt-testinRStudiocalculatingaconfidenceinterval(CI)of95%. 2.3.2 Isotope analysis: water samples,incubationexperiments&airsamples

The isotope analysis for thewater samples inthe 110 mL glass bottles, the incubationexperiment bottles and for the air samples inthe Wheaton vials was made in a gaschromatograph isotope ratio massspectrometer (GC-IRMS: GC = Trace GC UltrafromThermoScientificandIRMS=DeltaVPlusfromThermoElectronCorporation).Allsampleswere preconcentrated in a PreCon with loopsize of 10 mL for samples with headspaceconcentration < 25 ppm and a 2mL loop forsamples>25ppm(measuredpreviousintheGCdescribed in section 2.3.1 Concentrationanalysis).Theflowofthecarriergas(helium)inthe GC-IRMS was 2.0 mL/min (34 psi),temperatureofthecombustionovenat1020°C,GC injector temperature of 150°C and splitvalveonat117mL/min.Thereferencegasusedin the isotope analysis was CO2 (100%). Theaverage,standarddeviationandlinearityofthereferencegas(n=8)wasmeasuredeachdayofanalysis.Calibrationcalculationsof the 13C/12Cratioofthereferencegasweremadedailyfromastandardof100%CH4.Themeasurementsofthestandardweremadewithdirect injections(n = 4) with standard deviation < 0.1 for thecalculated δ13CCH4 each day of analysis. In thecalibration calculation, the ratio and isotopecomposition (δ13CCH4) of the reference gas vs.thestandardwasrecalculated intotheδ13CCH4ofthereferencegasvs.VPDB.Measurementofa standard of 4.96 ± 2.06 ppm (n = 1) in thePreCon was made in the beginning, in themiddleandattheendofthedaytocheckthequality of the method. The values werecomparedwithearliertestedvaluesmeasuredbythelaboratorystaff.Themethodforheadspacepreparationusedisdescribed in section 2.3.1 Concentrationanalysis. The amount of headspace for watersamples analysed with a 10 mL loop waspreparedwitha12mLheadspaceandsampleswith a 2 mL loop with 4 mL headspace.

Extraction of headspace gas was donewith asyringe flushed with argon, where 11mL gaswasusedforthe10mLloopanalysisand3mLforthe2mLloopanalysis,providinga1mLflushof the injection inlet.Toobtain theδ13CCH4vs.VPDB (Vienna Pee Dee Belemnite) of thesamples,thefollowingequationwasused:

δ13CCH4=RS

RVPDB-1 *1000

(eq.10)δ13CCH4=isotopecompositionofthesamplevs.VPDBRS= ratioof

13C/12C in thesample, calculated fromthedailycalibratedratioofthereferencegasRVPDB=givenratioof

13C/12CinVPDBFromallofthemeasurementsofthe4.96ppmstandardmadeinthePreConduringtheentireanalysisperiod(n=20),astandarddeviationofthe calculated δ13CCH4 from the PreCon wascalculated to 0.7%. For single samples, thisstandarddeviationwasusedtoobtainerrorsoftheresults.Forduplicateandtriplicatesamples,the total standard deviation was calculatedaccording to equation 9. The results of theexperimentsamplesinJuneobtainedfromtheNiskinbottleatB1andFifångsdjupetwerealsocompared to the results obtained from 0 mdepthatthesamesitesusingatwo-sidedpairedt-test.

2.3.3 Data analysis of water-air-fluxes:manualfluxchambers&LGR

ManualfluxchambersWater-air-fluxes(mgm-2day-1)ofCH4,CO2andN2O from the manual flux chambers werecalculated using the concentrations (ppm)analysed as described in section 2.3.1Concentrationanalysis.Firstly,itwasexaminedhowtheconcentrationhadchangedovertimeby plotting ppm vs. time. From this, it wasshown that equilibrium of the gases wasachievedinthechambersafterapproximately1- 2 days after placement of the chambers.Therefore, the air samples collected after theequilibrium was reached were neglected infurther description. Firstly, calculation of the


14

mass of the gases in the chambers for eachsamplewasmade:

m= P*P'*VC*(M*103)

RT

(eq.11)m=massofthegas(mg)P=airpressureintheheadspace(101325Pa)P’=partialpressureofthegas inthesample(ppmvaluecalculatedfromthepeakareamultipliedwith10-6)VC=volumeofthechamber(0.006m3)M=molecularweightofthegas(CH4=16gmoles-1,CO2 = 44 gmoles-1, N2O = 44 gmoles-1) correctedwith103toobtainresultsinmgR=thegasconstant(8.314Pa*m3moles*K-1)T=airtemperatureduringanalysis(293K)Secondly, to calculate the flux where only 2samples were collected before reachingequilibriuminthechambers,thisequationwasused:

F=m2-m1

At2-t1

(eq.12)F=water-air-flux(mgm-2day-1)m2 =mass of the gas from the second air sample(mg)m1=massofthegasfromthefirstairsample(mg)A=areaofthechamber(0.07069m2)t2=dayofthesecondsamplet1=0To obtain the fluxeswhere > 2 sampleswerecollected before reaching equilibrium in thechambers, a linear regression analysis wasmade from the calculated mass of the gasmeasured from each day (mg day-1). The fluxcalculationwasmadeby theslopegiven fromthelinearregression(k):

F= kA

(eq.13)The standard deviation for the fluxes werecalculated from the GC standard deviation of7.5%.

LGRmeasurementsThewater-air-fluxes of CH4 and CO2 from theLos Gatos Research Infrared Spectrometer(LGR) was calculated by firstly converting themeasurement time from seconds into days.Secondly,alinearregressionanalysiswasmadein R Studio for measurement with normaldistributed data to obtain a slope (k) of thechange in concentration during themeasurementfromstarttoend(ppmday-1)andconfidence interval (CI) of 95%. Formeasurements with non-normal distributeddata, a bootstrap (resampling) regressionanalysis were made (replicates = 10 000) toachieve a slope and a basic CI = 95%. Thisregressionhasshowntobeavalidmethodtousewhen having non-normal distributed data(Wang, 2001). Thirdly, the ideal gas law wasusedtocalculatethek-valueinppmday-1intomolesday-1:

k2=P*(k*10-6)*V

RT

(eq.14)k2=slope(molesday-1)P=airpressureduringthemeasurementk=slope(ppmday-1)correctedwith10-6V=totalvolumewhichincludesthechambervolume,thetubesusedbetweenthechambertotheLGRandtheinsideoftheLGR(total=0.006406m3)R=thegasconstant(8.314Pa*m3moles*K-1)T=airtemperatureduringthemeasurementThis equation was also used to calculate theminimumandmaximumk2,wherethekvaluewasexchangedbythefirstandthirdpercentilevalues from the CI. Fourthly, the minimum,average and maximum k2 values werecalculated intowater-air-fluxes (mgm-2 day-1)withtheequation:

F= k2*(M*103)

A

(eq.15)The general error of the measurements withtheLGRwas<0.01%andthereforedisregardedinthecalculationsoftheconfidenceintervals.


15

AtmospherictransectsThe known coordinates noted during theatmospheric transectmeasurements togetherwiththeCH4concentrationsatthesesiteswereprocessed in QGIS. To obtain coordinates forthegas concentration fromeach specific timemeasurement, lineswere drawn between theknowncoordinates.Thetool“Randompoints”wereusedtoaddthenumberofmeasurementpointsobtainedbetweentheknownpoints.Therandom points were sorted so that eachcoordinatefittedtherightgasconcentrationateachtimemeasurement.

2.3.4 Qualitative analysis ofphytoplanktongenera

The water samples for qualitative analysis ofphytoplankton genera were examined in aninvertedmicroscopeatresolutionof400x.Thepurposewastodoaqualitativeanalysistoseewhich genus dominated in each sample. Thenumber of individuals of the differentphytoplanktongenerawerecountedper1µLinthe microscope. The average and standarddeviationofthenumberofindividualsper1µLwerecalculatedcontinuouslypersampleduringthe analysis and when the values werestabilised,theanalysiswasended.


16

3.Results

3.1 Water concentrations and isotopecomposition: water profiles and surfacewaters

TheresultsarepresentedintheorderofresultsfromthewaterprofileatB1,thewaterprofileatFifångsdjupetandsurfacewatersatJulaftonsFyr and at the chamber sites. Further detailsregarding sampling parameters (time, airtemperature, weather conditions etc.) duringthesamplingarepresentedinAppendix3. 3.1.1Waterprofile:B1

The results from the concentration analysis,isotopeanalysesandCTDmeasurementsofthewaterprofileatB1arepresentedinfigure6andtable 2. The results of the concentration andisotopecompositionfromthedepthof30minSeptember is believed to be an analyticalanomaly.Thesuspicionisthatwaterleakedintothe Niskin bottle while obtaining the sample.Though,notestscouldbeexecutedafterwardstocontrolthissuspicion(salinitymeasurementsetc.), so the result is still presented in thegraphs.When observing the results from the variousparameters,therearecommonfactorsthatcanbedistinguished:i)inAugustandSeptember,itis seen that changes in O2 concentration andchanges instable isotopecomposition(δ13CCH4vs.VPDB) coincidedwith the thermocline andthe halocline in all threemonths. The δ13CCH4changed either above, within or below thethermocline and halocline, with lightercompositions in the surface waters, ii) inSeptember,changesinCH4concentrationwerenotedatthesamedepthsasmentionedabove,iii) there were seen decreasing values of O2concentrationsinJuneandAugustinthemiddleof the water column. Changes in O2concentration was seen >1 m above thethermoclinesinAugustandSeptember.

JuneThe thermocline and halocline in June wereshownwithatemperaturedropfrom9.8°Cto7.2°Candwithasalinityincreasefrom6.3PSUto6.5PSUbetweenthedepthsof9.5-10.5m.Below these depths, the lowest value in O2concentration was found at 18m depth withvalueof10.4mg/L.Theδ13CCH4providedlightercompositionaboveandwithinthethermocline,comparedtobelow:10m=-54.9±0.4‰versus20m = -44.3 ± 0.3‰. The CH4 concentrationshowed a significant higher value within thethermoclineat10mcomparedto0mand20mdepths:10m=31±3.4nMversus0m=24±1.9nMand20m=22±2.4nM.NosignificantdifferenceswereseenfortheconcentrationofCO2,thoughatrendoppositetotheCH4valuesarenoted.Thevaluesweregenerallyhigherat0 m and 20m and lower at 10 m and 40mdepth.TheN2Oconcentrationwassignificantlylowerat0mcomparedto20mand40mdepth:0m=7.4±0.6nMversus20m=8.7±0.7nMand40m=8.9±0.7nM.Theeuphoticdepth(1% of the surface value of thephotosynthetically active radiation (PAR)) wasseenat thedepthof 28m. The various timeswhentheCTDmeasurementswereconductedarenoted in thePAR-graphs toprovideeasierinterpretationofthePAR-data,withatimeforthis measurement at 09:00. No distinctdifferences in turbiditywas noted throughouttheprofile,although,slightlyhighervalueswereseentowardsthebottom.AugustFromtheresultsinAugust,thethermoclineandhaloclinewerefoundbetweenthedepthsof3-24m,withatemperaturechangefrom18.6°Cto6.8°Candsalinity from6.1PSUto6.9PSU.Simultaneously,adecreaseinO2concentrationwasfoundwithinthethermocline,withlowestconcentration at 13 m depth of 7.8 mg/L. Achange in the isotope composition was seenwithinthethermocline.Valuesatdepths≤15mwere more depleted (range between -60.0 ±0.4‰to-57.1±0.4‰)thanatdepthat≥25m(rangebetween-46.1±0.3‰to-45.1±0.3‰).The CH4 concentration showed continuouslyincreasing concentrations towards to seafloorwith concentration at 0 m of 32 ± 3.7 nMcomparedto40mof51±3.8nM.Therewerenosignificantdifferenceswithintheprofilein

FannyAxelsson

Bachelorthesisgeochemistry30hp

17

Time:09:00

Time:11:30

Time:14:00

Figure 6. Results of the water profile at B1 from the concentration analysis (CH4, CO2, N2O), isotope analysis (δ13CCH4 vs. VPDB) and CTD measurements (O2, temperature, salinity,

photosyntheticallyactiveradiation(PAR),turbidity).Eachrowshowsthethreedifferentmonths,whererow1=June,row2=Augustandrow3=September.Eachcolumnsharesthesamex-axis.Errorbarsrepresentsthestandarddeviation.TheshadedboxeshighlightthethermoclineineachmonthandthedashedlinesinthePAR-graphsshowstheeuphoticdepths.ThetimeoftheCTDmeasurementisalsonotedinthePAR-graphs.Thecircledresultsat30mdepthinSeptemberarethevaluessuspectedtobecontaminated.


18

theCO2concentration.Thevaluesrangedfromsurface concentration of 200 ± 22 µM tobottomconcentrationsof230±18µM.TheN2Oconcentrationwashigherinthebottomwaterswith6.7±0.5nMat0mdepthsand8.3±0.6nM at 40m depths. The euphotic depth wasalsofoundwithinthethermoclineandhaloclineatadepthof19m(at11:30)aswellasadeclineinturbidityat7.5m.SeptemberIn September, two boundaries of thethermocline and thehaloclinewere seen thatcoincided with changes in O2 concentration,isotope composition and CH4 concentration.Theupperboundarieswerefoundbetween22-30mandthelowerbetween33-36m.Withineach part, a decline in O2 concentration andchangesofδ13CCH4andCH4concentrationwasseen.Thetotalchangefromtheupperpartofthefirstthermocline/haloclinetothelowerpartof the second thermocline/halocline was:temperature=14.8-5.9°C,salinity=6.3-7.1

PSU,O2=10.3-5.7mg/L.Theδ13CCH4becamelighterbetweenthedepths0-20m,from-58.0± 0.4‰ to -60.1 ± 0.4‰, respectively.Thereafter, it distinctively changed to -46.0 ±0.3‰at25mdepth,whileitwasevenheavierat 40 m depths of -37.0 ± 0.2‰. The CH4concentrationprovidedvalueswithinthesamerangebetween0-20m:from23±2.3nMto20±1.6nM.At25mdepth,itsharplyincreasedto110±9.2nM,whileitthereafterdecreasedto72±8.3nMat40mdepths.Theconcentrationof CO2 showed an increasing trend (with nosignificantdifferences)fromsurfacewaterto25mdepth: from200±17µMto250±19µM.This trend ceased at 35 m, while theconcentrationdecreasedat40mdepthto180±22µM.Asimilar,butoppositetrendwasseenfor the N2O concentration. It decreased from7.2±0.9nMat0mto6.5±0.5nMat20m,whilehighervalueswereseenat25and35mdepth.Thenasecondincreasewasseenat40m depth to 8.4 ± 0.8 nM. The euphotic wasfoundat 17mdepth (at 14:00). The turbidity

DepthB1

CH4(nM)(aver.±stand.dev.)

δ13CCH4(‰)(aver.±stand.dev.)

CO2(µM)(aver.±stand.dev.)

N2O(nM)(aver.±stand.dev.)

Comments

0 J:24±1.9A:32±3.7S:23±2.3

J:-56.4±0.4A:-60.0±0.4S:-58.0±0.4

J:230±18A:200±22S:200±17

J:6.8±0.6A:6.7±0.5S:7.2±0.9

DTT

5 A:32±2.5S:20±1.6

A:-59.4±0.4S:-58.1±0.4

A:220±20S:210±18

A:7.1±0.5S:6.9±0.7

TT

10 J:31±3.4A:38±3.2S:19±2.1

J:-54.9±0.4A:-58.6±0.4S:-58.1±0.4

J:220±18A:220±17S:220±18

J:7.4±0.6A:7.5±0.6S:6.7±0.5

DTT

15 A:41±3.5S:21±2.9

A:-57.1±0.4S:-59.0±0.4

A:210±16S:230±17

A:7.7±0.6S:6.4±0.5

TT

20 J:22±2.4A:41±3.3S:20±1.6

J:-44.3±0.3A:-49.1±0.3S:-60.1±0.4

J:240±19A:220±17S:230±17

J:8.1±0.7A:8.1±0.6S:6.5±0.5

DDD

25 A:48±3.6S:110±9.2

A:-45.9±0.3S:-46.0±0.3

A:220±18S:240±19

A:8.2±0.6S:7.4±0.6

DD

30 A:46±3.2(S:41±4.4)

A:-46.1±0.3(S:-55.4±0.4)

A:210±16(S:230±18)

A:8.0±0.6(S:6.8±0.5)

DD,A

35 A:52±4.7S:130±10

A:-45.9±0.3S:-44.3±0.3

A:230±24S:240±18

A:8.3±0.7S:7.4±0.6

DD

40 J:28±6.1A:51±3.8S:72±8.3

J:-36.2±0.2A:-45.1±0.3S:-37.0±0.3

J:230±18A:230±18S:180±22

J:8.2±0.7A:8.4±0.6S:8.4±0.8

DDD

Table2.ResultsfromthewaterprofileatB1inJ=June,A=AugustandS=SeptemberforCH4,δ13CCH4,CO2andN2O(average±

standarddeviation).Inthecommentcolumn:D=duplicatesamples,T=triplicatesamples,A=analyticalanomaly.


19

increasedsharplybelowthethermocline,whichindicatesofastrongthermocline.When comparing the three months to eachother, it is seen that the thermocline andhaloclinewasweakest inAugust compared totheothermonths.Theuppermixed layerwas10 m in June, 3 m in August and 22 m inSeptember.Highestvalesinthesurfacewaters(≥ 20 m) was found in August and lowest inSeptember. Though, in the bottom waters inSeptembershowedhighestvaluesandJunethelowest.Theδ13CCH4showedsimilartrendsforallthree months, where the surface water waslighter than the bottom waters. However,August showed generally lightest valuesthroughoutthecolumnandJunetheheaviest,with a few exceptions. No differences wereseenintheCO2concentration,exceptat40mdepth, where September concentration waslower than the other months. For the N2Oconcentration, September provided lowervaluesthanJuneat10mdepthaswellaslowerthanbothJuneandAugustat15and20m.

3.1.2Waterprofile:Fifångsdjupet

The results from the concentration analysis,isotope analysis and CTDmeasurements fromthewaterprofileatFifångsdjupetarepresentedinfigure7andtable3.Similarresultsareshownas presented for B1: i) changes in CH4concentration, isotope composition and O2concentration coincidedwith the thermoclineand halocline, ii) two boundaries of thethermocline was seen in August, as inSeptember at B1, iii) therewas decreasingO2concentrationinJuneinthemiddleofthewatercolumn,asseeninJuneandAugustatB1,withdecreasing concentration >1 m above thethermoclines iv) isotope composition waslighterinsurfacewatersthaninbottomwatersinJuneandSeptember.Noteworthy,therangesofisotopecompositionthroughouttheprofilesare narrower at Fifångsdjupet than at B1. Nopronounced haloclines as normally observedwasseen,onlyslightlyincreasedvaluesinJuneand August andwith a sharp decreasewithinthethermoclineinSeptember.

JuneThethermoclineinJunewasfoundbetween7.5- 10 m depth, with a temperature decreasefrom12.5°Cto8.0°C.Withinthethermoclineat8mdepth,anincreaseofthesalinityfrom6.1PSU was found as well as decreased O2concentration within the thermocline withlowestvalueof9.0mg/L.TheCH4concentrationincreasedbelowthethermocline(0m=18±1.9nMvs.22m=35±2.8nM),whiletheδ13CCH4within the thermocline was lighter than theupper and lower samples, with value at 9 mdepthof-57.5±0.4‰.Thesampleat0mdepthshowed similar composition as the bottomsamples. TheCO2 concentrationdidnotdiffersignificantly, although, the bottom valuesshowedofgenerallyhighervalues:0m=220±19µMvs.22m=230±20µM.Similarshapeoftheprofileseenintheδ13CCH4graphisalsoseenin the N2O graph, with generally lowerconcentration at 9 m depth within thethermocline(7.7±0.6nM)thanatthesurfaceand in the bottom. However, no significantdifferenceswereseen.Theeuphoticdepthwasfound at 14 m depth (at 11:00) and with ahigherturbiditynotedwithinthethermocline.AugustInAugust,twodistinctthermoclineswereseen,the firstbetween7 - 9mwitha temperaturedrop from 18.5°C to 17°C. The second werefound between 11 - 12.5 m, where thetemperaturedecreasedfrom16.3°Cto13.5°C.Slightlysmallchangeswithinthesedepthswasalso seen for the salinity: firstly, an increasefrom6.1to6.2PSUandthereafteranincreaseup to 6.3 PSU at 12.5 m depth. The O2concentration decreased within boththermoclines, from 9.7 mg/L in the surfacewater until stabilised at 12.5 m depth at 5.6mg/L. The CH4 concentration increased belowthethermocline,from21±1.9nMat9mdepthto57±5.0nMat18mdepth,whiletheδ13CCH4showedlightervaluesat9mdepthof-57.7±0.4‰andat18mdepth-60.1±0.4‰.TheCO2concentration was similar between 0-18 mdepth (~200±15µM),whilea trendof lowerconcentrationwasseenat22mdepthof190±14µM.TheN2Oconcentrationwassignificantlyhigherat18mdepth(8.2±0.6nM)thanat0m(6.8±0.5nM)and9mdepth(6.9±0.5nM).Theeuphoticdepthwasfoundat12mdepth(at

FannyAxelsson

Bachelorthesisgeochemistry30hp

20

Time:11:00

Time:13:00

Time:10:45

Figure7.ResultsofthewaterprofileatFifångsdjupetfromtheconcentrationanalysis(CH4,CO2,N2O),isotopeanalysis(δ13CCH4vs.VPDB)andCTDmeasurements(O2,temperature,salinity,

photosyntheticallyactiveradiation(PAR),turbidity).Eachrowshowsthethreedifferentmonths,whererow1=June,row2=Augustandrow3=September.Eachcolumnsharesthesamex-axis..Errorbarsrepresentsthestandarddeviation.TheshadedboxeshighlightthethermoclineineachmonthandthedashedlinesinthePAR-graphsshowstheeuphoticdepths.ThetimeoftheCTDmeasurementisalsonotedinthePAR-graphs.


21

13:00). No distinct changes in turbidity wasseenwithintheprofile.SeptemberThethermoclineinSeptemberwasbetween15- 16.5 m, with temperature decrease from14.3°Cto11.0°C.Notably,thesalinityshoweddecreased values within these depth withlowestvalueof6.1PSUandhighestof6.5PSU.TheO2concentrationwasfoundtobestabilisedwithin thesedepthat5.6mg/L fromprevioussurface values of 9.7 mg/L. The CH4concentration and the δ13CCH4 showed higherconcentration and heavier composition belowthethermocline:9m=24±2.3nMand-59.6±0.4‰vs.18m=39±2.9nMand-55.1±0.4‰.TheoppositewasseenintheCO2concentrationcompared to the CH4, with decreasingconcentrationinthebottomwaters:9m=170±13µMvs.18m=140±11µM.However,theN2OconcentrationshowedsimilartrendastheCH4,withincreasingconcentrationtowardsthebottom. Although, no significant differenceswereseen.Theeuphoticdepthwas13mdepth(at10:45),whilenodifferencesoftheturbiditywasseenwithintheprofile.WhencomparingtheCH4resultsbetweenthemonths, the concentration throughout thewater columns shows similar trend of higherconcentrationsinthebottomwaterscompared

to in the surfacewaters.While in thebottomwaters at the depth of 22 m, August andSeptemberprovideddistinctivelyhigherresultsthaninJune.Regardingtheδ13CCH4inJune,thetrenddiffersbetweenthemonths.InJune,thesampleat9mdepth (within the thermocline)was lighter than the other samples in theprofile.InAugust,theheaviestcompositionwasfoundinthesurfacewater,whiletheprofileinSeptember was similar to the ones from B1,with the lightest composition in the surfacewater. Trends distinguished for the CO2concentration and for the N2O concentrationwerelowerconcentrationsofCO2towardsthebottom in August and September and higherconcentrationofN2Otowardsthebottominallthreemonths. September showed the lowestvaluesofCO2at9-22m,whileJuneshowedthehighest at 22 m. The only difference seenbetween themonths forN2O is at0mwhereJuneishigherthanAugust.

3.1.3 Surface water: Julaftons Fyr &chambersites

The results of the CH4, CO2 and N2Oconcentration as well as of δ13CCH4 of thesurfacewaterat JulaftonsFyr,chamber1andchamber2arepresentedinfigure8andtable4.

DepthFif.

CH4(nM)(aver.±stand.dev.)




Comments

0 J:18±1.9A:22±2.5S:25±2.3

J:-54.7±0.4A:-56.8±0.4S:-59.7±0.4

J:220±19A:200±15S:180±18

J:8.1±0.6A:6.8±0.5S:7.4±0.7

TTT

9

J:20±1.5A:21±1.9S:24±2.3

J:-57.5±0.4A:-57.7±0.4S:-59.6±0.4

J:220±21A:200±15S:170±13

J:7.7±0.6A:6.9±0.5S:7.8±0.6

DTT

18 J:34±2.5A:57±5.0S:39±2.9

J:-54.1±0.4A:-60.1±0.4S:-55.1±0.4

J:220±17A:200±15S:140±11

J:8.6±0.6A:8.2±0.6S:8.2±0.6

DDD

22 J:35±2.8A:68±5.1S:58±4.6

J:-53.2±0.4A:-60.4±0.4S:-55.0±0.4

J:230±20A:190±14S:120±9

J:8.9±0.7A:8.0±0.6S:8.8±0.7

DDD

Table3.ResultsfromthewaterprofileatFifångsdjupetinJ=June,A=AugustandS=SeptemberforCH4,δ13CCH4,CO2andN2O

(average±standarddeviation).Inthecommentcolumn:D=duplicatesamples,T=triplicatesamples.


22

The results from Julaftons Fyr of the CH4concentrationshowedsignificanthighervaluesinAugustandSeptembercomparedtoinJune.The results from the isotope analysis of thesample obtained in June could unfortunatelynotbeusedduetoanalyticalerror.Though,theδ13CCH4waslighterinAugustthaninSeptemberwithvaluesof-60.0±0.4‰and-57.2±0.4‰,respectively. The CO2 concentrations wassignificantlyhigher inJunethaninSeptember.NodifferenceswereseenbetweenthemonthsfortheN2Oconcentration.

Inthesurfacewateratthechambers,itisseenthattheCH4concentrationsignificantlydifferedbetweenthemonths,butwithsimilartrendsforthechambers:highestinAugustandlowestinSeptember. Though, chamber 1 showedgenerallyhighervaluesthanchamber2ineachmonth.Theδ13CCH4resultsshowsthatchamber1inAugustwasheaviercomparedtoJuneandSeptember. The opposite trend is seen atchamber2,whereAugustcompositionislightercompared to June and September. Whencomparingtheresultsbetweenthechambers,

Figure8.ResultsfromtheconcentrationandisotopeanalysisfromthesurfacewatersatJulaftonsFyr,chamber1andatchamber2.Sharedx-axisforallgraphs.a)CH4concentration(nM)witherrorbars(standarddeviation),b)CO2concentration(µM)with error bars (standard deviation), c) N2O concentration (nM)with error bars (standard deviation), d) isotopecompositionofδ13CCH4vs.VPDB(‰)witherrorbars(standarddeviation).


23

itisseenthatboththecompositioninJuneandAugust were heavier at chamber 1 than atchamber 2. No differences were seen inSeptember. At chamber 1, no differences areseenforCO2orN2OnorforCO2atchamber2.Thoughatchamber1,AugustconcentrationofN2OwaslowerthaninSeptember.

3.2Incubationexperiment

Theresultsfromtheconcentrationandisotopeanalysis of the dark incubation samplescollected with the Niskin bottle from B1 and

FifångsdjupetinJunearepresentedinfigure9.It was realised that the method for theincubation experiments obtained with aplankton net provided uncertainties in theresults. More information regarding themethodandresultsarethereforepresentedinAppendix 1. In figure 9 below, comparablevaluesfromB1andFifångsdjupetinJune(figure6&7 June=0m,minimumandmaximumoftheerrorbars)areshownasdashedlines.Fromthe two-sided paired t-test, no significantchange in the isotope composition in theexperiment samples were seen. Though CH4,CO2andN2Odecreasedduringthe5weeksofincubation(CI=95%).Theamountofdecrease

Site CH4(nM)(aver.±stand.dev.)




Comments

JulaftonsFyr

J:23±1.7A:32±3.4S:35±2.9

- A:-60.0±0.4S:-57.2±0.4

J:220±19A:210±15S:180±14

J:7.7±0.6A:6.7±0.5S:7.5±0.6

D,XTT

Chamber1

J:99±7.8A:151±11S:48±5.4

J:-55.6±0.4A:-54.2±0.4S:-56.3±0.4

J:210±19A:200±17S:190±15

J:6.9±0.6A:6.7±0.6S:7.6±0.6

DTT

Chamber2

J:50±5.0A:141±11S:38±3.0

J:-56.7±0.4A:-59.1±0.4S:-56.6±0.4

J:220±17A:210±17S:180±15

J:7.0±0.5A:6.5±0.5S:7.7±0.6

DTT

a) b) c) d)

Figure9.Results fromtheconcentration and isotope analysisof thedarkexperiment samples fromJunewith5weeks incubation. Fif.=Fifångsdjupet.Thedashedlinesrepresenttheminimumandmaximumvalueofeachparameterfromtheoriginalprofilesamplesat0mdepthatB1andFifångsdjupetfromthesamemonth(shownwithmatchingcolourscheme).a)CH4concentration(nM),b)isotopecompositionofδ13CCH4vs.VPDB(‰),c)CO2concentration(µM),d)N2Oconcentration(nM).Notethatnoneofthey-axesstartatzero.

Table4.ResultsfromthesurfacewatersatJulaftonsFyr,chamber1andchamber2inJ=June,A=AugustandS=SeptemberforCH4,δ

13CCH4,CO2andN2O(average±standarddeviation).Inthecommentcolumn:D=duplicatesamples,T=triplicatesamples,X=analyticalerrorduringisotopeanalysis.


24

was similar for both siteswhenobserving thetotalchangecomparedtotheoriginalsamples(table 5). Further results from the t-tests arefoundinAppendix1.Table5.Theresultsfromthetwo-sidedpairedt-testofthechange of CH4, δ

13CCH4, CO2 and N2O in the darkexperimentsamplesfromB1andFifångsdjupetinJune(5weeksincubation)withCI=95%.Negativevaluesshowadecrease.

Site CH4change(nM)

δ13CCH4change(‰)

CO2change(µM)

N2Ochange(nM)

B1 -2.68±0.04

-0.3±4.3

-138±27

-4.1±0.8

Fifångs-djupet

-3.7±0.3

-0.5±4.3

-127±31

-4.7±0.9

3.3Water-air-fluxes

Theresultsofthewater-air-fluxesofCH4andofCO2measuredbytheLGRandfromthemanualflux chambers are presented in the subchapters below. Though, only some of themeasurements of CO2 and none of themeasurement of N2O provided a statisticallysignificantfluxduringthedaysofmeasurementin themanual fluxchamber.These resultsaretherefore not presented in following figures,butprovidedinAppendix4:figureA4.1.Morespecific information regarding parameters(time,airtemperature,weatherconditionsetc)duringtheairsamplingatthechambersitesarepresentedinAppendix4:tableA4.1-A4.3,andwithgraphsof theconcentrationresults (ppmday-1 at the chambers and from the LGR

measurements, figure A4.2 - A4.3). Furtherdetails regarding the parameters during theLGRmeasurementsarepresentedinAppendix3.

3.3.1Methane water-air-fluxes & isotopecomposition

TheCH4water-air fluxes fromthemanual fluxchambers and from the LGR are presented infigure10andtable6andtheδ13CCH4oftheairsamples from the chambers in figure 11. Allfluxes are positive, i.e., indicating a flux ofmethane from water to air. The fluxes weregenerallyhigherat the shallower sites in Juneand August (Julaftons Fyr, Askö and chambersites). Referring to the figure 10 below, bothchamberscanbeusedfortrendanalysesaswellastheLGRresultsatB1,withthehighestfluxesoccurredinAugustof1.3mgm-2day-1atB1,5.0±0.4mgm-2day-1atchamber1and4.2±0.4mgm-2 day-1 at chamber2. By comparing theresultsbetweenchamber1and2itisseenthatthe fluxes at chamber 1 (chamber locatedtowards open waters) were generally higherthan chamber 2, though still showing similartrend in fluxes. There were two differentmeasurementsmadeforeachchamberinJunewhere the first measurements (Mon - Tues)werehigherthanthesecond(Wedn-Thurs)forboth chambers. The second measurementsshowed similar amount of fluxes as inSeptember for both chambers. Whencomparing the LGR measurement at Askö inJunewiththechambermeasurementsitisseenthatitwassimilarresultaschamber1inJune,

Figure10.TheCH4water-air-flux(mgm-2day-1)measuredbytheLGR(B1,Fifångsdjupet,Tallholmen,JulaftonsFyr,Askö)andwiththemanualfluxchambers(chamber1&2).TheCI=95%oftheLGRresultsandthestandarddeviationsofthechamberresultsarepresented.TwomeasurementoccasionswereperformedinJuneatthechambersites,notedasthe1stmeasurement(Mon-Tues)and2ndmeasurement(Wedn-Thurs).


25

eventhoughthemeasurementoftheLGRwasconductedinthevicinityofchamber2.The fluxes at Fifångsdjupet and Julaftons Fyrwere higher in September than in June, incontrast to theother sites. The fluxes in Junewere0.51mgm-2day-1atFifångsdjupetand1.1mgm-2day-1atJulaftonsFyr.Themeasurementat Tallholmen in June provided also similarresultsasFifångsdjupet:0.57mgm-2day-1.The δ13CCH4 of the air samples from thechambers showed the same trend for allmeasurements,wheretheisotopecompositionbecame lighter with time. The atmosphericstarting values were in the range of -52.0 ±0.3‰to-49.2±0.3‰atchamber1andof-52.7±0.3‰to-49.7±0.3‰atchamber2.Theonly

measurement showing a deviating trend ischamber 2 in August, where the isotopecompositionatday2isheavier(-58.6±0.4‰)thanatday1(-59.5±0.4‰).Thiscouldbeanindication of leakage from the chamber(Appendix5:5.4).

3.3.2Carbondioxidewater-air-fluxes

TheresultsfromtheCO2water-air-fluxesvariesbetween the sites where both positive andnegativefluxesareseeninfigure12andtable6.AcommonfactorforB1andFifångsdjupetisthat negative fluxes occurred at both sites inJune and August. The fluxes in August wereevidentlymorenegativethaninJunewith-1500

Figure11.Isotopecompositionofδ13CCH4vs.VPDB(‰)witherrorbars(standarddeviation)oftheairsamplesfromeachdayofsamplecollecting.June1st=thefirstmeasurement(Mon-Tues),June2nd=thesecondmeasurement(Wedn-Thurs).

Figure12.TheCO2water-air-flux(mgm-2day-1)measuredbytheLGR(B1,Fifångsdjupet,JulaftonsFyr,Askö)andwiththemanualfluxchambers(chamber2).TheCI=95%oftheLGRresultsandthestandarddeviationsofthechamberresults are presented. Two measurement occasions were performed in June at chamber2, noted as the 1stmeasurement(Mon-Tues)and2ndmeasurement(Wedn-Thurs).


26

mg m-2 day-1 and with -1100 mg m-2 day-1,respectively.Comparedtotheothersites,highpositivefluxesareseenatJulaftonsFyrforbothJuneandSeptember;590mgm-2day-1andmgm-2 day-1, respectively. The LGR results fromAskö in June differs distinctively from theresultsofthechambers.ThefluxoftheLGRwas-500mgm-2day-1incontrasttochamber2with13mgm-2day-1.

3.4Atmospherictransects

The transects of atmospheric CH4 from June,August and September aredisplayed in figure13. The results in June and August showedsimilar results, with the exception of thegenerallyhigherconcentrationsatJulaftonsFyrinJune(~1.89ppm).Thistrendisseenneitherin August nor in September. Otherwise, theatmosphericCH4concentrationsattheareaofFifångsdjupet were generally lower (~ 1.87

ppm)thancloserthesouth-westerncoastlineofAskö and with increasing concentrationtowards the station (~ 1.88 - 1.89 ppm). TheresultsinSeptemberisbelievedtobeskewedinaccuracy, since the concentration range wasgenerallyhigher (1.91 -1.93).Thesuspicion isthattheLGRwasinneedforcalibration,whichwasnotpossibleduringtheAskövisit.Nootherdata(fluxes,waterconcentration)supportsthehigh atmospheric September concentrations.Thoughwhenobservingthegeneraltrend,itisnotedthatthedaywhenthetransectbetweenFifångsdjupet - Askö was measured (15/9)showed lower concentrations (~ 1.91 - 1.92ppm)thanatthedayoftheB1-Askö-transect(14/9, ~ 1.92 - 1.93). Similar wind conditionswereobservedforbothdays(Appendix3).

Site CH4flux(mgm-2day-1)(aver.±stand.dev.)

CO2flux(mgm-2day-1)(aver.±stand.dev.)

Comments

B1 J:0.46±0.02A:1.3±0.007S:0.31±0.002

J:-170±6A:-1500±7S:-130±1

LGR

Fifångsdjupet J:0.51±0.03A:0.40±0.004S:0.57±0.002

J:-220±18A:-1100±7.4S:280±2

LGR

Tallholmen J:0.57±0.0005 - LGR,X

JulaftonsFyr J:1.1±0.005S:1.9±0.008

J:590±5S:640±3

LGR

Askö J:2.1±0.01 J:-490±3 LGR

Chamber1 J1st:2.0±0.6J2nd:1.2±0.3A:5.0±0.4S:0.83±0.07

- - - -

11,X1,X2,X

Chamber2 J1st:0.72±0.1J2nd:0.25±0.03A:4.5±0.3S:0.46±0.1

J1st:13±9J2nd:-16±14A:61±15

-

1,X2,X22,X

Table6.ResultsfromtheCH4andCO2water-air-fluxesmeasuredbytheLGRandatthechambersitesinJ=June,A=AugustandS=September(LGRresults:average±CI=95%,chamberresults:average±standarddeviation).Inthecommentcolumn:1=1daymeasurementtimeinthemanualfluxchambers,2=2daysmeasurementtimeinthemanualfluxchambers,X=nosignificantfluxofCO2.


27

3.5 Qualitative analysis of phytoplanktongenerainthewater

The qualitative analysis of dominatingphytoplanktongenerainthewatersamplesarepresentedintable7-9andcommongeneraareshown in figure 14. The dominating genus(noted“++++”inthetables)ineachmonthatallsites besides at Fifångsdjupet in August werethecyanobacteriaAphanizomenonspp.,whichwere also accompanied by the cyanobacteriaSkeletonema spp. in September at B1. AtFifångsdjupetinAugust,thedominatinggenuswasfoundtobetheciliateTintinnidspp.,whichis a zooplankton. This genus was includedanyway due to its high presence. It was alsopresentinthewaterfromAsköexperimentareain August and in the September sample fromFifångsdjupet. Other general common generafoundwasthecyanobacteriaLyngbyaspp.andDolichospermum spp. Also, Nodularia spp.(cyanobacteria)andChaetocerosspp.(diatom)were found in each sample, but with lowpresence.Noaccumulationofalgaewasseenatthe water surface in August during sampling,but chlorophyll data (µg/L) fromwerehighestduringthismonth(Appendix3).Whencomparingthenumberofgenerafoundineachmonthit isseenthattherewashighervariation among the genera found in June.Cyanobacteria was the dominating type ofphytoplankton in August, where less diatomgenerawerecomparedtoinJune.

Figure13.AtmospherictransectsofCH4concentration(ppm)in a) June (range1.87 -1.89ppm),b)August (range1.87 -1.89ppm)andc)September(1.91-1.93ppm).NotethattheconcentrationrangedifferinSeptembertotheothermonthsandcannotbecomparedbythecolourscheme.


28

June Asköexperimentarea Typeofphytoplankton Presence

Aphanizomenonspp. Cyanobacteria ++++ Lyngbyaspp. Cyanobacteria +++ Dolichospermumspp. Cyanobacteria ++ Chaetocerosspp. Diatom ++ Nodulariaspp. Cyanobacteria ++ Skeletonemaspp. Diatom + Fragilariaspp. Diatom +

Melosiraspp. Diatom + Naviculaspp. Diatom + Stephanodiscusspp. Diatom + Thalassoiraspp. Diatom +

Aug Asköexperimentarea Typeofphytoplankton Fifångsdjupet Typeofphytoplankton Presence Aphanizomenonspp. Cyanobacteria Tintinnidspp. Ciliate(zooplankton) ++++ Lyngbyaspp. Cyanobacteria Dolichospermumspp. Cyanobacteria +++ Tintinnidspp. Ciliate(zooplankton) - - +++

Dolichospermumspp. Cyanobacteria - - +++ Nodulariaspp. Cyanobacteria Nodulariaspp. Cyanobacteria ++ Chaetocerosspp. Diatom Chaetocerosspp. Diatom + - - Naviculaspp. Diatom + - - Fragilariaspp. Diatom +

Sep B1 Typeofphytoplankton Fifångsdjupet Typeofphytoplankton Presence

Skeletonemaspp. Diatom Aphanizomenonspp. Cyanobacteria ++++

Aphanizomenonspp. Cyanobacteria - - ++++

Lyngbyaspp. Cyanobacteria Lyngbyaspp. Cyanobacteria ++

Chaetocerosspp. Diatom Dolichospermumspp. Cyanobacteria +

Dinophysisspp. Diatom Tintinnidspp. Ciliate(zooplankton) +

Fragilariaspp. Diatom Chaetocerosspp. Diatom +

Stephanodiscusspp. Diatom Nodulariaspp. Cyanobacteria +

Nodulariaspp. Cyanobacteria Skeletonemaspp. Diatom +

Table7.Theresults fromthequalitativemicroscope analysis ofdominatingphytoplankton genus/genera (++++) from theAsköexperimentareainJune.The“+”-signsshowstherelativeamountofeachphytoplanktongeneraoccurringinthesamples.

Table8.Theresults fromthequalitativemicroscope analysis ofdominatingphytoplankton genus/genera (++++) from theAsköexperimentareaandFifångsdjupetinAugust.The“+”-signsshowstherelativeamountofeachphytoplanktongeneraoccurringinthesamples.

Table 9. The results from the qualitativemicroscope analysis of dominating phytoplankton genus/genera (++++) from B1 andFifångsdjupetinSeptember.The“+”-signsshowstherelativeamountofeachphytoplanktongeneraoccurringinthesamples.


29

A) B)

C) D)

E) F)

Figure14.Commonphytoplanktongeneraoccurringinthewatersamples.Allimagesarephotographedfromthe400xenlargementofthemicroscope.A&B)Dolichospermumspp.(cyanobacteria)C)Aphanizomenonspp.(cyanobacteria)D)Skeletomenaspp.(diatom)E)ThreeTintinnidspp.(ciliate)andNodulariaspp.(cyanobacteria)inthemiddle,F)Nodulariaspp.


30

4.DiscussionWhatisfoundinsomeofthedatainthisreportareseeminglyinterestingresultsthatarenotinlinewithwhataconventionalmodelofawaterprofile would predict. But before addressingthosefindings,itisdiscussedabouttheresultsthatare in linewithwhatwouldbepredicted.Usually,awatercolumnprofileduringsummertimewouldbepredictedtohavestratificationsin the profile from the thermocline and thehalocline. Thiswould provide distinct changesin dissolved gas concentrations and isotopecompositions in the profile (Axell, 1998). Insurface waters with high concentrations ofgases with respect to the atmosphere, therewould be positive water-air-fluxes withdissolved gases emitting from the water.Measuredwater-air-fluxesofCH4 in theBalticSeahaveshownvaluesof0.16-1.9mgm-2day-1(Bange et al., 1994). Similar values have alsobeen measured in the area around Askö(Brüchertetal.,unpublisheddata).Mostoftheresultspresentedinthisreportareinlinewiththese predictions: stratifications in the watercolumn prevented mixing of gases and CH4water-air-fluxesgenerallyaround0.5-2mgm-2day-1(eventhoughfluxesupto5mgm-2day-1was measured as well). The methanemeasurements in the atmospheric transectswere all close to the regional atmosphericvalues of ~ 1.89 ppm (Dlugokencky, 2017),except the concentrations in Septembershowing higher values. As mentioned in theresults,theSeptemberdatamaybeskewedandhavetobecorrectedbypost-calibrationoftheLGR instrument. The δ13CCH4 values of theatmosphere measured at the chamber siteswere 13C-depleted when comparing to theaverage atmospheric compositionofmethaneof-47.5‰(Morimotoetal.,2017;Nisbetetal.,2016).However, the concentrationsof CH4 ofthe atmosphere samples was near aconcentration limit of which the GC-IRMSprovided more 13C-depleted results than thetruevalues,asshownintestsperformedbythelaboratory staff. This could have skewed theresults,givinggenerallylightervalues.

Considering the results of CO2, the dissolvedCO2 in the water is a part of the carbonatesystem, that is dependent on the pH of thewater (Schulz, 2006). The CO2 results arethereforenoteasytofurtherinterpretwithoutdata of the pH. In some of the months, thesurfacewaterconcentrationswerehigherthaninthebottomwater,whileinothersitweretheopposite.Also,theCO2fluxesdifferedbetweenthe months and sites, where some werenegative and some positive. There are nostraightforwardexplanationsforthisresult,butonecanassumethatatthemeasurementssiteswith negative fluxes the photosynthesis rateswerehigher.ThemeasuredN2Oconcentrationsare similar to previous measured values (C.Olsson, unpubl. data). No significant fluxes ofN2O was measured, which complies with thefactthatthesurfaceconcentrationswereclosetothesaturationconcentrationofN2O(~7nMfromSarmientoandGruber,2006).Regardingthe findings of dominating phytoplanktongenerainthesurfacewaters,thesewereinlinewith genera usually occurring during summertime in the research area (Stal et al., 2003;WasmundandUhlig,2003).Also,areportfromSMHI (2017) that summarised phytoplanktongenera occurring near the study site showedpresenceofsimilargeneraasfoundhere.Furtherontothemoreinterestingfindingsthatarenotinlinewithwhatwouldbepredictedina conventional model. What would bepredictedisasedimentsourceofmethanethatdiffuses upwards though the water column.Under these circumstances, it would beexpected less amount of CH4 aswell asmore13C-enriched CH4 in the surfacewater than inthebottomwaters(SchneidervonDeimlingetal., 2011). This is due to CH4 oxidation duringtheascentinthewatercolumnthatdecreasestheconcentrationandenrichestheCH4with13C(Whiticar, 1999). Also, stratifications in thewater prevent mixing of the dissolved gaswithinthewatercolumn.TheCH4inthesurfacewater would also be in equilibrium with theatmosphere, which would provide similarisotope values in the water as in theatmosphere(Knoxetal.,1992).AconventionalmodelwouldalsopredictanO2concentrationprofile with decreasing concentrations below


31

the thermocline. That would be due tostratificationsinthewater,duetolessamountofphotosynthesisindeeperwatersandduetorespirationoforganicmatterusingtheO2inthewater (Axell, 1998). However,what is seen inthewater profiles at B1 and Fifångsdjupet donot correspond to these predictions. At bothsites,itwasmoreenrichedδ13CCH4valuesinthebottomwatersthaninthesurfacewaters,withexception at Fifångsdjupet in August. Theδ13CCH4valuesinthesurfacewatersweremore13C-depleted compared to the atmosphericcomposition and with concentrations > 700%larger than the saturation concentration(Sarmiento and Gruber, 2006). Also, a CH4concentration maxima was seen within thethermoclineatB1inJune.Fromtheseresults,itis assumed that the CH4 in the surfacewaterabove the thermocline would have to haveotherCH4sourcesbesidesaverticaltransportofdiffusedgasfromthesediment.RegardingtheO2concentration,thevaluesatB1inJuneandAugustandatFifångsdjupetinJunedecreasedin themiddle of the water column, within orbelowthethermoclinetothenfurtherincreasebelow. It is assumed that lateral water inputcould have occurred, providing the higher O2concentrationsinthebottomwatercomparedto in themiddle of thewater columns. Thesefindings have led to three hypotheses, wherethefirstdiscussesalateralwaterinputandthetwo latter discusses other CH4 sources thandiffusedgasfromthesediment:

• Hypothesis 1) assumption of a lateralinput of bottomwaters that providedtherelativehighconcentrationsofO2inthe bottom waters. It is discussedwhether a possible water input couldhaveaffectedtheδ13CCH4valuesinthebottomwateraswell.

• Hypothesis 2) it is proposed thatebullition bubbles containing 13C-depletedCH4reachesuptothemixedlayerinthewatercolumn.Ifdissolvinginthemixedlayer,theδ13CCH4valuesofthebubblesmixwith the compositionofthegasinthewater.ItissuggestedtohappeninsituatFifångsdjupet,whileebullition could occur in shallowercoastal waters with water currents

transporting the 13C-depleted CH4 outtoB1.

• Hypothesis 3) suggestion of a CH4

productioninthephoticofthesurfacewaterzoneatthechambersitesandatB1.

Thesehypothesisesarefurtherelaboratedandtested in various ways in the sub-chaptersbelow.

4.1 Hypothesis 1: lateral input ofisotopicallyheavybottomwaters

Hypothesis1considersalateralinputofbottomwater that effected theO2 concentration andpossibly the isotope composition. Thedecreased values of O2 in the middle of thewatercolumnsatB1inJuneandAugustandatFifångsdjupetinJunecontradictsapredictedO2profile.Theprofilesof theδ13CCH4values inallmonths at B1 and in June and September atFifångsdjupet also contradicts a predictedprofile.Itcouldbeassumedthatlateralinputofwater effected the profile of the O2concentration: either with input of lessoxygenatedwater in themiddle of the watercolumnorwithhigheroxygenatedwaterinthebottomwaters.Ifproceedingwiththelatter,itcould be argued that the water transportedlaterally into the areas could contain 13C-enriched CH4. That could provide anexplanationfortheheavyisotopecompositionseen in the bottom waters in the monthsmentioned above. Stratification in the watercolumn prevented mixing with the surfacewaterandtherebyprovidingdistinctchangesinisotopecompositions.If focusing more on Fifångsdjupet, aconventionalprofileofδ13CCH4valueswasseenin August. It can be assumed that the lateralwater input ceased from June into August aswell as that the water stratification becameweaker,asseenintheresults.Thiscouldhaveprovided an increased vertical upwardtransport of diffused CH4 from the sediment.SupposingthatthediffusingCH4wasmore13C-


32

depleted then the bottom water, then theisotopecompositionwouldbecome lighter,asalso seen in the profile in August. Whenconsidering the results from September, noindicationsof lateralwater inputwereseen intheO2data.However,theδ13CCH4valueswereyet again more 13C-enriched in the bottomwater. If remaining on this hypothesis, theexplanation for this result could be that thestrong stratification from the thermocline inSeptember inhibited the vertical transport ofCH4thatwasallowedinAugust.Whenthegasin thebottomwaterwasoxidised, theδ13CCH4values became heavier and leading to theprofileasseeninSeptember.Regarding the site at B1, the possible lateralinputwouldhaveceasedintoSeptembersincethe O2 data showed a conventional profile inthat month. However, no indications of averticalupwardtransportasasinglesourceofCH4 to the surface water were inferred inSeptember.Theδ13CCH4valuesremainedmoreenrichedinthebottomwatercomparedtothesurfacewater.Itissuggestedtobeduetothesame reason as for Fifångsdjupet: inhibitionfrom a strong stratification of the water andoxidationofCH4inthebottomwater.To test if this hypothesis of a lateral input ofisotopically heavy bottom water is valid, onewould need data of water current directionsduringthesummerof2017andoftheisotopecomposition of the diffusing CH4 from thesedimentatFifångsdjupet.Unfortunately,noneof this data have been found and furtherconclusions cannot be currently made. Butassuming that this hypothesis is a validexplanationfortheseresults,thequestionstillremains regarding the source(s) of the CH4 inthe surface waters. The CH4 concentrationincreasedatB1fromJuneintoAugustandalsoat Fifångsdjupet from June into September.Consequently, there must have been othersourcescontributing to theCH4 in thesurfacewater to sustain the concentration increase.Further hypotheses of CH4 sources arethereforeexaminedbelow.

4.2Hypothesis2:methaneebullition

4.2.1EbullitioninsituatFifångsdjupet

The second hypothesis to explain the isotopecomposition in the water column atFifångsdjupet considers gas ebullition in thearea at Tallholmen, as referred to in theintroduction. Approximately 600 gas-seepagesites have been identified in the unpublisheddata (E. Weidner, unpubl. data). Oneexplanationfortheresultsoftheδ13CCH4valuescould be that bubbles containing CH4 fromseepagesites in thesedimentare transportedinto the mixed layer above the thermoclinewithoutbeingdissolvedoroxidisedduringtheascent. If the bubbles are trapped and thendissolvedinthemixedlayer,theoriginalisotopecomposition would be preserved. This couldprovidethedissolvedCH4inthesurfacewateralighter composition than in bottom waterswheretheCH4alreadyhasbeenoxidised.Thisisillustratedinfigure16.InastudybySchneidervon Deimling et al. (2011), the authorsconcluded that bubbles are able to betransported to themixed layer. Bymodelling,theyestimatedthat<4%ofthebubbles fromebullitionsitesintheNorthSea(depth~70m)reachedthemixedlayer.Tofurtherinvestigatetheebullitionhypothesis,it was calculated how many bubbles sec-1seepage-site-1wereneededtoreachuptothemixedlayerbetweenJunetoAugust.Detailsareelaborated in Appendix 5: 5.1. In thecalculation, the assumptions were a steady-state of CH4 concentration between June toAugustandwithebullitionasasinglesourceofCH4tothesurfacewater.Theobservationthat<4%ofbubbles fromseepagesites reach themixed layer was assumed to be valid in thisscenario.Thetotalseepageflowwascalculatedaswell,meaning thenumberof bubbles sec-1site-1neededtoseepfromeachsitetoresultin4% reaching themixed layer. Various bubblessizeswereusedinthecalculation,toprovidearangeofflowneededtooccur.Theresultsarepresentedintable10.


33

Table 10. Results from the calculation of number ofbubbles sec-1 seepage-site-1 needed to reach themixedlayerattheFifångsdjupetareabetweenJune-Aug.

Bubblediameter(mm)

Bubbles reachingthe mixed layer(bubblessec-1site-1)

Totalseepflow(bubbles sec-1site-1)

2 9 2253 3 754 1 255 0.6 156 0.4 10

The results show that the total seepage flowfromthegas-seepagesitesattheFifångsdjupetareaisneededtobebetween10-225bubblessec-1 site-1whereby0.4 -9bubbles sec-1 site-1from this flow are needed to reach up to themixedlayer.Thetotalseepflowprovidesafluxofapproximately10-4molessec-1site-1fromthesediment.Withlogicalreasoning,theresultforthe 2mm bubble size with 225 bubbles sec-1site-1isseeminglyahighnumber.However,theresults are credible when comparing to thestudybySchneidervonDeimlingetal. (2011).Theyfoundatotalebullitionfluxfrom550sitesof1.2molesofCH4year-1withaveragebubblediametersof4.5mm.Thisisequivalenttoafluxof 7 * 10-5moles sec-1 site-1,which is not toodifferentfromwhatwascalculatedhere.Aninsitu ebullition source of CH4 to the surfacewater at this site is therefore supported bythese results. Though, geophysical acousticdata would be needed to provide furtherinformationofbubblessizesandseepageflowattheebullitionsites.

To expand this hypothesis to B1, where noseepage sites have been noted (as toknowledgeoftheauthoratthetimeofwriting),informationregardingtheavailabilityofCH4inthe gas phase in the sediment is needed. Astudy by Sawicka and Brüchert (2017) showsthat the CH4 concentration in the top 5 cmsediment at the sites B1 and H6 (located inHimmerfjärdenestuarynorthofFifångsdjupet:58°04’08”N, 17°40’63”E) did not exceed theCH4 saturation concentration in theirmeasurements. In this regard, no bubbleswouldbeabletoescapefromthesedimentintothe water column. This is also observed inThang et al. (2013) at the site H2 (east ofFifångsdjupet: 58°56”04’Nʹ17°43”81ʹE).Inthatstudy,theconcentrationofCH4inthetop20cmof sediment cores did not either exceed thesaturationconcentration.Withthesestudiesinmind,aninsituebullitionsourceofCH4intothemixed layer at B1 is not supported. Theebullition hypothesis is therefore furtherelaboratedconsideringthissite.

4.2.2 Ebullition at shallower areas withlateralinputofsurfacewaterstoB1

TheelaborationoftheebullitionhypothesisforB1 considers an ebullition source at theshallower areas closer to the coastline, at thechambersitesand JulaftonsFyr.Theassumedrelative13C-depletedCH4fromtheebullitionisbelieved to be laterally transported with

Figure16.Anidealisedandsimplifiedillustrationofhypothesis2(notinscaleorwithaccuratebathymetry).Right:2.1)insituebullitioncontributes to aCH4 input to the surfacewater at Fifångsdjupet.96%of thebubbles are assumedtobedissolvedinthebottomwaters,while4%isexpectedtoreachuptothemixedlayer.Left:2.2)ebullitionatshallowersiteswithlateralsurfacewatertransportcontributestoaCH4inputtoB1.


34

currents in the surface water from theshallowerareasouttotheareaofB1(figure16).The basis for the hypothesis is that similarisotopecompositionsinthesurfacewaterwerefound at B1 and at the shallower sites. Also,higher CH4 concentrations were found at theshallowerareascomparedtoatB1,enhancingthe idea of a CH4 source to B1 from theshallower areas. If ebullition takes place atthese sites, it would be fair to assume that alarger amount of bubbles would reach thesurfacewaterduetotheshallowdepths(~10–15 m). It is also considered likely that thebubbleswouldbetransportedallthewayuptothewater surface,whichwould lead todirectemissionsintotheatmosphere(Schmaleet.al.,2005).A way to investigate this hypothesis is bysearchingfor indicationsofdirectemissionsinthedataofthemeasuredwater-air-fluxatthemanual fluxchambers.Theassumption is thatthe isotope composition of the possibleebullition bubbles deviates from thecomposition in the surface water, since nooxidation is believed to have altered thecomposition of the bubbles. Therefore, if theisotope composition of thewater-air-flux (nottobeconfusedwiththeisotopecompositionoftheair samplesof the chamberspresented infigure11)highlydeviatesfromthecompositionin the surfacewater, then that could provideindications of direct emissions from anebullition source. The isotope composition ofthe fluxes was calculated to test this. ThecalculationisfurtherelaboratedinAppendix5:5.2. The calculated values of the fluxes werecomparedwiththeisotopecompositionofthesurface water to obtain the difference inisotope composition (∆13CCH4). For the sake ofpresentation,thefluxesfromthethreemonthsof the two chambers were grouped into fourgroups:groupN,D,DN1andDN2,where“N”referstothatthefluxesoccurredduringnighttime and “D” during day time. The fluxes ingroupNwerecoupledwiththefluxesingroupDandalsogroupDN1coupledwithgroupDN2.Thisisclarifiedintable11.The results from of the ∆13CCH4 values arepresentedinfigure17andtable12.Aschematicoverview of the results of chamber 2 in

September is also presented in figure 18 toprovideabetterunderstandingofthegroupingofthedaysandforbetterdistinguishingofthevariousparametersusedinthefluxcalculationinAppendix5.Table 11. Table of the grouped fluxes. The column“Sampling”showsbetweenwhichsamplingoccasionsthatthefluxesoccurred.

Group Sampling Hoursbetweensampling

Day/night

N 1stto2nd ~12 Night

D 2ndto3rd ~12 Day

DN1 1stto2nd ~24 Day&nightDN2 2ndto3rd ~24 Day&night

The result for group DN2 showed a positive∆13CCH4 value, thus with a flux with heavierisotope composition than the surface water.However,theassumptionisthatleakagemighthaveoccurredduringthesamplingbetweenthe2ndand3rdsamplingoccasionatthechamberingroup DN2. That would have mixed the airinside of the chamber with air in theatmosphere, therebygiving thechamberairaheavier isotope composition. That would inreturnaltertheresultsforthecalculationmadehere.Furtherexaminationofapossibleleakageis given inAppendix 5: 5.4. The results of the∆13CCH4 from group DN2 is thereforedisregardedinfurtherdiscussion.Table 12. Results from the difference in isotopecomposition ∆13CCH4 (‰) ± standard deviation betweenthecalculatedfluxesandthesurfacewateratthechambersites.Negativevaluesshowofmore13C-depletedfluxthansurfacewaterandviceversa.J1st=June1stmeasurement,J2nd=June2ndmeasurement,A=August,S=September.

Group Chamber1:∆13CCH4(‰),min-max

Chamber2:∆13CCH4(‰),min-max

N J2nd:-2.7±0.4S:-0.5±0.4

J2nd:0.2±0.4S:-0.4±0.4

D J:-S:-0.9±0.4

J2nd:-2.6±0.4S:-4.1±0.4

DN1 J1st:-2.6±0.4A:-3.8±0.4

J1st:-1.7±0.4A:-0.7±0.4

DN2 A:- A:0.9±0.4

Inordertodrawmoresolidconclusionsontheimportance of direct emissions for the fluxes,


35

the isotope composition of CH4 bubbles fromthesedimentmustbemeasuredatthespecificsites.Itisalsoimportanttoconsiderthatifthefluxescontainbothwater-air-fluxesanddirectemissions,thentheisotopecompositionwillbea mixture of the compositions from twosources. Though,what is interesting to see intheresultisthatthecompositionofthefluxes

in group N were close to zero (exception ofchamber 1: June 2nd: group N), meaning thatthefluxesdidnotdiffermuchfromthesurfacewater. GroupD andDN1, on the other hand,wereall lighterthanthesurfacewateraswellasgenerallylightercomparedtogroupN.Thiscouldshowthatthefluxesoccurringduringdaytime are lighter than fluxes occurring during

Figure17.Resultsfromthecalculationofthedifferenceinisotopecomposition∆13CCH4vs.VPDB(‰)fromthecalculatedfluxcompositionatthechamberssitescomparedtothecompositionofthesurfacewateratthechambersitesforeachmonth.Negativevaluesshowoflightercompositionofthefluxthanthewaterandviceversa.Theshadedbackgroundsrepresentthetime the fluxesoccurredduring:darkblue=night time,yellow=daytime, green=day&night time. June1st = the firstmeasurementmadeinJune(Mon-Tues),June2nd=thesecondmeasurementmadeinJune(Wedn-Thurs).

Figure18.SchematicoverviewoftheparametersusedintheequationinAppendix5:5.4aswellastheresultsfromchamber2inSeptember(figure17)asanexample:δ13CF=CH4fluxnight,δ

13CX=airinthechamberatthe1stsamplingandδ13CTOT=airinthechamberinthe2ndsampling.Thevaluespresentedinthefigurearetheaverageresultsfromchamber2inSeptember,wherethedayfluxes(-60.7‰)arelighterthanthenightfluxes(-57.0‰).GroupN:fluxbetweenthe1stand2ndsampling,groupD:fluxbetweenthe2ndand3rdsampling.Thechamberairbecomeslighterwithtimeduetomixingwiththe13C-depletedflux.


36

nighttime.Areasonforthisresultcouldbeduetovariationinwindstrengthbetweendayandnight that could influence the flux, sincewindspeed has a direct effect on the flux(Wanninkhof et al., 2009). However, noindicationsofaspecificrelationshipwereseenintheamountoffluxwithdaytime/nighttime(Appendix 5: 5.3) and it could be argued thatthere would therefore not be a relationshipwiththe isotopecompositionaswell.AnotherreasonfortheresultcouldbeaCH4productionsource in the photic zone during day timeinfluencingthefluxisotopecomposition.Tofurtherverifyiftherecouldoccurebullitionin the shallower areas, geophysical acousticdatawouldbeneeded.ToalsofurtherverifyiftherecouldbelateraltransportofCH4fromtheshallower sites to B1, information regardingwater currents are necessary. None of thesedata were possible to obtain within theframeworkofthiswork.

4.3 Hypothesis 3: methane production intheoxygenatedsurfacewater

Frompreviousdiscussion,itwasapparentthatday light presumably had an effect on theisotope compositionof thewater-air-fluxes atthe chamber sites in all three months. ThiscouldindicateofaCH4productioncoupledwithbiological activity occurring under lightconditions. This supports themain hypothesisfor this report, i.e., thatmeasurablemethaneproduction occurs in the oxygenated surfacewaterinthephoticzone.Thisfindingdoesalsosupport the main hypothesis that thisproduction significantly contributes to a CH4water-air-flux, since the concentrations at thechamber sites increased in August during thealgal bloom. The production pathway in themain hypothesis was believed to behydrogenotrophic methano-genesis. Asmentioned in the introduction, this pathwayuses H2 as a main component which can beproduced during nitrogen fixation bycyanobacteria (Bothe et al., 2010) and occursmostly under light conditions (Hong, 1992).However, it cannot be concluded whetherhydrogenotrophic methanogenesis could be

thepathway,moredatabeforedoingsowouldbeneeded.Consideringtheresultsforthedarkincubationexperiment, no indications of CH4 productionwere shown. Though, the CO2 and N2Oconcentrations provided indications ofbiologicalactivity.Iflightincubationshadbeenconductedwiththesamemethod,itwouldhadbeeninterestingtocompareifanydifferenceswereseenbetweenthedarkandlightsamples.If higher CH4 concentrations and/or lighterisotope values had been seen in the lightsamples, further conclusions could have beenmade regarding a CH4 production under lightconditions. The conclusions to be drawn fornowisthatnoCH4productionoccurredunderdarkconditions.However,anindicationofCH4productionintheoxygenatedsurfacewateratB1inJuneistheconcentrationmaximaofCH4withinthethermocline.Thisisinlinewithwhatother studies have found regarding methaneproduction in oxygenated surface waters; aproductioninconjunctionwiththethermocline(SchneidervonDeimlingetal.,2011;Tangetal.,2014). InthestudybySchneidervonDeimlinget al. (2011), the authors found methanemaxima within the thermocline as well aslighter isotope composition compared tobottomandsurfacewaters.Theysuggestedaninsitubiologicalproductionfortheirresults.One final indicationofmethaneproduction intheoxygenated,photiczoneistherelationshipbetween the CH4 and CO2 water-air-fluxesmeasured by the LGR at B1 and Askö (in thevicinity of the chamber sites). Graphs of CH4versus CO2 concentrations from the LGRmeasurementsarepresentedinAppendix6.Anegative linear relationship is seen in allmeasurementsatB1,themeasurementatAsköin June and at Fifångsdjupet in August (r2 >0.98). Thus, increasing concentrations of CH4wereseenduringdecreasingconcentrationsofCO2. This could be interpreted as a CH4productionoccurringduringuptakeofCO2.Thesecond component in the hydrogenotrophicmethano-genesisisCO2andthisresultcouldbean indication of this pathway. However, bothmeasurementsmade at Julaftons Fyr and theSeptember measurement at Fifångsdjupetshowed the opposite: increasing CO2 with


37

increasingCH4(r2>0.98).Thiscansuggestthatthereisactuallynotatruerelationshipbetweenthegases,butinsteadthattheconcentrationofboth gases changes linearly independently ofeachother.

4.4Finalcomments

Onealternativenotdiscussed in this report ismethane production in aggregates ofphytoplankton.Withintheseaggregates,anoxicconditions are formed, thus creating anenvironment in the surface water of whichmethaneproductioncanoccur(Grossartetal.,2011). During algal blooms, the conditions ofthis kindofproductionwould increasedue tolarger amount of phytoplankton in thewater.However, it would be assumed that theproductionwouldbehigherindarkconditionswhen photosynthesis is shut off, giving lessoxygen in thewater. Though, that contradictswhat isbelievedtobefoundhere.Onewouldalso consider a higher amount of methaneproductioninthesedimentduringalgalblooms,whichcouldexplain thehigherconcentrationsfound inAugust.Due to the larger amountofcyanobacteria decomposing at the seafloorduringtheblooms,conditions forhigherratesofmethaneproductionaregiven.Itshouldbeemphasisedthecomplexityofthesubject that is investigated in this report. ThedissolvedCH4inthewateriseffectedbyvariousparameter, changing in both time and space:biological activity, the structure of the watercolumn,windconditions,temperature,salinity,water currents, fluxes from the sediment aswellastotheatmosphere,accesstoproductioncomponents and soon. To therefore create abroad understanding of the subject, a lot ofparametersareneededtobeinconsideration.

5.ConclusionsThisstudyhasshownthatthesurfacewaterattheareaaroundAsköhavehighconcentrationsofmethaneandisasourcetotheatmosphere.

It is also shown that thewater south of Askö(B1)musthavehadaCH4sourcenotonlyfroma vertical upward transport of CH4 diffusingfrom the sediment. A suggestion of anothersource is CH4 ebullition from the sediment intheshallowerwatersatthesoutherncoastlineof Askö (chamber sites) with water currentspossibly transporting thegas laterally towardsthe open coastal waters at B1. It is alsosuggested that a CH4 production could occurunderlightconditionsinthemixedlayeratB1and around the chamber sites, a productionpresumablyhigherduringthealgalbloom.Thelatter supports the main hypothesis for thisstudy:amethaneproductionintheoxygenatedsurfacewaterthatsignificantlycontributestoaCH4water-air-fluxandthatincreasesduringthealgalbloom.InthewatersofthenorthernsideofAsköinthemoreenclosedbasinatFifångsdjupet,averticalupward transport of CH4 is reasonable as asingle source to the surface water in August.However, the amount of transport must hadbeenhighenoughtosustaintheconcentrationincreaseseenduringthesummer.Ahypothesisfor a second methane source is thereforepresented, in situebullition from the 600 gasseepagesitesattheseafloorinthenearbyareawith bubbles reaching the mixed layer. Ifebullitionwould be the only source of CH4 tothemixedlayer,atotalseepageflowtosustainthe CH4 concentration from June into Augustwas estimated to10 - 225bubbles sec-1 site-1with 4% of these bubbles reaching themixedlayer. In comparison to other studies, thesenumbers are fairly reasonable. However, todraw any further conclusion regardingebullitionasasource,theflowattheseepagesites at Fifångsdjupet would need to beinvestigated.

AcknowledgementsFirst and foremost, I would like to give mygreatest gratitude to my supervisors VolkerBrüchert at the Department of Geological


38

Sciences at Stockholm University and HellePlougattheDepartmentofMarineSciencesattheUniversityofGothenburg.Yourknowledge,guidance and enthusiasm have made thisbachelor thesispossibleandhaveencouragedmetoexplorewhatformewasaquiteunknowntopic. Iamespecially thankful toVolker forallthevastamountofe-mailsand longmeetingsthat always provided me with insights. I amgrateful for all the help and support that thestaffatAsköLaboratoryprovidedmeduringmyvisitstothestationandoutonthewaterwiththeresearchvesselR/VLimanda,inbothsunnyandrainyweatherconditions.Iamalsogratefulto my loving Toni Axelsson, who without anyquestiongavehis lastweekofvacationtothetaskof beingmy field assistance andhas alsosupportedme throughout the entire process.And Jonas Fredriksson,whohelpedme in thefieldinSeptemberandjoinedmeinthestruggleof rowing a boat for 90 minutes during acreakingsoundfromtheoars.IwouldalsoliketothankHeikeSiegmundattheStableIsotopeLaboratory (SIL) at StockholmUniversity, whowith enthusiasm supported me during theisotopeanalysis. I alsowant to thank JoachimJansen, Lena Kautsky and other staff atStockholmUniversitythatinvariouswayshavehelpedme.Lastly,Iwouldliketothankfriendsandfamilythathaveshowninterestandjoyforthisbachelorthesisduringtheprocess.

ReferencesAngelis,M.A.D.,Lee,C.,1995.Methaneproductionduring

zooplanktongrazingonmarinephytoplankton.Oceanogr.Lit.Rev.7,539.

Axell,L.B.,1998.OnthevariabilityofBalticSeadeepwatermixing. J. Geophys. Res. Oceans 103, 21667–21682.https://doi.org/10.1029/98JC01714

Bange, H.W., Bartell, U.H., Rapsomanikis, S., Andreae,M.O., 1994. Methane in the Baltic and NorthSeas and a reassessment of the marineemissions of methane. Glob. Biogeochem.Cycles 8, 465–480.https://doi.org/10.1029/94GB02181

Berg,A.,Lindblad,P.,Svensson,B.H.,2014.Cyanobacteriaasasourceofhydrogenformethaneformation.World J. Microbiol. Biotechnol. 30, 539–545.doi:10.1007/s11274-013-1463-5

Björck,S.,1995.AreviewofthehistoryoftheBalticSea,13.0-8.0 ka BP. Quat. Int. 27, 19–40.

https://doi.org/10.1016/1040-6182(94)00057-C

Bothe,H.,Schmitz,O.,Yates,M.G.,Newton,W.E.,2010.NitrogenFixationandHydrogenMetabolisminCyanobacteria. Microbiol. Mol. Biol. Rev. 74,529–551.https://doi.org/10.1128/MMBR.00033-10

Bousquet, P., Ciais, P., Miller, J.B., Dlugokencky, E.J.,Hauglustaine, D.A., Prigent, C., Van der Werf,G.R., Peylin, P., Brunke, E.-G., Carouge, C.,Langenfelds,R.L.,Lathière,J.,Papa,F.,Ramonet,M.,Schmidt,M.,Steele,L.P.,Tyler,S.C.,White,J., 2006. Contribution of anthropogenic andnatural sources to atmospheric methanevariability. Nature 443, 439–443.https://doi.org/10.1038/nature05132

Carstensen, J., Andersen, J.H., Gustafsson, B.G., Conley,D.J., 2014. Deoxygenation of the Baltic Seaduring the last century. Proc. Natl. Acad. Sci.111, 5628–5633.https://doi.org/10.1073/pnas.1323156111

Clark, I.,2015.GroundwaterGeochemistryandIsotopes.Taylor&FrancisGroup.

Codispoti,L.A.,Brandes,J.A.,Christensen,J.P.,Devol,A.H.,Naqvi, S.W.A., Paerl, H.W., Yoshinari, T., 2001.The oceanic fixed nitrogen and nitrous oxidebudgets: Moving targets as we enter theanthropocene?Sci.Mar.65,85–105.

Conrad, R., 2005. Quantification of methanogenicpathways using stable carbon isotopicsignatures: a review and a proposal. Org.Geochem., Stable isotope applications inmethane cycle studies 36, 739–752.https://doi.org/10.1016/j.orggeochem.2004.09.006

Damm,E.,Helmke,E.,Thoms,S.,Schauer,U.,Nöthig,E.,Bakker, K., Kiene, R.P. 2010. Methaneproductioninaerobicoligotrophicsurfacewaterin the central Arctic Ocean. Biogeosciences7:1099–1108

Deppenmeier, U., Müller, V., Gottschalk, G., 1996.Pathways of energy conservation inmethanogenic archaea. Arch. Microbiol. 165,149–163.https://doi.org/10.1007/BF01692856

Dlugokencky, E.J., P.M. Lang, A.M. Crotwell, J.W.Mund,M.J. Crotwell, and K.W. Thoning (2017),Atmospheric Methane Dry Air Mole FractionsfromtheNOAAESRLCarbonCycleCooperativeGlobal Air Sampling Network, 1983-2016,Version: 2017-07-28, Path:ftp://aftp.cmdl.noaa.gov/data/trace_gases/ch4/flask/surface/

Ferry, J.G., 1993a.Methanogenesis: Ecology, Physiology,Biochemistry & Genetics, Chapman & HallMicrobiologySeries.Springer.

Ferry,J.G.,1993b.FermentationofAcetate.SpringerLink304–334. https://doi.org/10.1007/978-1-4615-2391-8_7

Gelwicks, J.T., Risatti, J.B., Hayes, J.M., 1994. Carbonisotope effects associated with aceticlasticmethanogenesis. Appl. Environ. Microbiol. 60,467–472.


39

Grossart,H.-P., Frindte,K.,Dziallas,C., Eckert,W.,Tang,K.W., 2011. Microbial methane production inoxygenated water column of an oligotrophiclake. Proc. Natl. Acad. Sci. 108, 19657–19661.https://doi.org/10.1073/pnas.1110716108

Gülzow,W.,Rehder,G.,Schneiderv.Deimling,J.,Seifert,T., Tóth, Z., 2013. One year of continuousmeasurementsconstrainingmethaneemissionsfromtheBalticSea to theatmosphereusingashipofopportunity.Biogeosciences10,81–99.https://doi.org/10.5194/bg-10-81-2013

Hansson,M.,2006.CyanobakterieblomningariÖstersjön,resultat från satellitövervakning 1997-2005.SMHI.

Ho,D.T.,Bliven,L.F.,Wanninkhof,R.,Schlosser,P.,1997.The effect of rain on air-water gas exchange.Tellus B 49, 149–158.https://doi.org/10.1034/j.1600-0889.49.issue2.3.x

Hong,G.-F.,1992.TheNitrogenFixationanditsResearchin China, 1st ed. Springer-Verlag BerlinHeidelberg.

IPCC, 2013: Climate Change 2013: The Physical ScienceBasis. Contribution of Working Group I to theFifth Assessment Report of the Intergovern-mentalPanelonClimateChange [Stocker,T.F.,D. Qin, G.-K. Plattner, M. Tignor, S.K. Allen, J.Boschung, A. Nauels, Y. Xia, V. Bex and P.M.Midgley (eds.)]. Cambridge University Press,Cambridge,UnitedKingdomandNewYork,NY,USA, 1535 pp.http://www.climatechange2013.org/images/report/WG1AR5_ALL_FINAL.pdf

Johnson, K.M., Hughes, J.E., Donaghay, P.L., McN.Sieburth,J.1990.Bottle-CalibrationStaticHeadSpace Method for the Determination ofMethane Dissolved in Seawater. AnalyticalChemistry,vol.62,no.21,2408-2412

Karl, D.M., Beversdorf, L., Björkman, K.M., Church,M.J.,Martinez, A., Delong, E.F., 2008. Aerobicproductionofmethaneinthesea.Nat.Geosci.1,473–478.https://doi.org/10.1038/ngeo234

Karl,D.M.,Tilbrook,B.D.,1994.Productionandtransportof methane in oceanic particulate organicmatter. Nature 368, 732–734.https://doi.org/10.1038/368732a0

Klawonn,I.,Nahar,N.,Walve,J.,Andersson,B.,Olofsson,M.,Svedén,J.B.,Littmann,S.,Whitehouse,M.J.,Kuypers,M.M.M.,Ploug,H.,2016.Cell-specificnitrogen- and carbon-fixationof cyanobacteriain a temperate marine system (Baltic Sea).Environ. Microbiol. 18, 4596–4609.https://doi.org/10.1111/1462-2920.13557

Knox, M., Quay, P.D., Wilbur, D., 1992. Kinetic isotopicfractionation during air-water gas transfer ofO2,N2,CH4, andH2. J.Geophys.Res.Oceans97, 20335–20343.https://doi.org/10.1029/92JC00949

Kustmätsystem. 2017. Askö.https://www.smhi.se/kustmatsystem/asko(From the dates given in the tables of theappendices)

Lee,Z.,Weidemann,A.,Kindle,J.,Arnone,R.,Carder,K.L.,Davis, C., 2007. Euphotic zone depth: Itsderivation and implication to ocean-colorremote sensing. J. Geophys. Res. Oceans 112,C03009.https://doi.org/10.1029/2006JC003802

Mah, R.A., Ward, D.M., Baresi, L., Glass, T.L., 1977.Biogenesis of Methane. Annu. Rev. Microbiol.31, 309–341.https://doi.org/10.1146/annurev.mi.31.100177.001521

Morimoto,S.,Fujita,R.,Aoki,S.,Goto,D.,Nakazawa,T.,2017.Long-termvariationsofthemolefractionand carbon isotope ratio of atmosphericmethane observed at Ny-Ålesund, Svalbardfrom 1996 to 2013. Tellus B Chem. Phys.Meteorol. 69, 1380497.https://doi.org/10.1080/16000889.2017.1380497

Nisbet,E.G.,Dlugokencky,E.J.,Manning,M.R.,Lowry,D.,Fisher,R.E.,France,J.L.,Michel,S.E.,Miller,J.B.,White, J.W.C., Vaughn, B., Bousquet, P., Pyle,J.A., Warwick, N.J., Cain, M., Brownlow, R.,Zazzeri,G.,Lanoisellé,M.,Manning,A.C.,Gloor,E.,Worthy, D.E.J., Brunke, E.-G., Labuschagne,C., Wolff, E.W., Ganesan, A.L., 2016. Risingatmosphericmethane:2007–2014growthandisotopic shift. Glob. Biogeochem. Cycles 30,2016GB005406.https://doi.org/10.1002/2016GB005406

Penger, J., Conrad, R., Blaser, M., 2012. Stable CarbonIsotope Fractionation by MethylotrophicMethanogenic Archaea. Appl. Environ.Microbiol. 78, 7596–7602.https://doi.org/10.1128/AEM.01773-12

Reeburgh, W.S., Heggie, D.T., 1977. Microbial methaneconsumption reactions and their effect onmethanedistributionsinfreshwaterandmarineenvironments1. Limnol. Oceanogr. 22, 1–9.https://doi.org/10.4319/lo.1977.22.1.0001

Sarmiento, J., Gruber, N., 2006. Ocean BiogeochemicalDynamics. Princeton University Press, NewJersey.

Sawicka, J.E., Brüchert, V., 2017. Annual variability andregulation of methane and sulfate fluxes inBalticSeaestuarinesediments.Biogeosciences14, 325–339. https://doi.org/10.5194/bg-14-325-2017

Schmale, O., Schneider von Deimling, J., Gülzow, W.,Nausch, G., Waniek, J.J., Rehder, G., 2010.DistributionofmethaneinthewatercolumnoftheBalticSea.Geophys.Res. Lett.37, L12604.https://doi.org/10.1029/2010GL043115

Schneider von Deimling, J., Rehder, G., Greinert, J.,McGinnnis, D.F., Boetius, A., Linke, P., 2011.Quantification of seep-related methane gasemissionsatTommeliten,NorthSea.Cont.ShelfRes. 31, 867–878.https://doi.org/10.1016/j.csr.2011.02.012

Schmale, O., Greinert, J., Rehder, G., 2005. Methaneemissionfromhigh-intensitymarinegasseepsinthe Black Sea into the atmosphere. Geophys.


40

Res. Lett. 32, L07609.https://doi.org/10.1029/2004GL021138

Schulz, H.D., 2006. Quantification of Early Diagenesis:Dissolved Constituents in Pore Water andSignals in the Solid Phase, in: MarineGeochemistry.Springer,Berlin,Heidelberg,pp.73–124. https://doi.org/10.1007/3-540-32144-6_3

Scranton, M.I., Brewer, P.G., 1977. Occurrence ofmethane in the near-surface waters of thewestern subtropical North-Atlantic. Deep SeaRes. 24, 127–138.https://doi.org/10.1016/0146-6291(77)90548-3

SGU (Geological Survey of Sweden). 2015. MarineGeology, vector: resolution 1:100 000.https://maps.slu.se/get(Hämtat2016-10-22)

SMHI(SwedishMeteorologicalandHydrologicalInstitute).2017.Nyhetsarkiv:Sommarensalgblomningvartydligast i norra Östersjön.https://www.smhi.se/nyhetsarkiv/sommarens-algblomning-var-tydligast-i-norra-ostersjon-1.125392(From2018-01-03)

Stal, L.J., Albertano, P., Bergman, B., Bröckel, K. von,Gallon, J.R., Hayes, P.K., Sivonen, K., Walsby,A.E.,2003.BASIC:BalticSeacyanobacteria.Aninvestigation of the structure anddynamics ofwater blooms of cyanobacteria in the BalticSea—responses to a changing environment.Cont. Shelf Res., European Land-OceanInteraction 23, 1695–1714.https://doi.org/10.1016/j.csr.2003.06.001

Tang, K.W., McGinnis, D.F., Frindte, K., Brüchert, V.,Grossart, H.-P., 2014. Paradox reconsidered:Methane oversaturation in well-oxygenatedlake waters. Limnol. Oceanogr. 59, 275–284.https://doi.org/10.4319/lo.2014.59.1.0275

Thang, N.M., Brüchert, V., Formolo, M., Wegener, G.,Ginters, L., Jørgensen, B.B., Ferdelman, T.G.,2013. The Impact of Sediment and CarbonFluxesontheBiogeochemistryofMethaneandSulfur in Littoral Baltic Sea Sediments(Himmerfjärden,Sweden).EstuariesCoasts36,98–115. https://doi.org/10.1007/s12237-012-9557-0

Wang,F.K.,2001.Confidenceintervalforthemeanofnon-normaldata.Qual.Reliab.Eng.Int.17,257–267.https://doi.org/10.1002/qre.400

Wanninkhof, R., Asher, W.E., Ho, D.T., Sweeney, C.,McGillis, W.R., 2009. Advances in quantifyingair-seagasexchangeandenvironmentalforcing.Annu. Rev. Mar. Sci. 1, 213–244.https://doi.org/10.1146/annurev.marine.010908.163742

Wasmund,N.,Uhlig,S.,2003.PhytoplanktontrendsintheBaltic Sea. ICES J. Mar. Sci. 60, 177–186.https://doi.org/10.1016/S1054-3139(02)00280-1

Weiss,R.F.,1974.Carbondioxideinwaterandseawater:thesolubilityofanon-idealgas.Mar.Chem.2,203–215. https://doi.org/10.1016/0304-4203(74)90015-2

Weiss, R.F., Price, B.A., 1980. Nitrous oxide solubility inwater and seawater. Mar. Chem. 8, 347–359.https://doi.org/10.1016/0304-4203(80)90024-9

Whiticar, M.J., 1999. Carbon and hydrogen isotopesystematicsofbacterialformationandoxidationof methane. Chem. Geol. 161, 291–314.https://doi.org/10.1016/S0009-2541(99)00092-3

Whiticar, M.J., Faber, E., Schoell, M., 1986. Biogenicmethane formation in marine and freshwaterenvironments: CO2 reduction vs. acetatefermentation—Isotope evidence. Geochim.Cosmochim. Acta 50, 693–709.https://doi.org/10.1016/0016-7037(86)90346-7

Wiesenburg, D.A., Guinasso, N.L. 1979. EquilibriumSolubilitiesofMethane,CarbonMonoxide,andHydrogen inWater and SeaWater. Journal ofChemical andEngineeringData, vol. 24, no. 4,356-360.

Yamamoto,S.,Alcauskas,J.B.,Crozier,T.E.,1976.Solubilityofmethane in distilledwater and seawater. J.Chem. Eng. Data 21, 78–80.https://doi.org/10.1021/je60068a029


41

Appendix1A1.1.Resultstwo-sidedpairedt-testoftheincubationexperimentsamples

Theresultsfromthetwo-sidedpairedt-testperformedinRStudiofortheincubationexperimentpresentedinfigure9andtable5intheresultsaregivenbelowintableA1.1(B1)andtableA1.2(Fifångsdjupet).

A1.2Incubationexperiment:method

TheexperimentwasconducteddifferentlyinJunecomparedtoinAugustandSeptember,wherefigureA1.2demonstratesthedetails.Tocollectwatersamplesfortheincubationexperiment,theNiskinbottlewas used for two of the three experiments in June, to obtain samples containing low amount ofphytoplankton (as explained in2.2 Fieldmethods). A planktonnet (4.1µm)was used for theotherexperimentsinJuneattheAugAsköexperimentare,inAugustattheJuneAsköexperimentareaandinSeptemberatB1andFifångsdjupet.OnlydarkincubationexperimentswereconductedinJune,whileinAugustandSeptemberbothlightanddarkincubationsweremadetodistinguisheffectsoflightanddark conditions. In the latter, control samples were also made with filtered water, where allphytoplanktonwereremoved.Whencollectingsampleswiththeplanktonnet,thewaterwasfirstlypouredintoa2Lglassbottlewhenbeingoutinthefield.Thewaterfortheunfilteredincubationexperimentwaspouredthroughafunnelintoglassbottlesof110mLandfittedwithabutylstopperandaluminiumseal.Thefilteredsampleswerestreamedintothebottlesthroughasyringewitha0.45µmfilterattachedtoit.Thisremovedallthephytoplankton,whilebacteriaandarchaearemained.Thesamplesobtainedwiththeplanktonnet

t-value Df p-value Meandifference

CH4 276.58 2 1.307*10-5 2.7

δ13CCH4 -0.32172 2 0.7782 -0.32

CO2 22.267 2 2.011*10-3 140

N2O 22.66 2 1.942*10-3 4.1

t-value Df p-value Meandifference

CH4 59.477 2 2.826*10-4 3.7

δ13CCH4 -0.53092 2 0.6485 -0.53

CO2 17.656 2 3.192*10-3 130

N2O 23.422 2 1.818*10-3 4.7

TableA1.Resultsfromthetwo-sidedpairedt-testfromB1.Nullhypothesisthatthemeandifferenceisequalto0.Df=degreesoffreedom(n-1).

TableA2.Resultsfromthetwo-sidedpairedt-testfromFifångsdjupet.Nullhypothesisthatthemeandifferenceisequalto0.Df=degreesoffreedom(n-1).


42

wereincubatedfor24hours.TheJunesamplesfromtheNiskinbottlewerestoredinacoolingroomof15°Cfor5weeks,whilesamplesfromAugustandSeptemberwerestoredatthesurfacewaterbelowthe deck at the station at the Askö Laboratory. This was made to simulate similar environmentalconditionsastheconditionsofwheretheorganismswhereobtainedfrom.Toendtheincubation,1mLofZnCl2(zincchloride,50%concentration)wasadded.Theresultsfromtheplanktonnetmethodwerenotpresentedintheresultsofthisreportsincegaswaslostduringthesamplingandtransferringprocess.Nosampleswithstartingvaluesofconcentrationandisotopecompositionweremadeforthismethod,i.e.,samplesthatwerenotincubated.Noreferencevalueswerethereforeavailable tocompareanychanges.However, itwasassumedthat the filteredsampleswereinequilibriumwiththeatmosphere(concentrationCH4~2.6nM,CO2~12µM,N2O~7nM(SarmientoandGruber,2006),δ13CCH4nearatmosphericcompositionaround-50‰to-48‰).Theother samples transferred to the glass bottles through a funnel was assumed not being as muchequilibrated,butstillhavelostpartsofthegases.Though,noconfirmationofthiswasbeingabletobemade.

A1.3Incubationexperiment:results

TheresultsfromalloftheincubationexperimentsarepresentedinfigureA1.2.Noneoftheanalysedparametersinthefilteredsamplesshowedofexpectedvalues,asmentionedinthemethoddescription.TheCH4andN2Oconcentrationswerehigher,theCO2concentrationlowerandtheδ13CCH4waslighterthanexpected.Whencomparingtheresultsbetweenthemonthsforsamplesobtainedwiththesamemethod,nodistinguishingdifferenceswere seen.For theunfilteredsamples, theCH4concentrationwereinthesamerangeforbothlightanddarksamples,though,withaslightlyhighervalueinJuneatAsköexperimentarea:

• JunedarkAskö=14±2.3nM• Auglight=10±1.7nM• Augdark=12±1.0nM• SeplightB1=11±1.1nM• SepdarkB1=11±1.2nM• SeplightFifångsdjupet=10±0.75nM

FigureA1.2.Aschematicoverviewofhowtheincubationexperimentswereconductedforeachoccasion,whereleft=June,middle=Augustandright=September.


43

ThefilteredsampleinAugustwerelowerinCH4concentrationthantheexperimentfromFifångsdjupetinSeptember.However,itwasonlyasmalldifference:

• Aug=6.1±0.72nM• SepB1=8.1±1.8nM• SepFifångsdjupet=10±0.74nM

Theδ13CCH4valueswereinthesamerangeforallsamples(~-56‰to~-54‰),withvaluesslightlymoredepletedthanseeninthesurfacewaterofthestudysites.ExceptionofthisthoughwasatFifångsdjupetinSeptember,wheretheunfilteredsamplesprovidedvalueof-47.0±0.3‰andthefilteredsampleof-69.0 ± 0.5‰. As well the light experiment in August provided value of -51.9 ± 0.1‰. It must be

a)

c)

d)

FigureA1.3.ResultsfromtheconcentrationandisotopeanalysisoftheexperimentsamplesfromJune(left),August(middle)andSeptember(right)thatwerenotpresentedinthemainresults-section.Asko=Asköexperimentarea,Fif.=Fifångsdjupet,filt.=lightsamplesfilteredwith0.45µmfilter.a)CH4concentration(nM),b)isotopecompositionofδ13CCH4vs.VPDB(‰),c)CO2concentration(µM),d)N2Oconcentration(nM).Note:onlysinglesampleswereobtainedatFifångsdjupetinSeptember.

b)


44

mentionedthatduetotherelativelowconcentrationsofCH4intheincubationbottles,thereliabilityoftheresultsfromtheGC-IRMSwaslow.Testsperformedbythelaboratorystaffwithknownstandardsshowahighstandarddeviationatthelevelsofpeakareasthatthesesamplesshowed.Thisstandarddeviationisnotincludedintheerrorbarsinthegraph.Thoughafterconsideration,theseresultsarestill presented since the results from the duplicates and triplicates provided results in the samecompositionrange.However,thismighthaveskewedtheresultsatFifångsdjupetinSeptember,whereonlysinglesampleswereobtainedandnocomparisonofothersamplescanbemade.AllofthesamplesprovidedsimilarresultsinCO2andN2OconcentrationwithCO2rangebetween~70to~85µMandN2Obetween~2.1to~2.6nM.ThesevalueswereincloserangetotheJuneexperimentobtained from theNiskinbottle. Theonly smalldifferences seenwere lowerCO2values in the lightSeptemberexperiment fromFifångsdjupet (50±3.7µM)aswellashigherN2Oconcentration inthefilteredsamplefromB1andinthelightFifångsdjupetexperimentinSeptember(3.1±0.2nMforboth).Nofurtherinterpretationoftheresultismade,sincenoreferencesampleswereobtainedanditisnottrulyknownifthewaterwasinequilibriumwiththeatmosphereasassumed.


45

Appendix2Thecoefficientsusedinequation6whencalculatingtheBunsensolubilitycoefficientsforthegasesCH4,CO2andN2OarepresentedintableA2.

Gas A1 A2 A3 B1 B2 B3 Calculatedβ(molesL-1atm-1)

Reference

CH4 -68.8862 101.4956 28.7314 -0.076146 0.04397 -0.0068672 0.033461 WiesenburgandGuinasso(1979)

CO2 -58.0931 90.5069 22.294 0.027766 -0.025888 0.0050578 0.038187 Weiss(1974)

N2O -62.7062 97.3066 24.1406 -0.05842 0.033193 -0.0051313 0.027992 WeissandPrice(1980)

TableA2.CoefficientsfortheBunsensolubilitycoefficientβ(molesliter-1atm-1)ofthedrymolefractionofthegasesCH4,CO2andN2O.


46

Appendix3 TableA3.1-A3.3belowpresenttheinformationregardingthefieldconditionsduringthesamplingandmeasurementsatB1,Fifångsdjupet,Tallholmen,JulaftonsFyrandatAsköforeachoccasioninJune,AugustandSeptember.ThechlorophylldatawasgatheredfromthecoastalmeasurementsystematB1(Kustmätsystem,2017).TheremainingdatawasgatheredontheresearchvesselR/VLimanda.

SiteJune Date Time Temp.air(°C)

Temp.surf.water(°C)

Sal.(PSU)

Wind:speed(m/s)&direction

Airpress.(hPa)

Chlorophyll(µg/L)

Weather

B1 2017-06-28

09:00 15 12.9 6.3 2.8–4.0(ESE) 1011 0.69 Sunny

Fifångsdjupet 2017-06-28

11:10 15 15.8 6.3 3.5–4.5(ESE) 1012 0.69 -II-

JulaftonsFyr 2017-06-28

13:00 15 - - 3.6(SE) 1012 0.69 -II-

Askö(LGR) 2017-06-28

13:45 15 - - 3.3–7.2(ESE) 1012 - -II-

JuneAsköexp.area

2017-06-29

11:00 15 13.8 6.2 9.7(-) 1007 0.72 -II-

SiteSep Date Time Temp.air(°C)


Sal.(PSU)


Airpress.(hPa)

Chlorophyll(µg/L)

Weather

B1 2017-09-14

14:00 14 14.8 6.3 5.5–8.0(NNW) 985 2.4 Partlyrainandcloudy


10:45 13 14.9 6.3 5.0–6.5(NW) 997 2.1 Sunny


12:10 14 - - 3.4–6.0(NW) 998 2.1 Sunny

SiteAug Date Time Temp.air(°C)


Sal.(PSU)


Airpress.(hPa)

Chlorophyll(µg/L)

Weather

Askö(exp.area)

2017-08-02

09:00 20.3 18.5 5.9 2.0(-) 1009.5 2.7 Sunny

B1 2017-08-03

11:26 17 18.6 6.1 4.5–8.0(S-SSE) 1011 3.0 Sunny&windy


13:01 18 18.8 6.2 4.5–8.0(S-SSE) 1010 3.0 -II-


13:42 18 - - 4.5–8.0(S-SSE) 1010 3.0 -II-

TableA3.1.Samplingparametersobtainedduringthewatersampling,CTDandLGRmeasurementatthevarioussitesinJune.Temp=temperature,surf.=surface,sal.=salinity,press.=pressure.

TableA3.2.Samplingparametersobtainedduringthewatersampling,CTDandLGRmeasurementatthevarioussitesinAugust.Temp=temperature,surf.=surface,sal.=salinity,press.=pressure.

TableA3.3.Samplingparametersobtainedduringthewatersampling,CTDandLGRmeasurementatthevarioussitesinSeptember.Temp=temperature,surf.=surface,sal.=salinity,press.=pressure.


47

Appendix4TableA4.1-A4.3belowpresenttheinformationregardingthefieldconditionsduringtheairandwatersamplingsatthechambersitesforeachoccasioninJune,AugustandSeptember.Theairtemperature,windspeed,airpressureandchlorophylldatawasgatheredfromthecoastalmeasurementsystematB1(Kustmätsystem,2017).Thewatertemperatureandsalinityweremeasuredbyaconductivitymeter.

June Date Time Temp.air(°C)


Sal.(PSU)

Windspeed(m/s)

Airpress.(hPa)

Chlorophyll(µg/L)

Weather

Monday 2017-06-26

17:38,17:55

11.6 15.4,15.1 6.1,6.1

7.5 997.2 1.3 Sunny(rainearlierduringtheday)

Tuesday 2017-06-27

15:06,15:17

11.6 14.6,14.8 6.1,6.1

2.6 1007.5 0.82 Sunny

Wednesday 2017-06-28

17:10,17:30

15.0 14.4,14.8 6.1,6.1

5.7 1011.4 0.71 -II-

Thursdaymorning

2017-06-29

07:09,07:20

15.0 13.8,13.6 6.1,6.2

8.4 1009.3 0.71 -II-

Thursdayafternoon

2017-06-29

x,14.39

15.0 x,14.8 6.2 10.9 1006.2 0.82 -II-

Aug Date Time Temp.air(°C)


Sal.(PSU)

Windspeed(m/s)

Airpress.(hPa)

Chlorophyll(µg/L)

Weather

Monday 2017-07-31

15:22,14:55

20.4 19.1,20.2 5.9,5.8

4.7 1008.9 3.6 Sunny

Tuesday 2017-08-01

11:15,11:22

17.5 19.4,19.2 5.9,5.8

6.2 1015.3 2.5 Windy,abitcloudy

Wednesday 2017-08-02

09.43,09:52

18.5 19.2,19.4 5.9,5.9

2.0 1009.4 2.5 Sunny,alittlebitofrain

Thursday 2017-08-03

08:08,08:20

16.0 19.2,19.0 5.9,5.9

4.3 1011.5 2.1 Sunny(rainduringthewedn.-thurs.-night)

Friday 2017-08-04

07:59,08:07

16.1 19.3,18.8 5.9,5.8

6.4 998.2 2.7 Sunny&windy

Sep Date Time Temp.air(°C)


Sal.(PSU)

Windspeed(m/s)

Airpress.(hPa)

Chlorophyll(µg/L)

Weather

Thursday 2017-09-14

19:00,19:30

12.5 14.9,14.8 6.2,6.2

3.5 989.7 2.1 Rainonandoffduringtheday

Fridaymorning

2017-09.15

09:03,09:10

13.2 14.4,14.4 6.2,6.2

3.0 996.5 1.9 Sunny

Fridayafternoon

2017-09.15

14:44,14:55

15.5 14.9,14.8 6.2,6.2

1.4 999.4 2.2 -II-

TableA4.1.SamplingparametersobtainedduringtheairandwatersamplingatthechambersitesinJune.Temp=temperature,surf.=surface,sal.=salinity,press.=pressure,x=nosamplingmadeatchamber1.

TableA4.2.SamplingparametersobtainedduringtheairandwatersamplingatthechambersitesinAugust.Temp=temperature,surf.=surface,sal.=salinity,press.=pressure.

TableA4.3.SamplingparametersobtainedduringtheairandwatersamplingatthechambersitesinSeptember.Temp=temperature,surf.=surface,sal.=salinity,press.=pressure.


48

Theresultsinppmday-1fromoftheanalysisofCH4,CO2andN2OofalloftheairsamplesobtainedfromthechambersitesarepresentedfigureA4.1,eventheonesnotusedinthewater-air-fluxcalculations.Ifnosignificantdifferencewasseenbetweenthesamplesor if thestartingvalueofthesamplewasskewed(e.g.thefirstsampleofCO2inJune1stwith~600ppmintheatmosphere),nocalculationwasmade.Themeasurementresultsofppmday-1fromtheLGRmeasurementsfromeachsiteofCH4arepresentedinfigureA4.2andofCO2infigureA4.3.

FigureA4.1.TheresultsfromtheairsampleanalysisofCH4(top),CO2(middle)andN2O(bottom)fromeachsampleobtainedatchamber1(left)andchamber2(right)inJune,AugustandSeptember.Thecirclesshowwhichsamplesthatwereusedinthecalculationofthefluxes.However,thethirdsampleofCO2inJune2

nd(lightblue)wasnotincludedinthefluxcalculation.NosignificantfluxesweremeasuredofN2Oatall.Sharedlegendforallgraphs.


49

FigureA4.2.TheresultsfromtheLGRmeasurementofCH4(ppmday-1)inJune(a-e), August (f-g) and September (h-j). The black lines show the truemeasurementvalues,whilethered linesshowtheaverageslope(averagek-value)obtainedfromthelinearregressionanalysis.Theequations(y=kx+m)andther2-valuesabovethegraphsareassociatedwiththeredslopes,withallvaluesofr2closeto1(indicationofclosetoperfectlinearity).Themaximumandminimumk-value(CI=95%)arenotdisplayedinthegraphs.

a) b) c)

d) e) f)

g) h) i)

j)


50

FigureA4.3.TheresultsfromtheLGRmeasurementofCO2(ppmday-1)inJune(a-e), August (f-g) and September (h-j). The black lines show the truemeasurementvalues,whilethebluelinesshowtheaverageslope(averagek-value)obtainedfromthelinearregressionanalysis.Theequations(y=kx+m)andther2-valuesabovethegraphsareassociatedwiththeblueslopes,withallvalues of r2 close to 1 (indication of close to perfect linearity) except atTallholmen in June.No flux-calculationwasmade at Tallholmen due to thisresult.Themaximumandminimumk-valuesoftheotherresults(CI=95%)arenotdisplayedinthegraphs.

a) b) c)

d) e) f)

g) h) i)

j)


51

Appendix5

A5.1Hypothesis2.1:ebullitioninsituatFifångsdjupet

Themethod to calculate thenumberofbubbles sec-1 site-1needed to reach to themixed layerarepresentedhere.TheconcentrationinthesurfacewateratFifångsdjupetwas18±1.9nMinJuneand22 ± 2.5 nM in August. Therefore, the concentration of a steady-state was set to 20 nM in thecalculation.Assumingasteady-state,thebubbleCH4inputtothemixedlayerwillbethesameastheoutput of CH4 from themixed layer. To simplify the calculation, the only output considered to berequiredwasthewater-air-flux.OxidationwasnotconsideredasaCH4outputsincethedecreaseofCH4intheincubationexperimentswasconsideredtobenegligible.AlateraltransportofCH4outfromtheareawasassumedtobeequaltotheamountintransportintotheareaandwasneglectedaswell.ThefluxwasthereforerecalculatedasCH4emissions(molesday-1)forawatersurfacecovering7km2(2x3.5km,theareaofFifångsdjupetwhichincludedtheseepagesitesatTallholmen):

E=F*10-3 *A

M (eq.A5.1)

E=CH4emission(molesday-1)F*10-3=resultsoftheCH4water-air-flux(gm

-2day-1)A=area(m2)ofthewatersurfaceatFifångsdjupet(2000*3500m)M=molecularweightofCH4(16gmole-1)Secondly,thenumberofmolesofCH4perbubblewascalculated.Thiswasdoneforbubblediametersofd=2mm,d=3mm,d=4mm,d=5mmandd=6mmandwherethepressurewasassumedtobe2.5barsattheseafloorof24mdepth,withatemperatureof7°Candwithassumptionthatthebubblesdissolvedcompletelyinthemixedlayer:

nx=P*VxR*T

(eq.A5.2)

nX=numberofmolesofCH4perbubblewithxdiameterP=pressureattheseafloor(Pa)VX=volumeofthebubblewithxdiameter(m3)R=8.314(Pa*m3)(moles*K)-1

T=temperatureattheseafloor(K)Thereafter,thetotalnumberofbubbleswithxdiametersreachingthemixedlayerperwascalculatedusingtheCH4emissionasCH4input:

y=E

nx (eq.A5.3)

y=totalnumberofbubblesday-1Thirdly,thenumberofbubblessec-1site-1reachingthemixedlayerwascalculated:

yS=y

s*24*60*60 (eq.A5.4)

yS=totalnumberofbubblessec-1site-1

s=totalnumberofsites(600)


52

Lastly,thetotalseepageflowofbubblessec-1site-1wascalculated:

yT=ySf (eq.A5.5)

yT=totalseepageflow(bubblessec

-1site-1)f=fractionofnumberofbubblesreachingthemixedlayer(0.04)

A5.2Calculation inhypothesis2.2:ebullitionatshallowerareaswith lateral inputofsurfacewaterstoB1

The isotope composition of the fluxes was calculated to test hypothesis 2.2 too see if the fluxcompositionsdeviatedfromthecompositionofthesurfacewaters.Thecalculationwasmadewiththeequation:

δ13CF=δ13CTOT-δ

13CX*fXfF

(eq.A5.6)

δ13CF=calculatedisotopecomposition(‰)ofthefluxfF=fractionoftheCH4molesinthechamberaddedbytheflux(calculatedfromthemeasuredconcentrations)δ13CX=measuredisotopecomposition(‰)inthefirstairsampleobtainedfromthechamber(figure11)fX=fractionoftheCH4molescontainedinthechamberatthetimewhenthefirstairsamplewasobtainedδ13CTOT=measuredisotopecomposition(‰)inthesecondairsampletakenfromthechamber(figure11)ThechamberswereconsideredasclosedsystemswithCH4inputonlyfromfluxfromthewaterbelow.ThesurfacewaterwasseenasanopensystemwithCH4inputfromthesurroundingwater,leadingtoastable valueof isotope composition throughout themeasurement time. Therefore, theδ13CF valueswerecomparedwiththeisotopecompositionofthesurfacewatertoobtainthedifferenceinisotopecomposition(∆13CCH4).Infigure18,thevariousparameterinequationA5.6canbedistinguished.

A5.3Fluxvs.daytime/nighttimeinHypothesis2.2:ebullitionatshallowerareaswithlateralinputofsurfacewaterstoB1

Thefluxesfromthechambersitesweregroupedandusedinthetestingofhypothesis2.2,whereitwasthought that ebullition could occur at the shallow sites of Julaftons Fyr and the chamber sites. Norelationshipbetweendaytime/nighttimeandamountoffluxwerenotedaspresentedinfigureA5.3.1.InJune,thefluxesoccurringduringonlynighttime ingroupNwere lowerthantheonesduringdaytime.However,inSeptembertheoppositewasseen.NomeasurementsweremadeonlyduringnighttimeinAugust.


53

A5.4Leakagesuspicion

Here are further explanation regarding the assumption that there might have been leakage fromchamber2inAugustingroupDN2.Therewasaheaviercompositionfoundinthechamberatday2(3rdsampling) compared to day 1 (2nd sampling) (figure 11), which contradicts the other results ofcontinuingly lighter composition with time. If leakage had occurred, that would in return provideskewnessintheresultsinfigure17inthediscussionregardinghypothesis2.2,wheretheresultsshowedofheaviercompositionoftheCH4fluxthanofCH4inthesurfacewater.However,indicationofleakageis not seenwhenobserving the concentration results fromeachday (Appendix 4: figureA4.1: CH4:chamber2),wheretheconcentrationwas~3timesashighatday2(141ppm)comparedtoday1(44ppm). Itwas therefore calculated according to the equationbelowan estimationof the amount ofleakagethatcouldhaveoccurredifonlyconsideringtheisotopecomposition:

δ13CTOT=δ13CY*fY+δ13CX*fX+δ13CZ*fZ (eq.A5.7)

δ13CTOT=theaverageisotopecomposition(‰)inthechamberatday2(-58.6‰)δ13CY=isotopeaveragecomposition(‰)inthechamberatday1(-59.5‰)fY=fractionoftheCH4molesinthechamberatday1(0.30)δ13CX=isotopeaveragecomposition(‰)oftheatmosphereinAugust(-52.7‰)fX=fractionoftheCH4molesaddedfromapossibleleakage,i.e.,molesaddedfromtheatmosphere(tested)δ13CZ=thecalculatedfluxisotopecomposition(‰)fromday0-1atchamber2inAugust(groupDN1=-59.8‰)fZ=fractionoftheCH4molesaddedtheflux(tested)Thechamberairatday1contributedwith30%ofthetotalnumberofmoleswithinthechamberatday2, according to the concentration results (44 ppm/141 ppm * 100). The fraction added from theatmosphereandtheflux(assumingsamefluxcompositionascalculatedforday0-1)weretested.Whentestingvariouscontributingfractions,isseenthatapproximately55%couldhavebeenaddedfromthefluxand15%fromtheatmosphere.Thismeansthat is itpossiblethat15%ofthechamberaircouldhave been leaked between day 1 – 2 at chamber 2 in August. If that were the case, then the

FigureA5.3.1.GraphsoftheCH4fluxes(mgm-2day-1)atchamber1totheleftandchamber2totherightinthegroupeddaysN(nightonly),D(dayonly),DN1andDN2(dayandnight).June1st=thefirstmeasurementmadeinJune(Mon-Tues),June2nd=thesecondmeasurementmadeinJune(Wedn-Thurs).


54

concentration(ppm)measuredatday2wouldhavebeen~15%higherifnoleakagehadoccurred.Thiswouldhaveprovided~160ppminthechamberatday2insteadof141ppm.Thatwouldinreturnhaveprovidedatotalfluxatchamber2inAugustof5.2mgm-2day-1insteadofthe4.4mgm-2day-2,presentedinfigure11.Thisresultseemsfairlyreasonableandtheassumptionwastherebythatleakagecouldhaveoccurredsometimebetweenday1-2atchamber2inAugust.


55

Appendix6ThemeasurementresultsoftheCH4andCO2concentrations(ppm)fromtheLGRmeasurementswereplottedagainsteachotherandpresentedinfigureA6below.

Figure A6. The results from the LGR measurement of CH4 and CO2concentrations(ppm)plottedagainsteachother:June(a-e),August(f-g)andSeptember(h-j).Theredlinesshowthetruemeasurementvalues,while theblacklinesshowtheslopefromalinearregressionanalysis.Ther2-valuesabovethegraphsareassociatedwiththeredslopes.

a) b) c)

d) e) f)

g) h) i)

j)

bachelor thesis · 2018-06-08 · fanny axelsson bachelor thesis geochemistry 30 hp 5 the...

Documents