Interannual sedimentary effluxes of alkalinity in the southern North Sea

34 Pages·2017·11.49 MB·English

BiogeosciencesDiscuss.,https://doi.org/10.5194/bg-2018-58
ManuscriptunderreviewforjournalBiogeosciences
Discussionstarted:5February2018
c Author(s)2018.CCBY4.0License.
(cid:13)
Page 1
Interannual sedimentary effluxes of alkalinity in the southern North
Sea: Model results compared with summer observations
JohannesPa¨tscha,∗,WilfriedKu¨hna,KatharinaD.Sixb
aInstituteofOceanography,UniversityofHamburg,Germany
bMax-PlanckInstituteforMeteorology,Hamburg,Germany
Abstract
ForthesedimentsofthecentralandsouthernNorthSeadifferentsourcesofalkalinitygeneration
are quantified by a regional modelling system for the period 2000 - 2014. For this purpose a
formerlyglobaloceansedimentmodelcoupledwithapelagicecosystemmodelisadoptedtoshelf
seadynamicswheremuchlargerturnoverratesthanintheopenanddeepoceanoccurs. Totrack
alkalinitychangesduetodifferentnitrogen-relatedprocessestheopenoceansedimentmodelwas
extended by the state variables particulate organic nitrogen (PON) and ammonium. Directly
measured and from Ra isotope flux observationderived alkalinity fluxes from the sediment into
the pelagic are reproduced by the model system but calcite building and calcite dissolution are
underestimated. Bothfluxescanceloutintermsofalkalinitygenerationandconsumption. Other
simulated processesaltering alkalinity in the sediment like net sulfate reduction, denitrification,
nitrification and aerobic degradationare quantified and compare well with correspondingfluxes
derivedfromobservations. Mostofthesefluxesexhibitastrongpositivegradientfromtheopen
NorthSeatothecoastwherelargeriversdrainnutrientsandorganicmatter. Atmosphericnitrogen
depositionshowsalsoapositivegradientfromtheopenseatowardslandandsupportsalkalinity
generationinthesediments. Anadditionalsourceofspatialvariabilityisintroducedbytheuseof
a3D-heterogenousporosityfield. Duetorealisticporosityvariations(0.3-0.5)thealkalinityfluxes
varybyabout4%. Thestrongestimpactoninterannualvariationsofalkalinityfluxesexhibitthe
temporal varying nitrogeninputs fromlarge rivers directly governingthe nitrate concentrations
inthecoastalbottomwater,thus,providenitratenecessaryforbenthicdenitrification. Overthe
timeinvestigatedthealkalinityeffluxesdecreaseduetothedecreaseofthenitrogensupplybythe
rivers.
1. Introduction
1
Alkalinity generationfromanaerobic degradationin coastal sediments favorsthe marine uptake
2
3 capacityforatmosphericCO2. Thisisbecausethesepathsoforganicmatterdegradationinclude
∗Correspondingauthor
Emailaddress: [email protected](JohannesP¨atsch)
BiogeosciencesDiscuss.,https://doi.org/10.5194/bg-2018-58
ManuscriptunderreviewforjournalBiogeosciences
Discussionstarted:5February2018
c Author(s)2018.CCBY4.0License.
(cid:13)
Page 2
4 irreversibleprocesseslikeN2 productionandlossofreducedsulfateproducts.
InSeptember2011andJune2012Brenneretal.(2016)measuredalkalinityfluxesfromtheNorth
5
Sea sediment using severalsediment cores. For the southernNorth Sea they found a mean flux
6
7 of6.3mmolm−2d−1. Alkalinity effluxesintothe pelagicsystemcouldpartlydeterminethe rel-
ative high surface alkalinity concentrations (Fig. 1a) in the southern North Sea as observedby
8
Thomasetal.(2009)inSeptember2001. Togetherwithobservedconcentrationsofdissolvedinor-
9
ganiccarbon(DIC)(Bozecetal.,2006)thesesurfacealkalinityconcentrationscanbetranslated
10
11 into∆pCO2 values(pCO2ocean−pCO2atmosphere)whicharemainlyresponsiblefortheairseaex-
12 change of CO2 between ocean and atmosphere (Fig. 1b). In the southern North Sea positive
valuesindicateoversaturationandthusoutgassing,whereasinthenorthernpartsnegativevalues
13
14 result in an uptake of atmospheric CO2. When in a simple gedankenexperiment the observed
alkalinity fluxes by Brenneretal. (2016) would be reduced by 50 % from the beginning of the
15
year,thealkalinityconcentrationsespeciallyintheshallowsouthernNorthSeawouldbereduced
16
17 (Fig. 1c)andthecorresponding∆pCO2 valueswouldexhibitmuchstrongeroversaturation(Fig.
1d). ThissimpleexperimentignorestheseasonalityofthealkalinityfluxesandthefactthatDIC
18
fluxeswouldvaryinconcert.
19
In this paper we investigatethe variability of alkalinity generationand the efflux to the pelagic
20
zonebymeansofaregionalbiogeochemicalmodel. Thesecondchapterpresentsmethodsconcern-
21
ingthemodelsetup,particularwithregardtotheadaptationoftheformeropenoceansediment
22
model (Heinzeetal., 1999) to shelf sea conditions. Within the third chapter model results are
23
comparedwithobservationaldata. Inthefourthchapterweshowtheresultsofseveralscenarios
24
demonstratingthesensitivityofthetotalmodeldynamicsonenvironmentalsettingsduetochang-
25
ing alkalinity fluxes. One of these scenariospicks up the gedankenexperimentmentionedabove.
26
27 ItdemonstratesthestrongimpactofreducedalkalinityfluxesonthepCO2 (seeChapter4.2).
2. Methods
28
The simulationswereperformedwith the ecosystemmodel ECOHAM (Pa¨tschandKu¨hn, 2008)
29
using the nesting method focussing on the central and southern North Sea (50.88o to 57.28oN,
30
3.42oW to 9.25oE) (Pa¨tschetal., 2010). The model system includes the hydrodynamic model
31
HAMSOM(Backhaus,1985;Pohlmann,1996;Pa¨tschetal.,2017)andtheverticallyresolvedsed-
32
imentmodeloriginallydevelopedforthedeepopenocean(Heinzeetal.,1999). Thelattermodel
33
hasbeenadoptedtoshelfseadynamics,detailsarediscussedbelow. The3Dfieldsoftemperature
34
(T),salinity(S),advectiveflowandverticalturbulentmixingcoefficientscalculatedbyHAMSOM
35
areusedasforcingforECOHAM.ThetimestepofECOHAMis5minutes.
36
BiogeosciencesDiscuss.,https://doi.org/10.5194/bg-2018-58
ManuscriptunderreviewforjournalBiogeosciences
Discussionstarted:5February2018
c Author(s)2018.CCBY4.0License.
(cid:13)
Page 3
2.1. Thehydrodynamic Model
37
The 3D fields of temperature, salinity, advective flow and vertical turbulent mixing coefficients
38
calculated by the hydrodynamic model HAMSOM are used as forcing for the pelagic biogeo-
39
chemical model ECOHAM. HAMSOM is a baroclinic, primitive equation model using the hy-
40
drostatic and Boussinesq approximations. The current velocities are calculated using a first or-
41
der component-upstream scheme. The horizontal is discretized on a staggered Arakawa C-grid
42
(ArakawaandLamb,1977)witharesolutionof∆λ=1/3oand∆φ=1/5o.
43
In a first step the model was applied on a larger area including the Northwest European Shelf
44
(15.250oW-14.083oE,47.583oN-63.983oN)(Lorkowskietal.,2012). Forthislarge-domainrun
45
sea surfaceelevations ofthe semi-diurnallunar tide M2 wereprescribedatthe openboundaries
46
(Backhaus,1985). Thecorrespondingresultsoftemperature, salinityandsurfaceelevationwere
47
stored on the boundaries of the smaller model domain used in this study (black lines in Fig.
48
2). ThesedatawereusedasboundaryconditionsforthehydrodynamicmodelHAMSOMimple-
49
mentedonthesmallerdomainwithaverticalresolutionof5mintheupper50mandincreasing
50
resolutionbelow. The M2-tideis thus inducedimplicitely bythe prescribedsurfaceelevationat
51
theboundaries. DetailsofthenestingprocedurecanbefoundinSchwichtenberg(2013).
52
2.2. Thepelagic biogeochemical Module
53
In the same way as for the hydrodynamic model in a first step the biogeochemical model ran
54
onthelargermodeldomainandprovidedboundaryconditionsforthemodelonthesmallergrid
55
(Fig.2).
56
Thepelagicbiogeochemicalmodelincludes4nutrients(nitrate,ammonium,phosphate,silicate),
57
two phytoplanktongroups(diatoms andflagellates), twozooplanktongroups(micro- andmeso-
58
zooplankton),bacteria,twofractionsofdetritus(fastandslowlysinking),labiledissolvedorganic
59
matter,semi-labileorganiccarbon,oxygen,calcite,dissolvedinorganiccarbon,andtotalalkalinity.
60
Calcite formationis performedby flagellates, only. The molarratio of softtissue productionto
61
calciteproductionis10:1. Themodeldifferentiatesbetweennormalexudationbyphytoplankton,
62
the result of which is labile dissolved organic matter with Redfield composition corresponding
63
to the Redfield production, and an excess exudation of semi-labile organic carbon. The pelagic
64
moduleisdescribedindetailinLorkowskietal.(2012). Forthisstudyweincludedtheprognostic
65
alkalinitycalculationfromSchwichtenberg(2013). Thedifferentprocesses(Fi)andtheirinfluence
66
onalkalinityare:
67
F01-calcitedissolution
68 •
F02-calciteformation
69 •
F03-nitrification
70 •
BiogeosciencesDiscuss.,https://doi.org/10.5194/bg-2018-58
ManuscriptunderreviewforjournalBiogeosciences
Discussionstarted:5February2018
c Author(s)2018.CCBY4.0License.
(cid:13)
Page 4
F04-uptakeofnitrate
71 •
F05-releaseofammonium
72 •
F06-uptakeofammonium
73 •
F07-atmosphericdepositionofammonium
74 •
F08-atmosphericdepositionofnitrate
75 •
F09-uptakeofphosphate
76 •
F10-releaseofphosphate
77 •
Thesefluxesdeterminethechangeofalkalinity:
78
∂TA
=2(F01 F02 F03)+F04+F05 F06+F07 F08+F09 F10 (1)
∂t − − − − −
79
80
TogetherwiththedynamicsedimentmodulewhichexchangesTAandDICwiththepelagicsystem
81
itwaspossibletosimulatethefullcarbonatesystemprognostically.
82
2.3. TheSedimentModule
83
2.3.1. Theopenoceansedimentmodel
84
The original sediment model was developed by Heinzeetal. (1999) for the global ocean. This
85
model simulated accumulation, degradationand burial of particulate organicand shell material
86
and a diffusive pore water exchange with the overlaying ocean. It was applied mainly for the
87
deepoceanwithitslowamountsofincomingparticulatemattercomparedtotheshallowshelfsea
88
export. Thecorrespondingtimescalesoffluxvariationswereratherlarge(annualtodecadal)and
89
showednoseasonalsignal. Thismodelincludedthesolidcomponentsparticulateorganicmatter
90
(POM),calcite,opalandsiltexportedfromthepelagicandthedissolvedcomponentsphosphate
91
92 (PO4),dissolvedinorganiccarbon(DIC),alkalinity(TA),silicate(Si(OH)4),nitrate(NO3),oxy-
93 gen(O2),anddinitrogen(N2).
94
2.3.2. Theverticalresolution
95
The upper 156 mm of the sediment are resolved by 12 layers with increasing thickness (2 - 24
96
mm). Belowthedeepestlayeradimensionlessburiallayerisimplemented.
97
98
BiogeosciencesDiscuss.,https://doi.org/10.5194/bg-2018-58
ManuscriptunderreviewforjournalBiogeosciences
Discussionstarted:5February2018
c Author(s)2018.CCBY4.0License.
(cid:13)
Page 5
2.3.3. NewComponents
99
As the pelagic model delivers sinking particulate material with freely varying stoichiometry, we
100
differentiated benthic POM into the state variables particulate organic carbon(POC), nitrogen
101
102 (PON) and phosphorus (POP). Additionally we added ammonium (NH4) as product of the in-
complete aerobic degradation which can be oxidised by nitrification when oxygen is available
103
(Paulmieretal., 2009). Nitrite is notexplicitly included. The model combines the effectofsul-
104
fatereductionandreoxidationofreducedsulfatecompoundsasnetsulfatereduction(i.e.,sulfate
105
reductionminusreoxidation). Thedifferentreactionequationsincludingthealkalinitygeneration
106
arelistedintheappendix.
107
108
2.3.4. VaryingPorosity
109
Theeffectivityofseveralsedimentreactionsdependsontheporosity,i.e.,theportionofporewater
110
inagivensedimentvolume. Whiletheglobalcoarsegridsedimentmodelwasimplementedwith
111
a horizontally uniform porosity (Heinzeetal., 1999), in the presented shelf application varying
112
porositieswere takeninto account. The mainparts of the North Sea sediments consistofsand,
113
but there are also muddy areas and even rocky areas exist. The different sediment classes are
114
defined by the composition of grains with different diameters. W. Puls kindly delivered us a
115
North Sea wide map of such grain compositions (pers. comm.). As the sediment model uses
116
porosity values (P), the different grainsize distributions have to be mapped to porosityvalues.
117
Weusedthemediangrainsize(D50)tocalculatetheporosity(pers. comm. W.Puls):
118
Psurf =min(1,max(0.3,0.2603·1.20325D50)) (2)
D50=−log2d (3)
weredisthegraindiameterinmm.
119
120 TheresultingporosityvaluesPsurf fallintherange[0.3,1]. Onlyforrockysedimentstheporosity
isdefinedaszero(Fig. 2). AccordingtoHeinzeetal.(1999)porositiesP(z)indeeperlayerswere
121
definedinrelationtothetoplayer:
122
P(z)=Psurf·ek0·z(m) (4)
123 Fork0=2.12andPsurf =0.3thedeepestlayeratzk=12=−0.144mobtainsavalueofP(zk=12)=
0.22.
124
BiogeosciencesDiscuss.,https://doi.org/10.5194/bg-2018-58
ManuscriptunderreviewforjournalBiogeosciences
Discussionstarted:5February2018
c Author(s)2018.CCBY4.0License.
(cid:13)
Page 6
Process TurnoverRates OpenOcean Shelf unit Eqn. No
aerobicdegradation r1 1.160·10−13 2.000·10−10 mmmol3O2·s (7)
denitrification r2 1.157·10−7 1.736·10−3 1s (8)
sulfatereduction r3 1.157·10−9 3.472·10−9 1s (9)
calcitedissolution r4 1.000·10−13 1.000·10−8 mmolmC3O32−·s (10)
opaldissolution r5 1.000·10−12 1.000·10−11 mmomlS3iO2·s (11)
nitrification r6 1.157·10−4 1s (12)
Table1: Comparisonofopenocean(Heinzeetal.,1999)andshelf(thisstudy)turnoverrates.
2.3.5. TurnoverRates
125
The reaction equations and the chosen stoichiometries are described in detail in the Appendix.
126
Theseequationsuseturnoverrateswhichweremodifiedincomparisontotheoriginalopenocean
127
sedimentmodel(seeTable1)
128
2.3.6. TemperatureDependency
129
As the shallow water column in the North Sea exhibits strong seasonal temperature variations
130
(∆T > 15oC) a temperature dependency of both, the turnover rates (see Appendix) and the
131
verticaldiffusionwasimplemented.
132
133 AQ10valueof1.2foraerobicdegradation,denitrification,nitrification,sulfatereductionandthe
dissolutionofcalciteandopalwaschosen(seeAppendix).
134
The verticaldiffusion coefficientforall pore watertracersin the originalopenoceanmodel was
135
136 constant(dv=10−9ms2). Intheshelfmodelthecoefficientsweredefinedastemperature(T)and
porosity(P)dependent(Gypensetal.,2008):
137
dv= (d0+a·T)·P : P<0.4
 (d0+a·T)·P2 : P≥0.4
138 Theparametersd0andaaredefinedinGypensetal.(2008)(theirTable2)fordifferentgroupsof
139 porewatertracers: Thelowestcoefficientisdefinedforphosphate(dvpho(T10)=5.4·P·10−10ms2)
140 amediumcoefficient(dvtra(T10)=1.4·P·10−9ms2)isvalidforthebiogeochemicaltracersDIC,
nitrate,ammonium,alkalinityandsilicate. Thehighestcoefficientwasdefinedforthegasesoxygen
141
142 anddinitrogen(dvnit(T10)=1.6·P·10−9ms2), allatT10=10oCandaporosityP<0.4,which
istypicalforsandyground. Inordertotakeintoaccountadvectiveexchangeofporewaterwith
143
thepelagicsystemthecoefficientsfortheuppermostlayerwereincreasedbyafactorof10. This
144
factorwasdeterminedbyseveralsensitivity runsto balancethe exchangebetweenthe sediment
145
andthepelagic. ThesamefactorisusedbyNeumannetal.(2017)toswitchbetweendiffusiveand
146
BiogeosciencesDiscuss.,https://doi.org/10.5194/bg-2018-58
ManuscriptunderreviewforjournalBiogeosciences
Discussionstarted:5February2018
c Author(s)2018.CCBY4.0License.
(cid:13)
Page 7
advectivenitrateexchangebetweensedimentandpelagicintheGermanBight. Thetemperature
147
ofthesedimentwasdefinedasthetemperatureofthelowestpelagiclayer.
148
The vertical diffusion coefficient for DIC compareswell with the correspondingcoefficient given
149
byBurdigeandKomada(2013)(theirTable3)forT=50CandP=0.36.
150
151
2.4. ExternalData
152
Themeteorologicalforcing(Kalnayetal.,1996)andtheriverloadsofcarbon,alkalinity,nutrients
153
and organic compounds have been implemented according to Lorkowskietal. (2012). To treat
154
thesetracersmorerealisticallyinthisstudyalsodailyfreshwaterdischargeoftheriverswasused
155
156 (Pa¨tschetal., 2016). In this way the input of tracers from the rivers (mmold−1) can be an
effectivesourceorsinkdependingontheconcentrationsofthetracersintheriverwater.
157
The calculated shortwave incoming radiation has been reduced by 10% as it has been shown
158
that the sea surface temperature (SST) would otherwise be overestimated (compare Fig. 3 in
159
Lorkowskietal.(2012)).
160
TheatmosphericnitrogendepositionwasderivedfollowingGroßeetal.(2016),usingannualdata
161
fromtheEMEP(Cooperativeprogramformonitoringandevaluationofthelong-rangetransmis-
162
sions of air pollutants in Europe) model. As our simulation period exceeds the period of data
163
availablefromEMEPalong-termtrendaccordingtoScho¨ppetal.(2003)wasappliedinaddition.
164
Atmosphericdepositionisimplementedasinputsofnitrateandammonium.
165
2.5. TheExperiments
166
For eachexperiment describedbelow the biogeochemicalsimulationin the centraland southern
167
NorthSeaareaspunupover20yearsrepeatingtheyear2000untilallprocesseswereinequlibrium
168
anddidnotchangefromyeartoyear. Afterthisproceduretheyears2000to2014weresimulated
169
consecutively.
170
171
Differentexperimentsorscenarioswereperformed:
172
Thereferencerunwiththenewsedimentmoduleprovidesabasiswithrealisticboundary
173 •
conditionsandhorizontallyvaryingporosities.
174
Inordertoreproduceasituationwithoutanthropogenicinfluence,wereducedtheinorganic
175 •
andorganicriverinputofnitrogenandphosphorusto10%ofthereferencerun. Addition-
176
ally the atmospheric deposition of nitrogen was reduced to 28 %. This run more or less
177
reproducedthe”pristine conditions”Sernaetal.(2010)established.
178
BiogeosciencesDiscuss.,https://doi.org/10.5194/bg-2018-58
ManuscriptunderreviewforjournalBiogeosciences
Discussionstarted:5February2018
c Author(s)2018.CCBY4.0License.
(cid:13)
Page 8
Toanalysethe impactofthe new sedimentmodule onthe pelagic systemwecomparethe
179 •
resultsofthereferencerunwithresultsofthescenario”platerun”. Inthisscenarioasimple
180
sedimentmodulewasused,whichcollects,remineralisesandreleasesthesunkenparticulate
181
organicmaterialonatwodimensionalplate(Pa¨tschandKu¨hn,2008).
182
Inthereferencerunhorizontallyvaryingporositieswereused. Tostudytheinfluenceofthis
183 •
feature we conducted two additional model runs with basin wide uniform porosities: One
184
185 withtheminimumporosityPmin=0.3foundinthemodeldomainofthereferencerunand
186 onewiththemaximumporosityPmax=0.51.
3. Comparison with Observations
187
To get confidence into the adaptedsediment model we comparedsimulatedandobservedfluxes
188
between sediment and pelagic. Additionally, simulated pore water profiles were compared with
189
observedprofiles.
190
3.1. OxygenFluxes
191
Brenneretal. (2016) measured the total oxygen consumption of sediment cores which can be
192
comparedwithsimulatedoxygenfluxesintothesediment. Thecorrespondingavailabledataand
193
theirpositionsareshowninFig. 3(rectangles). Theunderlyingmapofsimulatedoxygenfluxes
194
at the time when observations were taken show reasonable values. Only in the German Bight
195
the model underestimates the measurements. An explanation for this effect is that particulate
196
organics(POM)importedbytheriversareconsideredasslowlysinkingdetritus. Asconsequence
197
the horizontalexportofPOM outofthe GermanBightis overestimatedandthe local flux into
198
thesedimentisunderestimated.
199
3.2. AlkalinityFluxes
200
Fig. 4showsthe comparisonoffieldwide averagedalkalinityeffluxesandthecontributionsfrom
201
aerobicdegradation,denitrification,netsulfatereduction,nitrificationandcalcitedissolutionfrom
202
observationsinSeptember2011(Brenneretal.,2016)andfromourmodelresultsforSeptember
203
2011. Forthe observationaldata only the spatial standarddeviationofalkalinity efflux is given
204
(see grey error bar in Fig. 4a). The temporal standard deviation of the simulated daily values
205
206 within September 2011is forall fluxes verysmall andnotshown(<0.003mmolm−2d−1). The
spatialstandarddeviationofthesimulatedSeptemberfluxesareshownaserrorbarsinFig. 4b.
207
Eventhoughthesimulatedeffluxlieswithinthehighspatialvariabilityoftheobservedalkalinity
208
efflux, the model ratherunderestimatesall contributions. Only the simulatedcontributionfrom
209
aerobic degradation is larger than the corresponding observation. The main deviation can be
210
attributedtothelowsimulatedcalcitedissolutionwithinthesediment.
211
BiogeosciencesDiscuss.,https://doi.org/10.5194/bg-2018-58
ManuscriptunderreviewforjournalBiogeosciences
Discussionstarted:5February2018
c Author(s)2018.CCBY4.0License.
(cid:13)
Page 9
In comparison to other models (Ridgwelletal., 2007; Lorkowskietal., 2012) the ratio of sim-
212
ulated particulate organic carbon to particulate inorganic carbon (POC:PIC) or the simulated
213
214 organiccarbontocalciteproduction(10molPOCd−1 : 1molcalcited−1)isrelativelylowmean-
inghighcalciteproductioninrelationtoorganiccarbonproduction. Nonethelessourmodelstill
215
leadstoanunderestimationofthecalcitedissolutioninthesedimentcomparedtotheanalysisof
216
Brenneretal.(2016). ThemeansimulatedcalcitedepositionintothesedimentinSeptember2011
217
218 was2.9mmolm−2d−1,whichisstillonly54%oftheobservedcalcitedissolutioninthesediment.
219
Fig. 5showsthecorrespondingsimulatedalkalinityfluxestothepelagicsystemfor15September
220
2011. ThealkalinityeffluxisstrongestintheGermanBightnearthemouthofRiverElbe. The
221
fluxdecreaseswithdistancefromthecontinentalcoast. ElevatedvaluescanbeseenofftheDanish
222
coast. Similarfeaturescanbe observedforthe contributorsaerobicdegradation,denitrification,
223
net sulfate reduction, and calcite dissolution. The distribution of the negative fluxes due to
224
nitrificationshowsalsoelevatedvaluesintheGermanBight.
225
Whenignoringthesedimentarycalcitedissolutioninboththesimulationandtheobserveddata,
226
the remaining alkalinity generationcompares better with the observations (Fig. 6). The colors
227
withinthe fourrectanglesidentify the strengthofthe observationaldatawhereasthe horizontal
228
distributionshownbythecoloredmapstandsforsimulations. OnlyintheinnerGermanBightthe
229
simulatedfluxappearsfartoolow. Anexplanationforthiseffectisthesameasforoxygenfluxes:
230
The export of POM out of the German Bight is overestimated and thus local remineralization
231
underestimated.
232
3.3. Profiles
233
DuringthecruiseHe-308inMay2009severalsedimentcoresintheGermanExclusiveEconomic
234
Zone (EEZ) were taken and investigated. The nitrate data are published by Neumannetal.
235
(2017), all data are archived in Pangaea (2017). We compare our results of the reference run
236
with observed data of oxygen, nitrate, phosphate and ammonium (Fig.7). To understand the
237
modelsensitivity,alsothecorrespondingprofilesofthe”pristineconditions”runareshown. The
238
239 positionofthe chosencoreisbetweentheGermancoastandtheislandHelgoland(54o 5′N, 8o
E).Thisareaisstronglyaffectedbyhighnutrientloadsfromthecontinentalriversandhighatmo-
240
sphericnitrogendeposition(Pa¨tschetal.,2010)resultinginsignifcantdifferencesinthesimulated
241
porewater concentrations of the reference run and the scenario ”pristine conditions” (solid and
242
243 dashedblacklines). Thesimulatedoxygenpenetrationdepth(concentration< 10mmolO2m−3)
is about0.5cm whichfits to the observations(Fig.7a). Itis about0.8cm inthe scenario”pris-
244
tine conditions”. In the upper 0.4cm the model underestimates in both scenarios the observed
245
246 oxygenconcentrations. Fig.7b shows the profiles of observedNOx including nitrate and nitrite
BiogeosciencesDiscuss.,https://doi.org/10.5194/bg-2018-58
ManuscriptunderreviewforjournalBiogeosciences
Discussionstarted:5February2018
c Author(s)2018.CCBY4.0License.
(cid:13)
Page 10
andthe profilesofsimulatednitrate. We think thatthis is apropercomparisonasobservedni-
247
248 trite concentrationsare low (< 0.8mmolNm−3, not shown). ObservedNOx concentrationsare
detectableonlyintheupper 2cm. Due touncalibratedmeasurementsdeepervaluesappeardis-
249
criminable from zero concentration, but they should be interpreted as zero concentration(pers.
250
comm. AndreasNeumann). Thesimulatedconcentrations(referencerun)reachverylowvalues
251
alreadyat 1cm depth, the ”pristine conditions” scenarioshows verylow concentrations already
252
at 0.5cm depth. Observedphosphate concentrations in Fig.7c indicate two mixing regimes: In
253
the upper 9cm the sediment core shows concentrations slightly increasing with depth, below a
254
strongergradientcanbeseen. Theupperpartappearswell-mixedwhileinthelowerpartmixing
255
decreases. Thiseffectmightbecausedbybioturbationandbioirrigationintheupper9cm. Asthe
256
latterprocessesarenotincludedinthemodelwegotamorehomogenouspictureofthephosphate
257
profiles. Themodel(referencerun)overestimatestheobservationalvaluesintheupperpartwhile
258
itunderestimatestheminthelowerpart. Asimilarpatterncanbeseenforammonium(Fig.7d),
259
where again the observational concentrations indicate an upper and a lower mixing zone. The
260
simulatedvaluesincreasebetweenthesurfaceandthe 5cmhorizon,belowtheyaremoreorless
261
constant. Thevaluesofthereferencerunaretoohighintheupper13cm. Thesehighsimulated
262
ammonium values might be caused by neglecting the process of anammox in the model. This
263
processtransformsreactivenitrogencompounds(ammoniumandnitrite)intoinertmolecularni-
264
trogen. SimilarhighammoniumconcentrationscanbefoundinLuffandMoll (2004)withintheir
265
Fig. 9.
266
267
4. Results
268
4.1. TemporalVariations
269
Thetemporaldevelopmentofmonthlyalkalinityeffluxes(2000-2014)withoutcalcitedissolution
270
ofanearcoaststation(54o N,8o E)showsanoveralldecreasingtrend(Fig. 8). Tounderstand
271
thisfeaturethesourcesofalkalinitygenerationandtheannualloadsofnitratebytheRiverElbe
272
(RadachandPa¨tsch,2007;Pa¨tschetal.,2016)intheGermanBight(53.9oN,8.9oE)areshown
273
(see Fig. 1a). Calcite dissolution is very variable and exhibits a decrease over the simulation
274
period. Becauseofitshighvariabilitywhichwouldoverwritethe nitrogen-relatedsignalscalcite
275
dissolutionisnotshown.
276
Aerobic degradationwith a distinct annualcycle appearsquite constantoverthe years. Sulfate
277
reductionismoreorlessconstant,whilenitrification(asnegativecontribution)showsapositive
278
trendinoppositetothenegativetrendofdenitrification. Thedarkbluelinerepresentsthenitrate
279
dischargeofthenearbyRiverElbe. Withstrongseasonalpeaksitexhibitsanegativetrendwhich
280
canexplainasimilartrendindenitrification.
281