Biogeosciences, 9, 4087-4097, 2012 www.biogeosciences.net/9/4087/2012/ doi: 10.5194/bg-9-4087-2012 © Author(s) 2012. CC Attribution 3.0 License. Biogeosciences The role of alkalinity generation in controlling the fluxes of CO2 during exposure and inundation on tidal flats P. A. Faber1, A. J. Kessler1, J. K. Bull1,1. D. M cKelvie2 3, F. J. R. M eysman4, and P. L. M. Cook1 1 Water Studies Centre, School of Chemistry, Monash University, Victoria 3800, Australia 2School of Chemistry, The University of Melbourne, Victoria 3010, Australia 3School of Geography, Earth and Environmental Sciences, University of Plymouth, Plymouth PL48AA, UK 4Department of Ecosystem Studies, Royal Netherlands Institute for Sea Research (NIOZ), Korringaweg 7, 4401 NT Yerseke, The Netherlands Correspondence to: P. A. Faber (peter.faber@monash.edu) Received: 2 April 2012 - Published in Biogeosciences Discuss.: 9 May 2012 Revised: 3 August 2012 - Accepted: 30 August 2012 - Published: 23 October 2012 Abstract. Dissolved inorganic carbon (DIC), gaseous CO 2 and alkalinity fluxes from intertidal sediments were investi­ gated during periods of exposure and inundation, using lab­ oratory core incubations, previously published field data and reactive transport model simulations. In the incubations and previous field data, it was found that during periods of alka­ linity production (attributed to the accumulation of reduced sulfur species within the sediment), the flux of DIC out of the sediment was greater during inundation than the gaseous CO 2 flux during exposure by a factor of up to 1.8. This find­ ing was supported by computational simulations which in­ dicated that large amounts of sulfate reduction and reduced sulfur burial (FeS) induce an alkalinity flux from the sed­ iment during high tide conditions. Model simulations also found that the amount of reactive Fe in the sediment was a major driver of net alkalinity production. Our finding that CO 2 fluxes can be significantly lower than total metabolism during exposure has implications for how total metabolism is quantified on tidal flats. 1 Introduction Tidal flats are highly dynamic and biogeochemically ac­ tive environments, that are characterised by the deposition of allochthonous organic matter as well as high in situ rates of primary production by microphytobenthos (Joye et al., 2009). The quantification of organic carbon produc­ tion and mineralisation in these environments is compli­ cated by the fact that tidal flats are periodically inundated. This means that fluxes of inorganic carbon should ideally be measured both during inundation (as dissolved inorganic carbon DIC = [CO 2 ] + [HCO(("] + [CO 3 - ]) and exposure (as gaseous CO 2 ). In practice, the quantification of inorganic carbon fluxes during the inundation of tidal flats is logistically complex because they are covered by a shallow layer of turbid water and often experience high currents. Moreover, the determination of DIC in aqueous samples is also more time consuming and technically challenging than the analy­ sis of CO 2 in gas samples. Typically, most studies of benthic metabolism on tidal flats only measure gaseous fluxes of CO 2 during exposure (Migné et al., 2005: Middelburg et al., 1996) or DIC flux during inundation (Cai et al., 1999). Relatively few studies have included direct measurement of inorganic carbon fluxes during both exposure and inunda­ tion of tidal flats and these studies show different outcomes. Alongi et al. (1999) found very similar exchange rates dur­ ing both inundation and exposure. By contrast, Gribsholt and Kristensen (2003) and Cook et al. (2004) found consistently higher rates of inorganic carbon exchange during inundation on un-vegetated tidal flat sediments, where the flux during in­ undation increased by a factor of ~ 2. A number of possible explanations exist for these differences. Firstly, solute trans­ port processes are very different between inundation and ex­ posure periods. During inundation, all three forms of the car­ bonate system (CO 2 , HCO^ and CO 3 - ) can be transported across the sediment interface, whereas during exposure, only gaseous CO 2 can diffuse into the atmosphere. Furthermore, Published by Copernicus Publications on behalf o f the European Geosciences Union. 4088 P. A. Faber et al.: The role of alkalinity generation bio-irrigation is w ell know n to enhance solute exchange in sedim ents (Kristensen, 1988). T his process w ill cease upon exposure, and w ould result in a reduction o f the C O 2 flux during exposure as com pared to the inundated DIC flux. S ec­ ondly, alkalinity generation, caused by the burial o f reduced m etal sulfides and/or calcium carbonate dissolution, is of­ ten observed at high rates in shallow w aters and intertidal sedim ents (Thom as et al., 2009; C ook et al., 2004; F ergu­ son et al., 2003). A lkalinity cannot escape the sedim ent dur­ ing low tide and, therefore, w ill tend to accum ulate near the sedim ent-w ater interface. T his w ill lead to a shift tow ards in2_ creased H C O 3 and CO 3 and decreased C O 2 and, hence, a reduced gaseous C O 2 flux during exposure. H owever, there are also reasons w hy the C O 2 flux m ight be enhanced during exposure. A change in C O 2 transfer during exposure to the atm osphere m ay occur due to the higher diffusion o f C O 2 in the atm osphere (IO - 5 com pared to 1 0 _ 9 m 2 s _1) and the decreased thickness o f the diffusive boundary layer during exposure (Brotas et al., 1990). To date, none o f these factors have been explored, exper­ im entally or theoretically. A basic understanding o f the ex­ change dynam ics o f inorganic carbon is required for the d e­ sign o f studies aim ing to quantify the fluxes o f inorganic car­ bon betw een tidal flats and the atm osphere and coastal w a­ ters. H ere, w e used data from controlled laboratory experi­ m ents and a num eric diagenetic m odel to investigate the dy­ nam ics o f inorganic carbon exchange in the intertidal zone. In particular, w e focus on the role o f alkalinity generation by anaerobic respiration in the retardation o f the C O 2 efflux during exposure relative to D IC export to the w ater colum n. 2 2.1 M aterials and m ethods E xperim ental M uddy sedim ent (Grain size: 48 % 300 p m -1 mm; 8 % 2 0 0 300 pm ; 3 9 % 100-200 pm; 5 % 6 2 -1 0 0 pm and p oros­ ity = 0.81 m l w ater cm - 3 ) w as collected from an intertidal flat located in the Yarra Estuary, A ustralia (37.833° S, 145.0229° E). The sedim ent w as sieved (1 m m mesh) to re­ m ove m acrofauna and hom ogenised. T he total inorganic car­ bon content o f the sedim ent w as < 0.1 % w / w . T he sedim ent w as then placed in a shallow tray and gently stirred w hilst b e­ ing aerated for 1 day to oxidise reduced solutes before being packed into 8 core liners (7 cm diam eter) to a depth o f 5 cm. This depth layer w as chosen to represent the diagenetically active depth. Figure 1 represents the experim ental setup that w as used. The cores had stirrer bars inserted to gently stir the w ater colum n. A peristaltic pum p system transferred am ­ bient sea w ater collected from P ort Phillip Bay in and out o f the cores (Salinity = 35, Tem perature = 20 °C ). The pum p­ ing regim e follow ed a regular six hourly cycle o f inundation and exposure that m im icked a sem i-diurnal tide. T he inlet o f the tubes in the core w as recessed slightly into the sedim ent Biogeosciences, 9, 4087-4097, 2012 6 hours High tid e condition Low tid e condition r \ I S to p p er Pum pand w a te rre se rv o ir Pum pand J . w a te rre selervoir rv o ir J rH ead sp a ce C J S ed im en t S edim ent core Fig. 1. Diagram of the experimental setup, showing the conditions of simulated tides. TA and DIC are sampled from the water col­ umn during inundation (high tide) and CO2 is sampled from the headspace during exposure (low tide). MS refers to a magnetic stir­ rer bar. surface, so that all the overlying w ater could be rem oved, leaving no ponding o f w ater on the surface during exposure. C arbon m ineralisation processes in natural system s can be highly dynam ic reflecting pulsed inputs o f organic m atter (OM ), for exam ple, deposition o f phytodetritus. This pulsed occurrence o f carbon m ineralisation is highly relevant, driv­ ing cycles o f solute reduction and oxidation, w hich in turn drives alkalinity production and consum ption. To sim ulate a pulse o f organic m atter, h a lf o f the cores had 0.38 g b aker’s yeast added to them after 71 days (+O M treatm ent), w hile the rem aining cores did not receive this organic m atter input (—OM ). D uring inundation, the flux o f D IC w as determ ined by covering the cores w ith sealed lids and taking 4 -5 w ater sam ples during sedim ent inundation. T he DIC flux w as d e­ term ined as the slope o f D IC concentration versus tim e m u l­ tiplied by the w ater height. T he pH w as sim ultaneously m ea­ sured as the w ater sam ple fo r alkalinity w as taken. D uring exposure, the flux o f gaseous C O 2 w as m easured by sam ­ pling the gas headspace o f the sealed core. A gain, the C O 2 flux w as determ ined as the slope o f C O 2 concentration v er­ sus tim e m ultiplied by the headspace height. C O 2 w as sam ­ pled using evacuated 3 ml draw blood collection vials (Vacutainers, B ecton D ickinson) and analysed using Flow Injec­ tion A nalysis (Satienperakul et al., 2004). A lkalinity sam ­ ples w ere filtered (Bonnet, 0.45 pm polyethersulfone) and preserved w ith 2 0 p lH g C l 2 (6 % w / v ) . T hey w ere stored in glass vials and refrigerated at 4 ° C for less than one w eek before analysis using a m odified G ran titration (A lm gren et al., 1983). pH w as determ ined at 20 °C using a pH electrode (Hach PHC301 connected to an H Q 40d m eter), calibrated w ith N BS buffers. D IC concentrations in the w ater colum n w ere calculated using alkalinity and pH, w ith the constants found in R oy et al. (1993). T he m ethod for total Fe analysis o f the dried sedim ent w as m odified from Lord (1982), u s­ ing a 24 h citrate/bi carbonate/dithi oni te extraction for easily extractable Fe, follow ed by a 24 h concentrated nitric acid www.biogeosciences.net/9/4087/2012/ P. A. Faber et al.: The role of alkalinity generation digestion. The extracts w ere then analysed using atom ic ab­ sorption spectroscopy (AAS). 2.2 M odel formulation The reactive transport m odel follow s the standard form u­ lation for early diagenetic m odels of m arine sedim ents (Boudreau, 1997; M eysm an et al., 2003). The m odel sim ­ ulates the depth profiles of solutes and solids, as w ell as the fluxes across the sedim ent-w ater interface, based on the mass balance equations Solutes: cp~dCr sr dx (pDi 9C . J2vi,kRk dx k dCf Solids: (1 - < p ) - ÿ j - = J 2 vi,kRk w here C pw and C f are the concentrations o f a solute and solid com pound, respectively. No advective processes are in­ cluded in the model, as negligible sedim entation occurred during the incubations. The porosity <p is assum ed constant w ith depth. For solutes, the only transport process is m olec­ ular diffusion. The diffusion coefficient Z)¡ is calculated as a function o f tem perature and salinity using the R package m arelac (Soetaert et al., 2010) and subsequently corrected for tortuosity according to the m odified W iessberg relation of Boudreau (1996). The quantities R t represent the reac­ tion rates, w here ry^ is the stoichiom eric coefficient o f the r-th species in the k-th reaction. The reaction set includes three m ineralisation pathw ays (aerobic respiration, dissim ilatory iron reduction and sulfate reduction) for two fractions of organic m atter (fast and slow decaying), the form ation of iron sulfide and pyrite, and the reoxidation o f reduced com ­ pounds in pore w ater (ferrous iron, free sulfide) and solid phase (iron sulfides and pyrite). The full set o f 13 reactions is given in Table 1. T he rate expressions for these reactions follow the standard kinetic rate law s and are given in Table 2. For sim plicity, w e have assum ed that processes such as nitri­ fication and denitrification have a negligible effect on alka­ linity fluxes in this system. 2.3 pH model The set o f acid-base reactions governing the pH dynam ics in the pore w ater includes the carbonate, borate, sulfide and w a­ ter equilibria. The im pact of dissociation reactions involving phosphate, am m onium , silicate and dissolved organic com ­ pounds w as assum ed negligible. The associated total alkalin­ ity and total species are hence defined as: A t = [H C O 3-] + 2 [C O 2“ ] + [B (OH)¡"] + [H S“ ] + [O H “ ] - [H+] S u m C 0 2 = [ C 0 2] + [H C O J] CO 2 - Sum BO H 3 = [B (O H )3] + [B (O H )“ ] S u m H 2 S = [H S “ ] + [H 2 S] www.biogeosciences.net/9/4087/2012/ 4089 Table 1. K inetic reactions included in the reaction set. R1 R2 {CFHOfi + O2 —> H2O {CH2 0 } f + 4 FeO O H + 8 H + ^ R3 R4 {CH20 } f + \ SO)]“ + \ H + - * CO 2 + \ H “ + H 2 0 {CH20 } s + 0 2 -* CO 2 + H 20 4 Fe2+ + CO2 + 7 H 2 O R5 {CH20 } s + 4 FeO O H + 8 H + -* 4 Fe2+ + C 0 2 + 7 H 20 R6 {CH20 } s + 1 SO 2“ + | H + R7 H S“ + 2 O2 - * SO)]“ + H+ R8 Fe2+ + \ 0 2 + I H 20 - * FeO O H + 2 H + CO2 + + H 20 R9 8 FeO O h = H S - + 15 H + -* SO^4_ + 12 H 20 RIO Fe2+ + H S“ - * FeS + H + R ll ¿W-* r FeO 1 CW T ow4 FeS + I 0 2 + 2 I 1 H120 OW Hll + SO)]“ + 2 H + R12 FeS + I SO)]“ R13 FeS2 + O2 | I H S - + I H + -* FeS2 + H 20 F e O O H + 2 SO',2 4H + h 2o 4 The associated equilibrium constants w ere calculated as a function o f tem perature and salinity using AquaEnv, a dedi­ cated R -package for acid-base and C O 2 system calculations (Hofm ann et al., 2010). 2.4 Boundary conditions and initial state The cyclic process o f inundation and exposure w as sim u­ lated by regular sw itching betw een two sets o f boundary conditions. U nder inundation, the concentration of solutes at the sedim ent-w ater interface w as fixed to that o f the overlying w ater (which w as set at constant pH and solute con­ centrations) , w hile a fixed flux w as im posed for the solids. T hese fluxes w ere set to zero for all solid com ponents dur­ ing the w hole sim ulation period (zero deposition assum ed - see above), apart from organic matter, w hich received a pulse input at 71 days. To m odel this pulse input, the organic m atter w as added uniform ly to the top 2 mm o f the sedim ent at a total concentration o f ~ 2500 pinoi C ^ O g “ 1 dry sed­ im ent and fractionated betw een the fast-decaying fraction (75% ) and the slow -decaying fraction (25% ), a ratio used by W estrich and B erner (1984) (who also included a third, non-reactive fraction in their model.) U nder exposed conditions, only the volatile solutes O 2 and C O 2 have an exchange w ith the overlying atm osphere, w hich is m odelled using the convective boundary flux J = ká ( C eq — C) w ith ká the piston velocity and C eq the gas con­ centration in equilibrium w ith the atm osphere. A t the lower boundary o f the sedim ent dom ain a no-flux condition w as im posed (fluxes across the bottom boundary are zero) for all com pounds over the w hole sim ulation period. The reduced Fe (FeS and FeS 2 ) concentration w as ini­ tially set to 0 and the O 2 concentration in the pore w a­ ter w as set to equilibrium w ith the atm ospheric concen­ tration. A uniform profile o f slow -decaying organic m at­ ter (~ 2500 pm ol C H 2 O g “ 1 dry sedim ent), and either 19 or 95 pinoi F e g “ 1 dry sedim ent w as imposed. These latter val­ ues w ere determ ined to be the extrem e low and high Fe Biogeosciences, 9, 4087-4097, 2012 4090 P. A. Faber et al.: The role of alkalinity generation Table 2. K inetic rate expressions for the reactions included in the reaction set. N ii in — (1 V ) f in si [ C H 2 O f ] , = ( 1 - ¥ > ) * s 1o w [ C H 2 O s Ri = [0 2] [0 2] + Ko, m [FeOOH] Rz = Rs = [S 0 4 ~ ] [S° r ] Ri = R o2 [FeOOH] + KpeOOH [ 0 2] + K q 2 KpeOOH R o2 2- [FeOOH] + ^FeOOH [ 0 2] + K q 2 ' + OL)A 2.6 [ ° 2] „s [0 2] + Kq m [FeOOH] r5 = r6 = Rl = # H 2S - O x [ 0 2 ] K o2 [FeOOH] + KpeOOH [ 0 2] + K q 2 ' K o2 KpeOOH [FeOOH] + KpeOOH [ 0 2] + K q 2 ' [S ° 4 ~ ] [S°r]- ■K S O 2 - HS- R& = Ç*Fe-Ox [pe2+ [0 2]_ Tfg = (1 - <P) ^FeOOH-FFS [FeOOH] [H SR w = U 'irS . lomi |^Fe j [HS ] ^11 = (1 - ¥>) ^FeS-Ox [FeS] [ 0 2] ^12 — (1 ~ (P) ^FeS2.form [FeS] H S~ ^13 = (1 - ¥ ,)*FeS2-Ox[FeS2] 0 2] concentrations, as m easured by a total Fe extraction at the conclusion o f the experim ent. 2.5 solved for the unknow n proton concentration using the it­ erative m ethod o f Follow s et al. (2006). U sing this proton concentration (or equally pH v alu e), the full spéciation could be calculated o f all chem ical species involved in the acidbase dissociation reactions. The resulting concentrations of the equilibrium species could then be em ployed in kinetic rate expressions (specifically, the pore w ater C 0 2 concen­ tration w as needed in the rate expression for the air-sea gas transfer o f C 0 2). Further details on this pH m odel procedure are given by Flofm ann et al. (2008). Description o f model runs The m odel w as set up to sim ulate the tidal cycles im posed in the laboratory core incubation experim ents, w hich w ere exposed and inundated alternately every 6 h. The model sim ulations provide flux estim ates at each point in tim e. The fluxes o f TA and DIC during inundation and C 0 2 fluxes during exposure are reported as the m ean over the 6 -h exposure/inundation period. T hese reported m ean values include transient (~ 30 min) spikes in DIC and TA im m ediately after inundation. The sim ulation extended over 120 days (i.e., 240 tidal cycles), w ith a pulse o f organic m at­ ter added on day 70, sim ulating the addition o f the organic m atter source in the laboratory experim ents. Table 3 lists an overview o f param eters values used in m odel sim ulations. T hese w ere determ ined from m easured properties, literature values or calibrated to the available dataset. All param eter values w ere kept constant during the w hole sim ulation pe­ riod (apart from the organic m atter input). Numerical model solution 2.7 A num erical solution procedure w as im plem ented in the open-source program m ing language R as fully detailed in Soetaert and M eysm an (2012). A reactive transport model essentially consists o f one partial differential equation (PDE) for each com pound. Together w ith the m ass action laws o f the acid-base equilibria, these PD Es form a differentialalgebraic system , w hich w as solved using an operator sp lit­ ting approach (Solution m ethod, 3b, as explained in Hofm ann et al., 2008). Follow ing the m ethod-of-lines, the R package R eacTran uses a finite difference schem e to expand the spatial derivatives of the PD Es over the sedim ent grid. This grid w as obtained by dividing the sedim ent dom ain (thickness L = 5 cm, the approxim ate depth o f the sedim ent cores in the incubation experim ent) into a uniform grid of 100 sedim ent layers. A fter finite differencing, the resulting set of O D Es w as integrated using the stiff equation solver vode from the R -package deSolve. A t each tim e step, the operator splitting approach o f the pH model required an additional pH equilibration step. To this end, the num erical integration o f the O D E system pro­ vided the values o f the reaction invariants (TA, S u m C 0 2 , SumF[2S, SumBOF[3) at each future tim e step. Subsequently, the nonlinear system o f algebraic acid-base expressions w as Biogeosciences, 9, 4087-4097, 2012 Sensitivity model runs The sensitivity o f C 0 2 flux to the piston velocity w as in­ vestigated using sim ulations, w here the piston velocity w as varied from 0.05 to 5 c m h _1. In each sim ulation, the initial FeO O H concentration w as 19 pm ol F e g - 1 dry sedim ent. For each value o f the piston velocity, the average flux o f C 0 2 during exposure w as calculated over the com plete sim ula­ tion period (30 tidal cycles). T his procedure w as repeated for three different initial values o f organic m atter concentration (2500, 1000 and 6 0 0 p m o lC g _1 dry sedim ent for the slowdecaying fraction, and 5, 2.5 and 1 .3 p m o lC g _ 1 dry sedi­ m ent for the fast-decaying fraction, respectively), to investi­ gate the sensitivity o f C 0 2 flux to the piston velocity under different am ounts of organic m atter loading. The availability o f reactive oxidised iron (FeOOH) con­ trols alkalinity production in the sedim ent and, hence, w e perform ed a second series of sim ulations to investigate the sensitivity of the m odel tow ards the initial sedim entary FeO O H concentration. To determ ine the effect of FeOOH availability on TA production, sim ulations w ere run w ith a fixed initial concentration o f C H 20 (~ 2500 pm ol C H 20 g - 1 dry sedim ent), but w ith various FeO O H concentrations (7.6, 19 and 95 pm ol F e g - 1 dry sedim ent). T he values of 19 and www.biogeosciences.net/9/4087/2012/ P. A. Faber et al.: The role o f alkalinity generation 4091 Table 3. O verview o f m odel param eter values. 7 6 Constant Value Units Porosity rho.sed (density of solid sediment) Piston velocity K.fast (highly labile organic matter) K.slow (less labile organic matter) K _02 (Monod constant for 0 2 consumption) K_FeOOH (Monod constant for FeOOH reduction) 0.8 2.6 - K _S02~ (Monod constant for 1.6 g cm ~ 2 1 25 cm h ~ 4 yr- 1 3.1 dry sedim ent Alkalinity 4 E E 33 Ï 2 LL 1 7 yr- 1 0.6 0.005 6 I o 200*rho.sed pmol cm~ 2 pmol cm~ 2 19 pmol Fe g DIC C 02 pmol cm~ 2 1X 3 4 .1 dry sedim ent Alkalinity 3 2 0 1.6 X IO7 p m o P 4 cm 2 yr~ 4 2.0 X IO4 p m o P 4 cm 2 yr~ 4 7 +OM co2 160 p m o U 4 cm 2 yr~ 4 0 p m o U 4 cm 2 yr~ 4 2.57 p m o P 4 cm 2 yr~ 4 1 X 10s p m o P 4 cm 2 yr~ 4 1 p m o P 4 cm 2 yr~ 4 95 pm ol F e g - 1 dry sedim ent are the low and high estim ates of sedim ent FeO O H concentrations in the core experim ent sedim ent, respectively. The value o f 7.6 pm ol Fe g - 1 dry sed ­ im ent w as chosen as a low value for a com parison w ith our estimates. 3 Í o -1 0 S 0 2~ reduction) k_H2S.Ox (Kinetic constant for H 2S oxidation) k_Fe.Ox (Kinetic constant for Fe oxidation) k_FeS.Ox (Kinetic constant for FeS oxidation) k_FeS2.Ox (Kinetic constant for FeS2 oxidation) k_Fe.H2S (Kinetic constant for F eO O H /H 2S redox reaction) k_FeS.form (Kinetic constant for FeS formation) k_FeS2.form (Kinetic constant for FeS2 formation) ■*= 5 9 5 pmol F e g DIC co 2 Results Experimental There w as little variation in the DIC fluxes over the entire course o f the experim ent in the —OM treatm ent w ith a m ean value o f 1.3 m m ol m - 2 h - 1 over the 120 days (Fig. 2). In the +OM treatm ent, DIC fluxes w ere the sam e as the —OM treat­ m ent w ithin experim ental uncertainty in the first 70 days of the experim ent before the organic m atter w as added (Fig. 2). A fter the organic m atter addition, DIC fluxes substantially increased to a m axim um o f ~ 5.0 m m ol m - 2 h- 1 , and then decreased again to ~ 2 . 0 m m ol m - 2 h - 1 over a tim e period of 40 days after organic m atter addition. A lkalinity fluxes w ere initially high in both —OM and +OM treatm ents, steadily decreasing from > 1.4 m m ol m - 2 h - 1 at the start of the ex­ perim ent to attain a negligible alkalinity flux after 60 days. The m ost likely explanation for this observation is that there www.biogeosciences.net/9/4087/2012/ DIC Alkalinity Alkalinity D ays Fig. 2. S im ulated m ean fluxes o f C 0 2 during exposure (C O g), dissolved inorganic carbon (DIC) and alkalinity during inundation in the com puter sim ulation w ith 95 and 19 pm ol F e g - 1 sedim ent ( d/ w) and laboratory core incubation fluxes w ith (+OM) and w ith ­ out (—OM) added organic m atter. E rror bars are standard deviations across 10 replicates. w as build-up of iron sulfides, w hose rate o f form ation slowed over tim e. We hypothesise that initially, a large pool of Fe (III) w as form ed during sedim ent pre-treatm ent, w hich w as subsequently reduced to m etal sulfides. D issim ilatoiy iron reduction produces alkalinity (TA : DIC = 8 , see R9 in Table 2), but as the pool o f Fe (III) is depleted, alkalinity pro­ duction w ill decrease over tim e. A fter the pulsed addition of organic matter, alkalinity fluxes increased rapidly to a m ax­ im um o f 3.0 m m ol m - 2 h - 1 before decreasing again rapidly. The C O 2 fluxes during exposure w ere initially very low in both treatm ents (~ O .lO m m o lm - 2 h- 1 ). In the —OM treat­ ment, the C O 2 flux increased after 60 days, coinciding w ith Biogeosciences, 9, 4087-4097, 2012 4092 P. A. Faber et al.: The role of alkalinity generation the decrease and cessation of alkalinity fluxes (fluxes not significantly different from zero, shown by the error bars). In the +OM treatment, there was an increase in CO 2 fluxes after the addition of organic matter, reaching a maximum of ~ 2.5 mmol m - 2 h- 1 . The CO 2 fluxes decreased again to ~ 0.60 mmol m - 2 h - 1 at the conclusion of the experiment. 3.2 M odel simulations To match with the laboratory experiments, the data from the model were truncated before 20 days (Fig. 2). From day 20 to 70, the simulation yielded inundated DIC fluxes between ~ 1.2 and ~ 1.4 mmol m - 2 h- 1 . Immediately after the OM addition at day 70, the DIC flux increased to a maximum of ~ 5.7 m m olm - 2 h- 1 . The TA flux decreased from ~ 0.48 to ~ 0.40 mmol m - 2 h - 1 before the OM addition, then in­ creased to a maximum of ~ 0.74 mmol m - 2 h - 1 after the ad­ dition. The CO 2 flux during exposure was between ~ 0.66 and ~ 1.0m m olm - 2 h - 1 before the OM addition, and rose to a maximum of ~ 3.6 m m olm - 2 h - 1 after the addition of the OM, then decreasing again to ~ 1.2 mmol m - 2 h - 1 by the end of the simulation. The peak flux during the simulation is shifted forward in time relative to the core incubations indi­ cating a delayed response of organic matter mineralisation under experimental conditions. This is expected, as the bac­ terial community first needs to grow to consume the bioavailable organic matter. In the computer simulation, no lag phase occurs, as the model formulation does not include popula­ tion growth and, thus, assumes an instantaneous response of microbial metabolism. In the model simulations, DIC fluxes were consistently higher during inundation than CO 2 fluxes during the cor­ responding exposure periods (Fig. 3). The discrepancy be­ tween DIC and gaseous CO 2 fluxes appears to increase with increasing respiration, shown by the divergence of the sim­ ulation data from the 1 : 1 line as DIC flux increases. This pattern was very similar to that previously observed for in­ tact sediments collected from tidal flats (Cook et al., 2004) (Fig. 3). The fluxes of gaseous CO 2 during exposure and the fluxes of DIC and TA during exposure were highly sensi­ tive to initial Fe concentrations in the sediment. This can be evaluated by calculating the mean fluxes over the pe­ riod before the organic matter pulse (Fig. 4). The high Fe concentration resulted in a lower CO 2 flux during ex­ posure (~ 0.65 vs. ~ 0.35 mmol m - 2 h - 1 for the 19 and 95 pmol F e g - 1 dry sediment). There was also a large dis­ crepancy between alkalinity fluxes, where the high Fe con­ centration resulted in a higher TA flux (0.80 mmol m - 2 h - 1 versus 1.4 m m olm - 2 h - 1 for the 19 and 9 5 p m o lF e g - 1 dry sediment simulations). The difference between DIC fluxes were smaller, with a flux of ~ 1 . 8 mmol m - 2 h - 1 for the simulation with 95 pmol F e g - 1 dry sediment and ~ 1. 6 mmol m - 2 h - 1 for the simulation with 19 pmol Fe g - 1 dry sediment. Biogeosciences, 9, 4087-4097, 2012 I n c u b a te d C o re s C o o k e t al 2 0 0 4 1:1 line C 0 2 vs DIC (simulation) 0 1 2 3 4 5 E x p o s e d C 0 3 flux (m m ol m"2 h"1) Fig. 3. F luxes o f dissolved inorganic carbon (DIC) versus exposed C O 2 fluxes for tw o experim ental datasets and a m odel sim ulation. Incubated core experim ents represent both the + and —O M trea t­ m ents. 3.3 Simulation profiles The profile of FeOOH in the top 0.5 cm of sediment changed dramatically upon the addition of the organic matter pulse to the top 0.2 cm (Fig. 5). The profile taken two tidal cycles be­ fore the pulse shows the FeOOH reaching ~ 15 pmol F e g - 1 dry sediment in the surface layer and decreasing with in­ creasing depth. One tidal cycle after the addition of organic matter, FeOOH has begun to oxidise, with the concentration in the top 0.2 cm falling to a maximum of 2 pmol F e g - 1 dry sediment in the surface layer. Below 0.2 cm, the FeOOH pro­ file is similar between the two times, as the organic matter concentration has not changed in this domain. 4 4.1 Discussion Alkalinity producing reactions The production of alkalinity can be attributed to several key processes: ammonia release, net denitrification, net sulfate reduction and to a lesser extent, dissolved organic carbon release (Hammond et al., 1999). Berner et al. (1970) pro­ posed the idea of sulfate reduction as an alkalinity producing process. It was noted that sulfate reduction coupled with the precipitation of the stable iron sulfides FeS and FeS 2 pro­ duces one mole of alkalinity per mole of carbon mineralised (Berner et al., 1970) (Eq. 1). 6CH 20 + 3 S 0 2 - + 2FeOOH FeS + FeS2 + 6HCO:r + 4H 20 (1) In this representation of sulfate reduction, the supply of Fe (III) to the sediment enables the burial of reduced sul­ fur compounds as iron sulfides (FeS) or pyrite (FeS2 ). This burial of reduced sulfur prevents sulfide oxidation and, www.biogeosciences.net/9/4087/2012/ P. A. Faber et al.: The role of alkalinity generation 4093 2.0 0.0 Ö 1.6 0.1 0.2 — F e OOH b e fo re OM addition ■■ O ■ F eO O H a fte r OM addition 0.3 0.6 0.4 0.7 0.6 0.5 0 2 4 6 8 10 12 14 16 ftm ol F e g"' d ry s e d im e n t 0.4 Fig. 5. A sim ulation show ing the FeO O H concentration before and 0.3 after an organic m atter pulse w ith initial FeO O H concentration of 1 9 p m o lg - 1 sedim ent (d/ w). Note: T he step around 0 .2 c m is due to the pulse o f organic m atter added to only the top 0.2 cm. 0.2 1.6 7 .6 pm ol F e g d ry se d im e n t -1 E 19 pm ol F e g “5 9 5 (omoi F e g d ry se d im e n t -1 d ry se d im e n t E E X D < I- 03 D C T3 C 0.2 20 10 0 30 40 50 Tim e a fte r F e an d o rg an ic c a rb o n addition (d ay s) Fig. 4. Fluxes o f dissolved inorganic carbon (DIC), exposed C O 2 and alkalinity fluxes for a sim ulation o f a 50 day period w ith an im ­ posed profile o f 7.6, 19 and 95 p m o lg - 1 sedim ent (d / w ) FeO O H . hence, it prevents the alkalinity generated by sulfate reduc­ tion being consum ed again (Eq. 2). H S- + 2 O 2 -> SO 4 - + H + (2) U pon exposure to oxygen, iron sulfides can also oxidise, leading to alkalinity consum ption (and, hence, no net alka­ linity generation by sulfate reduction) (Eq. 3). FeS + 2 .2 5 O 2 + 2 .5 H 2 0 F e( 0 H ) 3 + S 0 f2 + 2H + (3) However, the rate o f this reaction is typically low er than that o f free sulfide oxidation (see Table 3). M ore importantly, one requires a suitable transport process, that either m oves the re­ duced sulfur com pounds upw ards to the oxic zone (like biotu rbation), or transfers oxygen dow n into anoxic layers (like bio-irrigation). In the experim ental setup here, such transport processes are not present. H u and Cai (2011) review ed the factors controlling alka­ linity generation in ocean m argins in sedim ent and concluded www.biogeosciences.net/9/4087/2012/ that the burial o f pyrite and the denitrification o f N O 3 d e­ rived from the continents w ere the only net sources o f al­ kalinity in ocean m argin sedim ents. A s the cycling o f su l­ fur is closely linked w ith that o f iron, w e included both su l­ fur and iron geochem istry in the diagenetic m odel. In our experim ents, denitrification w as unlikely to be a significant source o f alkalinity generation because o f the lack o f NO3 in the overlying w ater o f our incubations. C oupled nitrifica­ tion denitrification w ill not results in significant net alkalin­ ity generation (Hu and Cai, 2011). Consequently, w e did not m odel the effect o f denitrification on alkalinity in this m odel. D issolved organic m atter and HPOjj- have been show n to contribute to alkalinity fluxes (Cai et al., 2010), but reactions involving sulfur (and it interactions w ith iron) are the m ajor processes controlling the alkalinity balance w ithin the sed ­ im ent and, hence, only these processes are considered. We are confident that the m odel used here represents the critical anaerobic alkalinity-generating processes that are im portant in our experim ental set up. 4.2 Model sensitivity towards piston velocity A t piston velocities > l c m h - 1 the CO 2 efflux reached a stable value and this asym ptotic CO 2 efflux increased p ro ­ portionally w ith initial organic m atter content (or total m in ­ eralisation) (Fig. 6 ). D ecreasing the piston velocity below 0 .5 c m h _ 1 reduced CO 2 fluxes during sedim ent exposure. T his can be attributed to the slow er efflux o f CO 2 for a given C O 2 concentration gradient (transport control), and a slow er flux o f O 2 into the sedim ent resulting in less aerobic respi­ ration. Indeed, w e found the rate o f aerobic respiration to decrease w ith decreasing piston velocity (data not shown). Initially w e thought that the alkalinity flux w ould rem ain Biogeosciences, 9, 4087-4097, 2012 4094 P. A. Faber et al.: The role of alkalinity generation E o Low m ineralisation rate M edium m ineralisation rate High m ineralisation rate E E X 0.6 =5 CM O o ■o o in o Q_ 0.4 0.2 X LII 0.0 0 1 2 3 4 5 6 P isto n V elocity (cm h'1) Fig. 6. E xposed flux o f C O 2 w ith v arying piston velocity and respi­ ration rate. Low, m edium and high m ineralisation rates refer to peak D IC productions o f approx. 600, 1300 and 2 6 0 0 p m o lm _2 h - 1 , re­ spectively. constant upon changing piston velocity. It w as noted, how ­ ever, that there w as an increase in alkalinity production by 7 % as the piston velocity w as decreased from 5 to 0.05 (data not shown) indicating that anaerobic respiration accounted for a greater proportion of the total respiration. H igher respi­ ration rates w ere m ore sensitive to decreasing piston veloci­ ties. We concluded that w hen the piston velocity is greater than l c m h - 1 , exchange w ill be lim ited by production of C O 2 (reaction control), rather than transport across the airw ater interface (Fig. 6 ). For the rem ainder of the sim ulations, the piston velocity w as fixed at 1 cm h _ 1 , the low est o f values found by various researchers across the w ater-air interface in estuarine and riverine system s, as sum m arised by Raym ond et al. (2 0 0 0 ). 4.3 Agreement o f simulations with experimental data There w as a general agreem ent betw een experim ental flux data and the m odel sim ulations. Both m odel and experim en­ tal data show ed that DIC fluxes from the sedim ent during tidal inundation are higher than C O 2 fluxes during exposure. This difference w as m ost pronounced at the start o f the exper­ im ent and after the addition of organic matter. A t the start of the experim ents there w as a rapid alkalinity generation w hich can be explained by the form ation of FeS as the Fe (III) ini­ tially present w ithin the surface sedim ent w as reduced. A fter the pulse of organic matter, the oxygen penetration in the sed­ im ent is reduced and a higher rate o f anaerobic carbon m in­ eralisation takes place. This leads then to increased rates of reduced sulfur production, FeS burial and net alkalinity pro­ duction. The reduction of Fe in the surface o f the sedim ent leads to a transient alkalinity flux and associated increase in the discrepancy betw een gaseous C O 2 flux and DIC flux (Fig. 5). This observation is seen in both the m odel and the Biogeosciences, 9, 4087-4097, 2012 laboratory incubations, although there is a slight offset due to the fact that there is a delay in the m axim um respiration rate in the laboratory incubations com pared to the model. In our sim ulations, reduced sulfur production continues after FeO O H is exhausted, but it is reoxidised as free su l­ fide diffuses upw ards into the oxygenated surface layers of sedim ent. N onetheless, a relatively sm all alkalinity flux still leaves the sedim ent, as there is som e build up of H 2 S in the sedim ent pore w aters (The diffusion o f sulfide into the oxic layer, and its subsequent oxidation is slow er than its produc­ tion w ithin the deeper layers o f the sedim ent). The dynam ics of this sim ulation/experim ent have strong environm ental relevance being representative o f a resuspen­ sion event or the deposition o f fresh sedim ent from the catch­ m ent and the deposition o f organic matter. This highlights the im portance o f environm ental dynam ics in controlling the relative loss o f C O 2 and D IC from intertidal sedim ent. In­ deed, in coastal sedim ents, periodically high fluxes o f alka­ linity are observed follow ing periods o f high organic m atter input (Ferguson et ah, 2003a). Hargrave and Phillips (1981) note the significance o f short-term organic m atter deposition, w ith high supply rates, relative to m ineralisation rates, caus­ ing a burial o f organic m atter in deeper, anoxic layers o f the sedim ent. 4.4 M odel sensitivity towards sedimentary iron The sensitivity o f the m odel (Fig. 7) to solid-phase iron (Fe) suggests that the delivery o f fresh (oxidised) Fe controls how m uch m ineralised carbon is exported to the atm osphere (as C O 2 ) versus the ocean (as alkalinity). In the sim ulation, a large input o f Fe leads to a higher rate o f alkalinity pro­ duction through F eS 2 burial (FeOOH consum ption), and a corresponding low er C O 2 efflux during exposure (Fig. 4). The sensitivity o f the m odel to Fe agrees w ith the literature (Hu and Cai, 2011; B erner et ah, 1970), w hich suggests that F eS 2 burial is the m ost im portant process controlling alkalin­ ity production. The present w ork show s that in high carbon system s such as intertidal sedim ents, Fe w ill be the lim iting species in pyrite burial, despite the role o f carbon as the m e­ diating species in pyrite burial globally (Hu and Cai, 2011). The tem poral change of TA fluxes also differs consider­ ably. TA flux stabilises after 30 days in the sim ulations w ith the low er Fe concentrations, w hereas at higher Fe concentra­ tions, TA flux continues to increase. This is due to the reduc­ tion o f all the oxidised Fe, creating a higher net production of alkalinity. For sim ulations w ith little or no Fe, there is a higher tem porary sink of H S - in the sedim ent pore w aters during intense alkalinity production (Fig. 7) as insoluble iron sulfides cannot form . Subsequent alkalinity consum ption oc­ curs as the H S - is reoxidised. The effect of the reoxidation of reduced sulfur can be seen in Fig. 7, w hich show s the sim ­ ulation w ith no FeO O H having low er pH and, thus, higher C O 2 concentration, than the sim ulation w ith 95 pm ol F e g - 1 dry sedim ent. T he alkalinity concentration (as indicated by www.biogeosciences.net/9/4087/2012/ P. A. Faber et al.: The role of alkalinity generation 4095 o.o CL <D TJ C 0.3 <D 95 FeOOH 0 FeOOH E I d 0.4 <0 0.5 6 .2 6.4 6 .6 6 .8 7 .0 7 .2 7.4 7.6 7.8 8 .0 0.0 0 .2 pH 0.4 0.6 0.8 1.0 1.2 1.4 C 0 2 (mM in porewater) 0.0 o 0.1 0.2 0 0.3 E I 0-4 0.5 2 .0 2.5 3 .0 3.5 4 .0 4.5 5.0 5 .5 6.0 2.0 2.5 DIC (mM in porewater) 3.0 3.5 4.0 4.5 5.0 1.0 1.2 TA (mM in porewater) 0.0 E o Q- 0 .2 o T5 0.5 0 10 20 ILimol FeOOH 30 40 0.0 0.2 g"1 dry sedim ent 0.4 0.6 0.8 HS" (mM in porewater) o.o 0.1 £-0 - 2 S 0.3 0.5 0 .0 0 0 .0 5 0 .1 0 0 .1 5 0.20 0 .2 5 0.3 0 0.3 5 0 2 (mM in porewater) Fig. 7. Profiles o f pH , C O 2, D IC , TA, FeO O H and H S an d O 2 from tw o sim ulations. E m pty circles represent a sim ulation w ith no FeO O H , w hereas filled circles represent 95 pm ol g _ 1 sedim ent (d / w) FeO O H . HCO^" and C O 3 - ) throughout the FeO O H containing profile is also considerably higher. D uring periods w here respiration is extrem ely high, H S - has the potential to by-pass the sm all oxic zone in the sedim ent and so the H S - can reach the w ater colum n before it is reoxidised. In this situation, the num eric sim ulation w ill show a net alkalinity flux from the sedim ent. In reality, im m ediate reoxidation o f H S - in the overlying w a­ ter colum n w ill directly consum e this alkalinity. W ith regard to carbon budgets, the effect o f anaerobic car­ bon m etabolism is im portant as higher alkalinity export to the ocean creates a higher C O 2 buffering capacity for the u p ­ www.biogeosciences.net/9/4087/2012/ take of atm ospheric C O 2 . The effect o f anaerobic alkalinity production on atm ospheric carbon uptake in coastal seas has been estim ated at up to 60 % o f the total uptake (Thom as et al., 2009). This is an upper estim ate, as it neglects alkalinity consum ed during nitrification (Hu and Cai, 2011). In regions w ith high anthropogenic disturbance, large am ounts o f ter­ restrial carbon is exported tow ards coastal oceans (Frankignoulle et al., 1996, 1998; A bril et al., 2002). Increased organic carbon loading in estuaries due to anthropogenic land use represents an anthropogenic carbon source to the atm osphere upon its breakdown. M ore anthropogenically Biogeosciences, 9, 4087-4097, 2012 4096 disturbed catchm ents w ill also export m ore sedim ent and, hence, m ore Fe(III) (Asselm an et al., 2003; W alling, 1999) to estuaries, w hich may partially offset larger C O 2 em issions from increased labile organic carbon export to estuaries. The significance o f this offset rem ains to be investigated. 5 Conclusions The laboratory data and field data (Cook et al., 2004) and com puter sim ulations clearly show that gaseous C O 2 fluxes during exposure on tidal flats are likely to under estim ate total carbon m ineralisation. This finding has clear im plica­ tions for research in this field, given that total inorganic car­ bon fluxes for intertidal sedim ents have generally been es­ tim ated using only gaseous C O 2 fluxes on the exposed sed ­ im ents (M igné et al., 2005; M iddelburg et al., 1996). The results o f this study dem onstrate the im portance o f consider­ ing both exposed and inundated fluxes in studies o f intertidal m etabolism , w hich m ay be quite different, depending on the extent to w hich there is a net accum ulation o f reduced so ­ lutes. Acknowledgements. P. Faber acknow ledges the assistance o f an A ustralian Postgraduate A ssociation scholarship. T his w ork w as supported by a M onash U niversity faculty o f science early career researcher grant, as w ell as M onash researcher accelerator grant to PC. We thank H ans R oy for stim ulating discussions on this w ork. E dited by: S. W. A. N aqvi References A bril, G., N ogueira, M ., Etcheber, H., C abecadas, G., Lem aire, E., and B rogueira, M .: B ehaviour o f organic carbon in nine contrast­ ing european estuaries, E stuarine, Coast. S h e lf Sei., 54, 2 4 1-262, 2002. A lm gren, T., D yrssen, D., and Fonselius, S.: D eterm ination o f a l­ kalinity and total carbonate, in: M ethods o f seaw ater an aly ­ sis, edited by: G rasshoff, K., E hrhardt, M ., and K rem ling, K., Springer-V erlag, Chem ie, W einheim , 9 9 -1 2 3 , 1983. A longi, D., Tirendi, F., D ixon, P., Trott, L., and B runskill, G.: M in ­ eralization o f organic m atter in intertidal sedim ents o f a tropical sem i-enclosed delta, Estuar. Coast. S h e lf S., 48, 4 5 1 -4 6 7 , 1999. A sselm an, N. E. M ., M iddelkoop, H., and Van D ijk, P. M.: T he im ­ pact o f changes in clim ate and land use on soil erosion, transport and deposition o f suspended sedim ent in the river rhine, H ydrol. Process., 1 7 ,3 2 2 5 -3 2 4 4 ,2 0 0 3 . B erner, R. A., Scott, M . R., and T hom linson, C.: C arbonate alk a­ linity in the pore w aters o f anoxic m arine sedim ents, Lim nol. O ceanogr., 15, 5 4 4 -5 4 9 , 1970. B oudreau, B. P.: The diffusive tortuosity o f fine-grained unlithified sedim ents, G eochim . C osm ochim . A c., 60, 3 1 3 9 -3 1 4 2 , 1996. B oudreau, B. P.: D iagenetic m odels and their im plem entation: M od­ elling transport and reactions in aquatic sedim ents, Springer, B erlin and N ew York, 1997. Biogeosciences, 9, 4087-4097, 2012 P. A. Faber et al.: The role of alkalinity generation B rotas, V., A m orim -F erreira, A., Vale, C., and Catarino, F.: O x y ­ gen profiles in intertidal sedim ents o f ria form osa (s. Portugal), H ydrobiologia, 207, 1 2 3-130, 1990. Cai, W., Pom eroy, L., M oran, M ., and W ang, Y.: O xygen and carbon dioxide m ass balance for the estuarine-intertidal m arsh com plex o f five rivers in the southeastern us, Lim nol. O ceanogr., 44, 6 3 9 649, 1999. Cai, W. J., Luther, G. W., C ornw ell, J. C., and G iblin, A. E.: C ar­ bon cycling and the coupling betw een proton and electron tran s­ fer reactions in aquatic sedim ents in lake cham plain, A quat. G eochem ., 16, 4 2 1 -4 4 6 ,2 0 1 0 . Cook, P., Butler, E., and Eyre, B.: C arbon and nitrogen cycling on intertidal m udflats o f a tem perate australian estuary i. B enthic m etabolism , Mar. E col.-Prog. Ser., 280, 2 5 -3 8 , 2004. Ferguson, A. J. P., Eyre, B. D., and Gay, J. M .: O rganic m atter and benthic m etabolism in euphotic sedim ents along shallow su b ­ tropical estuaries, northern new south w ales, australia, A quat. M icrob. Ecol., 33, 137-154, 2003. Follow s, M . J., Ito, T., and D utkiew icz, S.: O n the solution o f the carbonate chem istry system in ocean biogeochem istry m odels, O cean M odel., 12, 2 9 0 -3 0 1 , 2006. Frankignoulle, M ., Bourge, I., and W ollast, R.: A tm ospheric C02 fluxes in a highly polluted estuary (the scheldt), Lim nol, Oceanogr,, 41, 36 5 -3 6 9 , 1996. Frankignoulle, M ., A bril, G., Borges, A., Bourge, I., C anon, C., D elille, B., L ibert, E., and Théate, J.: C arbon dioxide em ission from european estuaries, Science, 282, 4 3 4 -4 3 6 , 1998. G ribsholt, B. and K ristensen, E.: B enthic m etabolism and sulfur cy­ cling along an inundation gradient in a tidal spartina anglica salt m arsh, Lim nol. O ceanogr., 48, 2 1 5 1 -2 1 6 2 , 2003. H am m ond, D., G iordani, P., B erelson, W., and Poletti, R.: D iag en e­ sis o f carbon and nutrients and benthic exchange in sedim ents o f the northern adriatic sea, M ar. C heni., 66 , 53 -7 9 , 1999. Hargrave, B. and Phillips, G.: A nnual in situ carbon dioxide and oxygen flux across a subtidal m arine sedim ent, Estuar. Coast. S h e lf S., 12, 7 2 5 -7 3 7 , 1981. H ofm ann, A. F., M eysm an, F. J. R., Soetaert, K., and M iddelburg, J. J.: A step-by-step procedure for pH m odel construction in aquatic system s, B iogeosciences, 5, 2 2 7 -2 5 1 , doi:10.5194/bg-5227-2008, 2008. H ofm ann, A. F., Soetaert, K., M iddelburg, J. J., and M eysm an, F. J. R.: A quaenv: A n aquatic a cid -b ase m odelling environm ent in r, A quat. G eochem ., 16, 50 7 -5 4 6 , 2010. Hu, X. and Cai, W. J.: A n assessm ent o f ocean m argin anaerobic processes on oceanic alkalinity budget, G lobal B iogeochem . Cy., 25, 1 -1 1 ,2 0 1 1 . Joye, S. B., de Beer, D., C ook, P. L. M ., and Perillo, G.: B iogeo­ chem ical dynam ics o f coastal tidal flats, in: C oastal w etlands: an integrated ecosystem approach, edited by: Perillo, G., Cahoon, D., and B rinson, M ., Elsevier, A m sterdam , T he N ether­ lands, 3 4 5 -3 7 4 , 2009. K ristensen, E.: B enthic fauna and biogeochem ical processes in m a ­ rine sedim ents: M icrobial activities and fluxes, in: N itrogen cy­ cling in coastal m arine environm ents, edited by: B lackburn, T. H. and Sorensen, J., Jo h n W iley and Sons, Chichester, 2 7 5 -2 9 9 , 1988. Lord, C. J.: A selective and precise m ethod for pyrite determ ination in sedim entary m aterials, J. Sedim ent. Res., 52, 6 6 4 -6 6 6 , 1982. www.biogeosciences.net/9/4087/2012/ P. A. Faber et al.: The role of alkalinity generation M eysm an, F. J. R., M iddelburg, J. J., H erm an, P. M . J., and Heip, C. H. R.: R eactive transport in surface sedim ents, II. M edia: A n object-oriented problem -solving environm ent for early diagene­ sis, C om put. G eosci., 29, 3 0 1 -3 1 8 , 2003. M iddelburg, J., Klaver, G., N ieuw enhuize, J., W ielem aker, A., De H aas, W., V lug, T., and Van der N at, J.: O rganic m atter m ineral­ ization in intertidal sedim ents along an estuarine gradient, Mar. E col.-Prog. Ser., 132, 1 5 7-168, 1996. M igné, A., D avoult, D., B ourrand, J. J., and B oucher, G.: B en­ thic prim ary production, respiration and rem ineralisation: In situ m easurem ents in the soft-bottom abra alba com m unity o f the w estern english channel (north brittany), J. Sea Res., 53, 2 2 3 229, 2005. R aym ond, P., Bauer, J., and Cole, J.: A tm ospheric C02 evasion, d is­ solved inorganic carbon production, and net heterotrophy in the york river estuary, Lim nol. O ceanogr., 45, 1 7 0 7 -1717, 2000. Roy, R., Roy, L., Vogel, K., Porter-M oore, C., Pearson, T., G ood, C., M illero, F., and C am pbell, D.: The dissociation constants o f carbonic acid in seaw ater at salinities 5 to 45 and tem peratures 0 to 45 °C , M ar. Cheni., 44, 2 4 9 -2 6 7 , 1993. www.biogeosciences.net/9/4087/2012/ 4097 Satienperakul, S., C ardw ell, T. J., Cattrall, R. W., M cK elvie, I. D., Taylor, D. M ., and Kolev, S. D.: D eterm ination o f carbon dioxide in gaseous sam ples by gas diffusion-flow injection, Talanta, 62, 6 3 1 -6 3 6 ,2 0 0 4 . Soetaert, K. and M eysm an, F.: Reactive transport in aquatic eco sy s­ tem s: R apid m odel prototyping in the open source softw are r, Environ. M odel. Softw., 32, 4 9 -6 0 , 2012. Soetaert, K., P etzoldt, T., and M eysm an, F. J. R.: M arelac: Tools for aquatic sciences, R package version 2.1.1., 2010. T hom as, H., Schiettecatte, L. S., Suykens, K., K oné, Y. J. M ., Shadw ick, E. H., Prow e, A. E. F., B ozec, Y., de Baar, H. J. W., and Borges, A. V.: E nhanced ocean carbon storage from anaerobic a l­ kalinity generation in coastal sedim ents, B iogeosciences, 6, 2 6 7 274, doi:10.5194/bg-6-267-2009, 2009. W alling, D.: L inking land use, erosion and sedim ent yields in river basins, H ydrobiologia, 410, 2 2 3 -2 4 0 , 1999. W estrich, J. T. and Berner, R. A.: T he role o f sedim entary organic m atter in bacterial sulfate reduction: The g m odel tested, Lim nol. O ceanogr., 29, 2 3 6 -2 4 9 , 1984. Biogeosciences, 9, 4087-4097, 2012