Are recent changes in sediment manganese sequestration in the euxinic basins of the Baltic Sea linked to the expansion of hypoxia?

Lenz, C.; Jilbert, T.; Conley, D.J.; Wolthers, M.; Slomp, C.P.

doi:https://doi.org/10.5194/bg-12-4875-2015

Articles | Volume 12, issue 16

https://doi.org/10.5194/bg-12-4875-2015

© Author(s) 2015. This work is distributed under
the Creative Commons Attribution 3.0 License.

https://doi.org/10.5194/bg-12-4875-2015

© Author(s) 2015. This work is distributed under
the Creative Commons Attribution 3.0 License.

Articles | Volume 12, issue 16

Research article

|

19 Aug 2015

Research article |

| 19 Aug 2015

Are recent changes in sediment manganese sequestration in the euxinic basins of the Baltic Sea linked to the expansion of hypoxia?

C. Lenz, T. Jilbert, D.J. Conley, M. Wolthers, and C.P. Slomp

Download

Final revised paper (published on 19 Aug 2015)
Supplement to the final revised paper
Preprint (discussion started on 25 Jun 2014)
Supplement to the preprint

Interactive discussion

Status: closed

AC: Author comment | RC: Referee comment | SC: Short comment | EC: Editor comment

- Printer-friendly version

- Supplement

RC C3404: 'Comment on manganese sequestration in Baltic Sea euxinic basins', Anonymous Referee #1, 10 Jul 2014
- AC C5600: 'Reply to reviewer 1', Conny Lenz, 01 Oct 2014
RC C3646: 'Review of: Are recent changes in sediment manganese sequestration in the euxinic basins of the Baltic Sea linked to the expansion of hypoxia?', Anonymous Referee #2, 21 Jul 2014
- AC C5616: 'Reply to reviewer 2', Conny Lenz, 01 Oct 2014

Peer-review completion

AR: Author's response | RR: Referee report | ED: Editor decision

AR by Conny Lenz on behalf of the Authors (17 Oct 2014) Author's response Manuscript

ED: Referee Nomination & Report Request started (31 Oct 2014) by Leticia Cotrim da Cunha

RR by Anonymous Referee #3 (13 Nov 2014)

RR by Anonymous Referee #1 (14 Nov 2014)

RR by Anonymous Referee #4 (28 Dec 2014)

Suggestions for revision or reasons for rejection

The purpose of the present study is to characterize processes of Mn enrichment in Baltic Sea sediments according to recent changes due to expanding hypoxia since the 80s.

I am not familiar with the specific physical and environmental processes that take place in the Baltic Sea. Therefore my review will focus on biogeochemical processes and data described in this manuscript.
First of all, I was struck by the lack of some data and calculations to support the conclusions and by the lack of some alternative explanations that could also explain the data. This is detailed below.
1) The sediment characteristics are not given : is the grain size constant? what is the profile of porosity? These parameters are needed to verify the flux calculation. The grain size of the different stations is a relevant piece of information to compare the stations. If the grain size changes within a vertical profile, the story is no more the same.
2) Some pore water and solid fraction data are lacking : how can the authors calculate the saturation index of a Mn-Ca carbonate without dissolved Ca2+ data? The redox geochemistry of Mn depends on the geochemistry of several electron donors or acceptors, such as carbon species, nitrogen species, oxygen, sulphur species, and iron species. Here, we have an interesting collection of data that includes high vertical resolution profiles of dissolved Mn, solid Mn, TOC, dissolved ammonium, particulate S and Mo, dissolved sulphate and sulphide, and alkalinity. Some samples of bottom water data are lacking (sulphate in Landsort Deep). Nothing about iron species. However, it is well known that Fe2+ is a reducing compound for Mn-oxides. Nothing about methane. Methane is probably a key parameter in the Landsort Deep, where the TOC content is very high and sulphate is close to zero. Methane is a reducing agent for Mn and Fe-oxides; methanogenesis may have a control on CO3 content (i.e. on solid carbonate stability). I am quite convinced that some iron (and maybe methane) data have been obtained by the authors. Did they attempt to publish these data in another paper? If these data exist, they must be included in the present study. If they do not exist the data set is incomplete and the conclusions given here are not valid.
3) Some simple mass balance calculations could support, confirm, or invalidate some hypothesis presented in the paper.

Detailed comments :
First sentence of the introduction : “Manganese (Mn) enrichments in sedimentary deposits are often used as an indicator of redox changes in the overlying waters” is not valid. Write “could be” instead of “are often”, because Mn-rich deposits have been used for other purpose, such as sedimentary source (in the arctic), sedimentary processes (contourite, turbidite), which have nothing to do with changes in the redox state of the water column.

Introduction : p2, line 17 : Mn-enrichment in permanently anoxic water also occurs in some hydrothermal settings (e.g. brine-filled deeps in the Red Sea)

Introduction p2 line 25 : Mercone et al 2001 is not an appropriate reference here, since it refers to sapropel deposition (which includes anoxic events)

Introduction p2 lines 29-30 : “The Baltic Sea provides an ideal environment for studies of redox-dependent Mn dynamics because of the large spatial and temporal variations in oxygen conditions over the past century” I am not sure that a place that experiences so many changes is an “ideal” environment to try to understand the Mn-redox behaviour, apart from when all the needed parameters are provided, which is not the case here.

Introduction p3 line 18 : see the work by Mouret et al 2009 (Benthic geochemistry of manganese in the Bay of Biscay, and the sediment accumulation rate. GeoMarine Letters 29:133–149) that describes the relationship between Mn-inventory in sediment and the sediment accumulation rate

Methods p5, lines 25 and 28 : do you mean 0.3 and 0.125 g (not mg)

Methods p5, line 29 : What about AVS? during acid attack, you may have lost sulphur from AVS, which is probably a major fraction of particulate S.

Methods : p6 line 3 and appendix1 : age model. The age model is valid only if the sediment is homogeneous (sieved sediment or sediment with initial constant grain size). Data of mass accumulation rate (appendix) can be validated only if the authors provide the porosity or the water content data. The appendix shows a column named “Observed 210Pb activity(mBq g‐1,salt corr.)” with values that are very different from initial data. This must be explained.

section 2.6 : saturation states : Ca2+ data not shown; pH data probably with an error (because of centrifugation, filtration) : this must be corrected or explained. A small error in pH change considerably the carbonate ion activity calculated with PHREEQ

p9, line 15-17 : “Mn is enriched in the surface sediment at Fladen, but is nearly absent at the LF1, BY5 and LF3 sites. At sites LL19, BY15, F80 and LD1, Mn is present again but is mostly observed at greater depth in the sediment. Sediment Mo is low at the Fladen, LF1, BY5 and LF3 sites…” Even if I read the manuscript carefully, I had difficulties to bring the sites symbols back to geographic sites. BY15, F80, LL19… should be changed by other names in the papers.

p11, line 11 : authors should provide the mass balance calculation : what is the Mn-inventory in the water column? What is the bottom surface area located below the redoxcline? …

p11, line 19-20. The conclusion given here are not supported by a quantitative discussion.

p12, discussion about Landsort Deep : The very low concentration in sulphate suggests that methane is present (probably in high amount). Author should explore the scientific literature that deals with lake biogeochemistry, because the sulphate depleted Landsort Deep sediments are like an organic-rich lake sediment. In the Landsort Deep CH4 represents probably a second source of organic C coming from below. This may have an impact on the whole benthic biogeochemistry.
Authors claimed that there is a deep pore water source of dissolved Mn in the Landsort Deep because the concentrations of Mn are higher than in other stations. I would rather suggest that dissolved Mn2+ is controlled by the equilibrium with a carbonate in the other stations, whereas it is not controlled by the same solid phase in the Landsort Deep.

The pore water Mn concentrations are by far the highest at the Landsort Deep site (> 1 mM versus <0.26 mM of Mn2+). According to the authors, this “deep” diffusive supply of Mn to the surface sediment may in fact explain why the formation of Mn enrichments at this site is more continuous than at the other sites :
A simple mass balance could be made to (in)validate this hypothesis : authors should compare the diffusive flux with the accumulation of authigenic Mn (based on sediment accumulation rate and Mn-inventory calculation)

p. 13, line 21 : what do the authors means with “the most euxinic periods”. Is it the periods with the highest S(-II) concentrations?

H2S concentration in the figures are showed as negative oxygen concentrations. Here the real S(-II) data should be showed, in order to a avoid to add complexity in the manuscript.

p. 14 line 13 “But what can inhibit the formation of these Mn carbonates?” the simplest explanation is not put forward here : the upper sediment is not enriched in Mn-carbonate, because this part of the sediment pore water is undersaturated with respect to a carbonate phase (see the fig. 3). The saturation is reach only below 10 cm depth approximately in these stations.

other hypothesis : high Fe2+ concentration at the sediment-water interface that prevent the accumulation of Mn-oxide (but Fe data are lacking)

NH4 data are presented here. However, there is no discussion concerning the role of NH4 as a reducing compound for Mn-oxide, despite the abundant literature on this topic.

P. 15 conclusion : first sentence : this conclusion must be corroborated by a mass balance calculation.

conclusion p15 lines 25-28 :” We show that the most recent sediments in the Fårö Deep and Gotland Deep contain low concentrations of Mn near the sediment surface. We postulate that this is due to the expansion of hypoxia over the past decades with the Mn oxides formed during inflows from the North Sea often being reduced so rapidly that the Mn2+ is lost to the water column.”
This text suggests that the Mn content is not simply low, but that is it lower than what it should be. However nothing in the ms support this conclusion. Low concentration of Mn in anoxic surface sediment is very common.

conclusion p15 line 31 : “Our results indicate that sediment Mn carbonates in the Baltic Sea no longer reliably record inflows of oxygenated North Sea water.” the “no longer” suggests that Mn carbonates reliably recorded inflows in the past. This has not been demonstrated here, nor presented in the introduction based on literature.

Tabl 1 : oxygen concentrations given in mL/L. Please use molar units

Hide

RR by P. L. Croot (24 Feb 2015)

Suggestions for revision or reasons for rejection

Overview:
This manuscript presents data from studies into the sediment chemistry from a series of stations in the Baltic Sea with a focus on Mn and its potential as a tracer of oxygenation events in the anoxic deep waters. The experimental approach is very strong, well described, and state of the art and the article is well written to the point and concise. The authors interpret their dataset in terms of differing modes for Mn carbonate formation in the sediments in response to periodic injections of oxygenated waters into anoxic deep waters. In the Gotland deep they suggest that MnCO3 previously formed from MnO2 that was precipitated following North Sea inflows, but now is absent as the anoxia is now basin wide. While in the Landsort Deep, MnCO3 and MnS form independent of inflow events as there is a deep pore water source of reduced Mn. These findings have important implications in terms of paleo reconstructions of oxygenation events. However at present the paper lacks an appreciation of the time scales involved nor the potential for other controlling processes (e.g. pyrite formation and/or basin shape/size) and how this impacts the Mn inventories and concentrations in the bottom waters and porewaters. There is a frequent referral to a potential recent change in that the generation of H2S after an oxygenation event is faster now but no explanation is given of how this can occur. Waht would be most useful however is some sort of description or evaluation (model or otherwise) of what the tipping point is for a record of an oxygenation event to be recorded. For instance if there is a short interval between oxygen events is it recorded as one or two events, alternatively if there has been no event for several years and sulfide has built up appreciably, how big an event has to occur for it to be recorded in say either the Landsort or Gotland basins? Overall though this manuscript should be acceptable after moderate revision as outlined below.

General Comments:

Timescales of response to an oxygenation event:
One aspect that is poorly explored in the manuscript at present is how quickly the system responds to an oxygenation event and how quickly does it return to the previous state. Laboratory data indicates that MnS is apparently oxidized rapidly when re-suspended in oxygenated seawater (Simpson et al., 1998; Zhang and Millero, 1994) and may form MnO2 or even Mn(III) species. Simple abiotic oxidation of Mn(II) is very slow (von Langen et al., 1997) but is accelerated by bacteria (de Vrind et al., 1986; Sujith and Bharathi, 2011; Tebo and Emerson, 1986; Tebo, 1991; Tebo et al., 2005). There are a also works that look at the kinetics of reduction of MnO2 by H2S (Yao and Millero, 1993) and this information would be useful for looking at the timescale that is involved on that side of the redox cycle.

Reactive Mn, Mn(III) in the Baltic:
A difference in reactive Mn between the Gotland deep and Landsort deep has been commented on recently in another work (Dellwig et al., 2012). In their study they found almost no reactive Mn, thought to be Mn(III) species, in the Gotland deep. While in the Landsort deep there was a strong gradient of increasing reactive Mn towards the central part of the basin. These authors suggested lateral intrusions of oxygen into Gotland deep prevented the formation of stable suboxic zone where reactive Mn could accumulate. It would be useful to discuss then the potential for reactive Mn to impact the formation of MnCO3 etc as presumably an oxygenation event may lead to an increase in Mn(III) within the sediments (Luther III et al.; Madison et al., 2011; Oldham et al.; Trouwborst et al., 2006) that may be supported by organic complexing agents found there.

It should also be noted that there is now considerable Mn data from the Baltic (Dellwig et al., 2012; Kashiwabara et al., 2009; Neretin et al., 2003; Pakhomova and Yakushev, 2013; Pohl and Hennings, 2005; Yakushev et al., 2009) which illustrate that Mn in the anoxic waters in the Baltic can vary appreciably due to the oxygenation events and this needs to be taken into account. The recent work by Pakhomova and Yakushev (2013) is a good place to start as they provide a minor compilation of values and a modelling effort. In this context the recent works on Mn cycling at the redoxcline in the Baltic (Dellwig et al., 2012; Schnetger and Dellwig, 2012) should also be included in the discussion as during an oxygenation event in the deep waters the same sub-oxic chemistry will be brought into play in the sediments.

Mn incorporation into pyrite:
One aspect of Mn chemistry in anoxic environments that appears to be under explored here but could be a key pathway is the earlier finding that dissolved Mn can adsorb onto FeS phases and be pyritized (Morse and Luther III, 1999). These authors found that Mn was more readily incorporated into pyrite when the degree of pyritization (pyrite Fe/(pyrite Fe + reactive Fe) was above 40%. Thus the coexistence of Mn and S in the sediments may also be related to FeS availability and not MnS formation per se. In this case the ratio of Fe to Mn, the degree of pyritization, as well as the sulfide concentration would be important. Pyrite has already been shown to be an important species for transferring Fe to the sediments in the Gotland deep (Fehr et al., 2010). Information on the pyrite and Fe(II) content in the sediments/porewaters would also then help to gauge if this was having an impact on the Mn cycle.

The age of Baltic seawater and mixing in the Gotland Basin:
One of the major themes of this paper is the influx of oxygenated water into the different basins of the Baltic. It would therefore be of interest to link the present work with some recent physical oceanographic studies of the water mass age in the Baltic (Meier, 2005) and the Baltic sea mixing experiment BATRE (Holtermann and Umlauf, 2012; Holtermann et al., 2012). In the latter an inert tracer (SF5CF3) was injected into the Gotland Basin at 200 m depth and followed over 18 months or so. BATRE is useful in the present context as it points to the role of typography in the mixing of the bottom waters in this region and to the potential response time of an oxygenation event.

Mn carbonate formation and solubility:
There are a number of other recent works that have looked at the formation and solubility of Mn carbonates in seawater (Jensen et al., 2002; Johnson, 1982; Luo and Millero, 2003; Neher-Neumann, 1994; Radha and Navrotsky, 2014; Wartel et al., 1990). Note that there are also papers that deal with the transformation of aragonite/calcite to Mn carbonate (Böttcher, 1997; Böttcher, 1998) and this pathway should be mentioned also in the discussion along with data on the carbonate system in the water column (Schneider et al., 2010; Ulfsbo et al., 2011). This would then help to make the case that MnCO3 formation is only occurring in the uppermost part of the sediments.

Molybdenum:
The conversion of molybdate to reduced molybdenum species is reasonably complex and depends on the sulfide concentration (Erickson and Helz, 2000; Wang et al., 2009; Wang et al., 2011), so it would be useful to include a line or two on this. Also it is worth pointing out the strong conservative behavior for Molybdenum in the Baltic (Prange and Kremling, 1985) as due to the low salinities encountered, there are strong vertical gradients in Molybdenum that are not found in other regions. This makes it harder to see small losses of Mo due to scavenging.

Specific Comments:

P3 Line 30: So how does this mechanism work is it that the pore water Mn2+ is lower as the reduction takes place in the water column?
P7 Line 20: Inclusion of the seawater value and error estimates for the flux would be useful to include either here or later in the manuscript as a check to the validity of the flux estimates.
P8 Line 18: Is the data on the saturation for FeS included here in the manuscript? As I could not find it listed in the main text.
P9 Line 4: It would be useful to also provide here the saturation status for FeS at these sites?
P10 Line 25: In terms of the key processes it would be helpful to state here that release of Mn from sediments with MnO2 present is driven by redox processes, while in the case of MnCO3 it is controlled by the concentration of the solutes and there is no apparent redox change for Mn.
P11 Line 2: Is there no Mn incorporation into pyrite occurring then?
P11 Line 26: Is this controlled by the high Mn in the overlying anoxic waters? If so how much of a reservoir effect is this?
P11 Line 28: When the oxygenated water comes in to the basin it will obviously take some time to oxidize the bulk of the Mn2+ (Yakushev et al., 2011) and to form that MnO2 that is deposited to the sediment so there is an important kinetic effect here. Also this thin layer of MnO2 is then presumably reduced at the sediment interface resulting in a locally high concentration of Mn2+ which leads to the formation of MnCO3. This all relies on slow diffusion and no resuspension of the Mn porewaters in the sediments.
P14 Line 24: Why does the sulfide return more rapidly? High overlying productivity driving more rapid deoxygenation?
P14 Line 32: I would agree with this as MnO2 in the water column most likely is associated with an organic layer which would reduce its ability to scavenge Molybdate. Most studies that have looked into the ability of MnO2 to scavenge molybdate do not consider the effect of organics (Kashiwabara et al., 2009).
P15 Line 24: This sentence needs some qualification as while there might be 8 µM Mn(II) in the bottom waters and shallowest porewaters this might still be insufficient to precipitate out MnCO3 if the carbonate concentration is too low (low pH and/or low alkalinity) (Böttcher, 1997; Böttcher, 1998). There would also presumably be a sulfide control on the Mn(II) concentration.
P15 Line 27: Is the expansion in terms of horizontal area and/or vertical extent? Is there any data that the inventory of Mn in the water column is increasing?
Table 2: It would be helpful here to also include the Mn(II) concentration in both the overlying water and at the top of the porewater profile to show that the flux is real. In this context if a dissolved Mn inventory for the anoxic waters was constructed it would allow a turnover time for this pool to be estimated using the fluxes listed here. This would help to answer the question of how long it takes to resupply the bottom waters after an oxygenation event.

References cited:

Böttcher, M.E., 1997. The transformation of aragonite to MnxCa(1 − x)CO3 solid-solutions at 20 °C: An experimental study. Marine Chemistry, 57(1–2): 97-106.
Böttcher, M.E., 1998. Manganese(II) partitioning during experimental precipitation of rhodochrosite–calcite solid solutions from aqueous solutions. Marine Chemistry, 62(3–4): 287-297.
de Vrind, J.P.M. et al., 1986. Manganese Oxidation by Spores and Spore Coats of a Marine Bacillus Species. Appl. Environ. Microbiol., 52(5): 1096-1100.
Dellwig, O., Schnetger, B., Brumsack, H.J., Grossart, H.P. and Umlauf, L., 2012. Dissolved reactive manganese at pelagic redoxclines (part II): Hydrodynamic conditions for accumulation. Journal of Marine Systems, 90(1): 31-41.
Erickson, B.E. and Helz, G.R., 2000. Molybdenum(VI) speciation in sulfidic waters: Stability and lability of thiomolybdates. Geochimica Et Cosmochimica Acta, 64(7): 1149-1158.
Fehr, M.A., Andersson, P.S., Hålenius, U., Gustafsson, Ö. and Mörth, C.-M., 2010. Iron enrichments and Fe isotopic compositions of surface sediments from the Gotland Deep, Baltic Sea. Chemical Geology, 277(3–4): 310-322.
Holtermann, P.L. and Umlauf, L., 2012. The Baltic Sea Tracer Release Experiment: 2. Mixing processes. J. Geophys. Res., 117(C1): C01022.
Holtermann, P.L. et al., 2012. The Baltic Sea Tracer Release Experiment: 1. Mixing rates. J. Geophys. Res., 117(C1): C01021.
Jensen, D.L., Boddum, J.K., Tjell, J.C. and Christensen, T.H., 2002. The solubility of rhodochrosite (MnCO3) and siderite (FeCO3) in anaerobic aquatic environments. Applied Geochemistry, 17(4): 503-511.
Johnson, K.S., 1982. Solubility of rhodochrosite (MnCO3) in water and seawater. Geochimica et Cosmochimica Acta, 46(10): 1805-1809.
Kashiwabara, T., Takahashi, Y. and Tanimizu, M., 2009. A XAFS study on the mechanism of isotopic fractionation of molybdenum during its adsorption on ferromanganese oxides. Geochemical Journal, 43(6): E31-E36.
Luo, Y. and Millero, F., 2003. Solubility of Rhodochrosite (MnCO3) in NaCl Solutions. Journal of Solution Chemistry, 32(5): 405-416.
Luther III, G.W., Madison, A.S., Mucci, A., Sundby, B. and Oldham, V.E., A kinetic approach to assess the strengths of ligands bound to soluble Mn(III). Marine Chemistry(0).
Madison, A.S., Tebo, B.M. and Luther Iii, G.W., 2011. Simultaneous determination of soluble manganese(III), manganese(II) and total manganese in natural (pore)waters. Talanta, 84(2): 374-381.
Meier, H.E.M., 2005. Modeling the age of Baltic Seawater masses: Quantification and steady state sensitivity experiments. Journal of Geophysical Research: Oceans, 110(C2): C02006.
Morse, J.W. and Luther III, G.W., 1999. Chemical influences on trace metal-sulfide interactions in anoxic sediments. Geochimica et Cosmochimica Acta, 63(19–20): 3373-3378.
Neher-Neumann, E., 1994. Studies on metal carbonate equilibria 26 - The Hydrogen carbonate complex of manganese(II) in acid solutions and a 3 M (Na)ClO4 ionic medium at 25-degrees C. Determination of the solubility product of MnCO3(s) Acta Chemica Scandinavica, 48(5): 393-398.
Neretin, L.N., Pohl, C., Jost, G., Leipe, T. and Pollehne, F., 2003. Manganese cycling in the Gotland Deep, Baltic Sea. Marine Chemistry, 82(3-4): 125.
Oldham, V.E., Owings, S.M., Jones, M., Tebo, B.M. and Luther Iii, G.W., Evidence for the presence of strong Mn(III)-binding ligands in the water column of the Chesapeake Bay. Marine Chemistry(0).
Pakhomova, S. and Yakushev, E.V., 2013. Manganese and Iron at the Redox Interfaces in the Black Sea, the Baltic Sea, and the Oslo Fjord. In: E.V. Yakushev (Editor), Chemical Structure of Pelagic Redox Interfaces. The Handbook of Environmental Chemistry. Springer Berlin Heidelberg, pp. 67-93.
Pohl, C. and Hennings, U., 2005. The coupling of long-term trace metal trends to internal trace metal fluxes at the oxic–anoxic interface in the Gotland Basin (57°19,20′N; 20°03,00′E) Baltic Sea. Journal of Marine Systems, 56(1–2): 207-225.
Prange, A. and Kremling, K., 1985. Distribution of dissolved molybdenum, uranium and vanadium in Baltic Sea waters. Marine Chemistry, 16: 259-274.
Radha, A.V. and Navrotsky, A., 2014. Manganese carbonate formation from amorphous and nanocrystalline precursors: Thermodynamics and geochemical relevance. American Mineralogist, 99(5-6): 1063-1070.
Schneider, B., Nausch, G. and Pohl, C., 2010. Mineralization of organic matter and nitrogen transformations in the Gotland Sea deep water. Marine Chemistry, 119(1–4): 153-161.
Schnetger, B. and Dellwig, O., 2012. Dissolved reactive manganese at pelagic redoxclines (part I): A method for determination based on field experiments. Journal of Marine Systems, 90(1): 23-30.
Simpson, S.L., Apte, S.C. and Batley, G.E., 1998. Effect of Short-Term Resuspension Events on Trace Metal Speciation in Polluted Anoxic Sediments. Environmental Science & Technology, 32(5): 620-625.
Sujith, P.P. and Bharathi, P.A.L., 2011. Manganese Oxidation by Bacteria: Biogeochemical Aspects
Molecular Biomineralization. In: W.E.G. Müller (Editor). Progress in Molecular and Subcellular Biology. Springer Berlin Heidelberg, pp. 49-76.
Tebo, B. and Emerson, S., 1986. Microbial manganese(II) oxidation in the marine environment: a quantitative study. Biogeochemistry, 2(2): 149-161.
Tebo, B.M., 1991. Manganese(II) oxidation in the suboxic zone of the Black Sea. Deep Sea Research Part A. Oceanographic Research Papers, 38, Supplement 2(0): S883-S905.
Tebo, B.M., Johnson, H.A., McCarthy, J.K. and Templeton, A.S., 2005. Geomicrobiology of manganese(II) oxidation. Trends in Microbiology, 13(9): 421-428.
Trouwborst, R.E., Clement, B.G., Tebo, B.M., Glazer, B.T. and Luther, G.W., 2006. Soluble Mn(III) in suboxic zones. Science, 313(5795): 1955-1957.
Ulfsbo, A., Hulth, S. and Anderson, L.G., 2011. pH and biogeochemical processes in the Gotland Basin of the Baltic Sea. Marine Chemistry, 127(1–4): 20-30.
von Langen, P.J., Johnson, K.S., Coale, K.H. and Elrod, V.A., 1997. Oxidation kinetics of manganese (II) in seawater at nanomolar concentrations. Geochimica Et Cosmochimica Acta, 61(23): 4945-4954.
Wang, D., Aller, R.C. and Sañudo-Wilhelmy, S.A., 2009. A new method for the quantification of different redox-species of molybdenum (V and VI) in seawater. Marine Chemistry, 113(3–4): 250-256.
Wang, D., Aller, R.C. and Sañudo-Wilhelmy, S.A., 2011. Redox speciation and early diagenetic behavior of dissolved molybdenum in sulfidic muds. Marine Chemistry, 125(1-4): 101-107.
Wartel, M., Skiker, M., Auger, Y. and Boughriet, A., 1990. Interaction of manganese(II) with carbonates in seawater: Assessment of the solubility product of MnCO3 and Mn distribution coefficient between the liquid phase and CaCO3 Particles. Marine Chemistry, 29(0): 99-117.
Yakushev, E., Pakhomova, S., Sørenson, K. and Skei, J., 2009. Importance of the different manganese species in the formation of water column redox zones: Observations and modeling. Marine Chemistry, 117(1–4): 59-70.
Yakushev, E.V. et al., 2011. Modeling the influence of oxygenated inflows on the biogeochemical structure of the Gotland Sea, central Baltic Sea: Changes in the distribution of manganese. Computers & Geosciences, 37(4): 398-409.
Yao, W. and Millero, F.J., 1993. The rate of sulfide oxidation by [delta]MnO2 in seawater. Geochimica et Cosmochimica Acta, 57(14): 3359.
Zhang, J.-Z. and Millero, F.J., 1994. Investigation of metal sulfide complexes in sea water using cathodic stripping square wave voltammetry. Analytica Chimica Acta, 284: 497-504.

Hide

ED: Reject (01 Dec 2014) by Leticia Cotrim da Cunha

ED: Reconsider after major revisions (06 Mar 2015) by Christoph Heinze

AR by Conny Lenz on behalf of the Authors (13 May 2015) Author's response Manuscript

ED: Referee Nomination & Report Request started (30 May 2015) by Katja Fennel

RR by Anonymous Referee #6 (19 Jun 2015)

Suggestions for revision or reasons for rejection

The authors investigated a very dynamic environment and one that is likely to yield great insights on the diagenetic behavior of manganese and the formation of manganese carbonates, an environment and topic to which I have given much thought. As such, this is a potentially very important contribution that will attract much attention among carbonate geochemists, diageneticists, and paleoceanographers. Hence, I read the manuscript with great interest.
The manuscript is clearly worthy of publication, but makes for very difficult reading because of the authors fail to properly describe the sedimentary environment, clearly state their working hypothesis and supporting evidence early on in the discussion. Instead the readers is kept guessing and many of the authors explanations initially appear as purely speculative. This being said, I believe that the manuscript could readily be modified to make it acceptable for publication. Detail comments follow. As noted above, many of my comments (choice of sampling methodology, sediment and pore water extraction procedures, handling of sediment cores and analysis) and frustrations about this manuscript stem from the lack of information (e.g., physical and chemical characteristics of the sediment and limitations) provided by the authors in the text. For example, is the porosity of the surface sediments really 99% as indicated in Table 3? If so, this explains why the authors can suck the sediment out of the cores, why the pore waters can easily be extracted by centrifugation, why the cores must embed their cores in resin prior to line scan analyses. The sediment porosities can be found in the appendices, but a one-sentence statement about this in the text would have eliminated much confusion.
Although fairly well written, the manuscript is long and makes for difficult reading. It is replete with grammatical errors, the abusive use of “However” at the beginning of sentences, the inappropriate use of “While” instead of “Whereas”, and the awkward and frequent use of “in line” where ‘consistent’ would have been more appropriate. For the benefit of the authors, I have edited most of the manuscript and attach an annotated version to this review.
Specific comments:
Abstract –
Page 1, lines 22-23: Where would this large pool originate from, if not from the accumulation of Mn diffusing out of the sediment?
Page 2, line 5: Why is the input of reactive Fe to these basins lower than in recent history?
(Note that the authors eventually, late in the discussion, provide an explanation and, thus, keep the reader guessing, speculating, and questioning the authors’ interpretations until the end. This makes for a poor manuscript.)

Introduction –
Page 2, line 28: derived from sulfate reduction? (see comment above in parentheses)

Materials and Methods –
Page 5, lines 19-20: Is this water truly representative of the bottom water as pore waters may be advected out of the sediment during corer penetration? This problem, which is normally neglected, is compounded by the small volume of overlying water recovered by the multi-corer. Given the high porosity of the sediment, this is not likely a problem, but, as indicated above, the authors should let the reader know of the high sediment porosity instead of the reader having to seek out this information in the appendices. Given the high sediment porosity, how was the sediment sampled with a multi-corer that cannot land/rest on a sediment surface? The only way I know to sample such high porosity sediment without disturbance is by a diver. Again, the authors need to add additional information.
, lines 21-22: Is the sediment porosity high enough that the sediment can be sucked out of the corer with a syringe? According to Table 3, sediment porosity is 99%! See previous comment.

Page 6, lines 12-13: This is a blanket statement that avoids doing a proper error analysis!
, lines 18-19: How long and at what temperature were the bottles equilibrated? The authors assume that all the methane will have been salted-out in the presence of the saturated NaCl solution. Is the reader expected to guess this? At the very least, the authors should refer the reader to a paper where the analytical principle is laid out.
, lines 27-28: Was the weight of residual sea-salts taken into consideration? Again, these data appear in the appendices, but the reader has to seek them out.
, lines 30-32: Was the TOC content normalized to the weight of the original sediment or to the decalcified sediment? Hopefully not the latter case since HCl will dissolve other phases than carbonates, including some of the OC.

Page 8, lines 1-3: Was the compression linear or did it vary with the porosity of the sample? What would be the impact of the latter on the age model based on the distribution of Mn and Mo? It would have helped the reader understand the choice of procedures if the authors had previously indicated that the sediment porosity was very high, as implied from Table 3 and indicated in the appendices.
, lines 19-22: Why not do XRF scanner analysis on the original core as is now often done. This would have avoided disturbance of the core (shrinkage) as a result of resin embedding? Again, it would have helped the reader understand the choice of procedures if the authors had previously indicated that the sediment porosity was very high, as implied from Table 3 and indicated in the appendices.
, lines 28-29: Why not from the gradient across the sediment-water interface itself? The concentration gradient between 0.25 and 2.5 cm below the SWI is likely very different than the across the SWI. On the other hand, the authors probably meant between the overlying waters and pore waters concentrations at 0.25 or 2.5 cm depth. This would make more sense, but is not what is written. Even so, depending on the thickness of the sediment oxic layer, the gradient between the OLW and 2.5 cm below the SWI may be very different than immediately across the SWI.

Page 10, line 11: Should probably be second order and include alkalinity. Although manganese oxide reduction generates alkalinity, much more alkalinity is produced through sulfate reduction. The authors should assess if the first process generates sufficient alkalinity to trigger Mn carbonate precipitation
, lines 22-24: What about the overlying waters, they must certainly serve as a boundary condition, as noted in Table 3.

Results –
Page 11, line 19: The authors should refer to pore water Mn instead of Mn2+ if they suspect the presence of Mn(III), as their analytical protocol does not allow this distinction.
, line 22: Same comment applies here as the authors' analytical protocol does not distinguish between dissolved Fe2+ and Fe(III).
, line 24: If so, this means that there is little authigenic CaCO3 precipitation and puts some constraints on the alkalinity (or [CO3]) that can be reached in the pore water. On the other hand, the Mn enrichments are always well correlated with Ca enrichments (Figure 4).

Discussion -
Page 14, line 14: Mn2+ is released to the pore water or the OLW, or both?
, line 22: How can an undersaturation with respect to Mn carbonate develop under anoxic conditions when alkalinity is constantly being generated?

Page 16, lines 6-7: Does this mean that the dissolved Mn pool reached a steady state in 1.5 years? If so the benthic flux must, at first, have been much higher than the values stated above.
, lines 8-11: See previous comment.
, lines 13-16: Unless the burial rate of Mn (in the form of carbonate or sulfide) has changed over time.
, line 17: These species (Mn2+), and particularly Mn(III), are not likely to be present as free ions in solution. Hence, it would be more appropriate to identify them as Mn(III) and Mn(II), respectively.
, lines 30-31: Probably because the pool of Mn is dispersed/diluted in the water column. If so, the authors should specify it, instead of having it implied.

Page 17, line 17: It could not occur in diffusive contact with the undersaturated overlying water column, but could occur a few cm below the SWI in a more consolidated sediment.
, lines 19-21: This conversion occurs stepwise and is relatively slow, probably on longer time scales than the formation of the Mn carbonates. The process is further limited by the diffusion of Mo from the overlying water column, although some might have been supplied by adsorption to Mn oxides settling to the sediment surface following the inflow of oxygenated waters.
, line 25: Unlike Mo, the Fe usually originates from the sedimentary solid phases directly or following Fe(III) reduction by OM.
, lines 26-28: This statement makes little sense as Mn(II) is unstable under oxic conditions. Perhaps the overlying water was oxic, but the environment in which the Mn carbonate precipitated must have been reducing to reduce the Mn(IV) oxides to Mn(II) and perhaps anoxic to provide the necessary alkalinity.

Page 18, line 4: in the Gotland Basin?
, lines 20-21: Could the MnS form at the expanse of/or replacing the Mn carbonates?
, lines 26-27: See previous ccomment.

Page 19, lines 10-12: How else than following inflows would Mn-rich particles be supplied to the sediment at Landsort Deep? The authors leave too much unsaid, if they have an explanation.
, line 30: What is responsible for these? From the previous text it would appear that the duration of the hypoxia and euxinia has increased between inflow events, but, as stated, it could also be interpreted as an increase in the strength of hypoxia (lower oxygen concentrations) or euxinia (higher sulfide concentrations. In the latter case, the authors should provide an explanation for this (e.g., increased inlfux of OM in response to eutrophication). (Again, the authors provide, later in the discussion, an explanation and, thus, keep the reader guessing, speculating, and questioning the authors’ interpretations until the end. This makes for a poor manuscript.)

Page 20, paragraph 1: The evidence and scenario presented by the authors in this paragraph should have appeared earlier in the discussion instead of letting the reader guess what mechanisms are responsible for the increased reducing capacity of the deep basin sediments (increased flux of OM due to eutrophication) and reduced shuttling of Fe to the deep basins (extension of hypoxic conditions to the shelves).
Page 20, line 4: The authors are skipping several steps and should add: ... the accumulation of Mn oxides at the sediment-water interface, and the re-establishment of reducing conditions, ...
, lines 10-11: Why? Increases sulfate reduction? (see comment above in parentheses)
, line 16: Why? Faster onset of hypoxia and euxinia rue to greater influx of OM? (see comment above in parentheses)
, lines 28-30: This argument could have been brought forward earlier in the discussion.

Page 21, lines 4-6: Again, this argument should have been raised earlier, instead of letting the reader guess.

Conclusions –
Page 22, line 28: It would be more logical to invert these statements as the loss of Mn2+ inhibits the formation of carbonates.

Other comments-
What of the role of AVS and their strong affinity for Mn(II)? The adsorption of Mn(II) to AVS can control the porewater Mn(II) concentrations and the conversion of AVS to pyrite could provide a deep source of Mn(II). (Morse and Luther (1999) mention that Mn(II) can also be pyritized?)
One would think that they (Mn and Mo enrichments) might be related since molybdate adsorbs strongly onto Mn oxides and could be scavenged from the water column as Mn oxides settle out, unless the adsorption kinetics is slow and significant concentrations can only be obtained through diagenetic recycling across the oxic-suboxic boundary.
How does the record of inflows over the past two decades compare to previous decades in terms of frequency and intensity?

Supersaturation of surface sediment pore waters following an inflow event is unlikely, if there is a finite oxic zone (as long as there is dissolved oxygen in the overlying waters since the production of metabolic CO2 will lower pH. What makes it supersaturated at depth? The accumulation of carbonate alkalinity from sulfate reduction?

Referee Report: PDF

Hide

RR by Anonymous Referee #7 (22 Jun 2015)

ED: Reconsider after major revisions (26 Jun 2015) by Katja Fennel

AR by Conny Lenz on behalf of the Authors (17 Jul 2015) Author's response Manuscript

ED: Publish as is (21 Jul 2015) by Katja Fennel

Download

The requested paper has a corresponding corrigendum published. Please read the corrigendum first before downloading the article.

Article (3748 KB)
Metadata XML