General comments:

In this paper, the authors carried out a thorough analysis of porewater and solid phase chemistry throughout a core of marine sediments, with a focus on patterns of Fe isotope composition. The goal was to use these patterns to explain patterns of Fe biogeochemistry throughout the depth of the sediments. The thoroughness of the differential extractions was really impressive. They were operationally defined (e.g. roughly corresponding to adsorbed Fe(II), poorly crystalline Fe(III), crystalline Fe(III)) but those operational definitions are still meaningful from the standpoint of Fe biogeochemistry. The authors found that patterns of Fe isotope compositions of Fe(II) and Fe(III) phases were inconsistent with those observed to result from dissimilatory/respiratory Fe(III) reduction in lab experiments. The authors indicate that adsorption/atom exchange does not contribute to the observed isotope patterns even though they also indicate that the rates of Fe(III) reduction are likely quite low.

I am struggling with the fermenter conclusion a little bit, because it seems like the authors are saying “patterns of Fe isotopes in all of these different pools are inconsistent with dissimilatory/respiratory Fe(III) reduction, so it must be from the fermenters dumping electrons.” Additionally, conduction of electrons from fermenters to methanogens would not result in a net reduction of the Fe (the fermenter would reduce it, but the methanogen would oxidize it). Granted, it is likely that an initial reduction would have to occur, because most hypotheses for interspecies electron transfer via Fe involve magnetite or pyrite, but after that initial reduction, no net redox change would occur with the Fe.

I am also struggling to see how the authors incorporated advection or diffusion of dissolved Fe(II) into their models and interpretation. Depending on the rates of Fe(III) reduction, those would be a major controller of the extents of atom exchange (i.e., is Fe(II) exported quickly enough that no atom/electron exchange can occur?).

Despite these criticisms, I think the work is important because it contributes to what we know about patterns of Fe isotope compositions in different Fe pools – It’s just hard to explain at this point. I think the authors have identified the major controller here: kinetics of Fe(III) reduction vs. kinetics of abiotic processes. I enjoyed reading this paper.

Specific comments:

Ln. 57. The authors do not include Fe(III) reduction by methanogens in their interpretation. The enzymology of that process is similarly understudied to that of fermenters.

ln. 345-351. avoid three sentence paragraphs. Also having trouble seeing how these observations are fitting into the broader story.

Ln. 620. this isn't completely true (benefitting both microbes). Unless the methanogen can use the reduced/conductive Fe phase as an electron donor, the Fe(III) just gets reduced and that's the end of it. This scenario is #1 on ln. 604.

Ln. 622-624. I think Nathan Yee’s group has done some work to address how fermenters reduce Fe(III).

Ln. 659-660. I agree with this, and think it’s the strength of the paper.

Technical corrections:

Ln. 19. Please remove “unsurprisingly”

Ln. 386. Please change “conclusive” to “consistent”

Ln. 436. Please change “wit” to “with”

General Comments:

Overall, this is a very interesting study, presenting a comprehensive dataset and a scientific approach that combines field data with modeling. The illustrations are clear, the manuscript is well-written, and the interpretations, while complex and occasionally probably speculative, remain cautious. This is well-explained throughout the manuscript.

This work contributes to advancing our understanding of the iron cycle in methanic sediments. For all these reasons, I recommend the publication of this work.

I have two major remarks, in addition to several minor points listed in the PDF as comments:

•  The authors appear to assume, given the chemical conditions, that dissolved iron is always reduced iron (Fe-II). However, what is referred to as dissolved iron is, in fact, filtered iron (0.1 µm). Was the redox speciation of iron measured in all the pore waters? In the oceanic water column, 'dissolved' iron encompasses a variety of physicochemical species, including small particles (<0.1 µm containing Fe-III) or colloidal complexes (which can also contain Fe-III). The process of 'non-reductive dissolution' of iron – which does not require reductive conditions as it can involve desorption or ligand-promoted dissolution – seems to be an important process for the release of 'dissolved' - in fact filtered - iron at the sediment-seawater interface (Radic et al. 2011, Labatut et al. 2014, Homoky et al. 2021). Could this 'non-reductive dissolution' process play a role in these methanic sediments?
•     The validation of the isotopic measurements appears too superficial. Potential artifacts (isotope fractionation, contamination) related to chemistry, preconcentration, purification, and partial dissolution seem not to have been thoroughly investigated. Yields and blanks from the various steps are not consistently reported. Repeatability seems not to have been quantified across the entire protocol (including the chemistry), and the error bars reported in the graphs seem too small. For instance, the repeatability of the instrument for pore waters appears to be 0.26 ‰, which in my view implies that no measurement of pore waters can have an error bar smaller than 0.26 ‰.

see the pdf for other comments

Very impressive and very nice work !