This paper is within the scope of the journal and I suggest publication after a minor revision addressing the comments in the attached PDF file.

The authors propose a simulation of the Cr and Cr isotopes cycles at a global oceanic scale using the Bern3D EMIC. Despite a limited data set, they propose a well argued discussion of the sources, transformation and sink of this oceanic tracer including its redox forms (CrIII and CrVI). Although the limited set of data is narrowing the interpretation fields, leading to somehow hypothetic discussions (eg on the OMZ CrIII behavior and the Arctic specific case), the consistency between the modeled and observed data give cofidence about the scientific approach.  The manuscript is well written and illustrated. This work deserves publication in Biogeosciences with minor revisions as listed below.

• Riverine inputs : the authors adopt a simplified parametrization to the CrVI and CrIII removal rates in the estuaries. Although the respective values of 20% and 80% are consistent with the known behavior of these tracers, the reader is missing argument justifying this choice. In addition, the sensitivity of the model to this input term was not tested, it would have been interesting to explain why.
• The same manner, what are the arguments justifying the increase of the OMZ reduction rates to 20 and then 30 nmol/m3/y ? The way it’s written looks like the authors played to the lotery which I don’t think it’s the case !
• The observation data base : although provided in the annex, a map representing the locations of the available (and selected) data is missing. I strongly recommend to propose a figures compiling all the stations used in this work (if I am not wrong, only few Topogulf stations are represented while the whole set of published data is listed in the excel file in annex)
• Finally, I was slightly disappointed to read that the discussion on the Med Sea results was skipped since I analyzed 3 profiles, quite homogeneous with depth and that could be compared to the North Atlantic ones : this semi-enclosed sea is a good place to test the relative influence of dust vs sediment sources.

see attached

Review of “Modeling the marine chromium cycle: New constraints on global scale

Processes” by Pöppelmeier, Janssen, Jaccard, and Stocker (bg2021-0106)

            The paper serves a useful purpose as a first-stab model for Cr in the global ocean. The model itself is well documented, and for many purposes, of sufficient complexity that many processes can be included. It makes estimates of things that are poorly known (most notably, the benthic flux of Cr), which gives observationalists targets to aim at. And if anyone disagrees with some of the model assumptions, then they are free to make their own model. Hence I recommend publication of a revised version of the manuscript.

            That said, there are many things in here that I wouldn’t have done if I were the one constructing the Cr assumptions. My major unhappiness is in the assumption of a (relatively large) globally uniform seafloor flux for Cr, although I would also have handled the OMZ assumptions differently. I think the manuscript bears some confusion on the ocean residence time of Cr. If the benthic flux of Cr comes from regenerated biogenic matter that removed it from the upper ocean, it doesn't shorten the residence time in the entire ocean-surface sediment system. I think that the river flux is the main constraint on that number, unless Cr release from aluminosilicates in sediments is significant. And there is no data upon which to know whether this happens.

            As the authors note, the model’s handling of oxygen deficient zones (ODZs, which should be distinguished from oxygen minimum zones (OMZs) is inadequate to represent them very well – even for simpler properties such as oxygen. Is there a companion paper on the nitrogen system in ODZs? I bet that it is similarly problematical. Probably someone needs to make a regional model that can do a decent job at representing ODZs before trying to include them in a global model.

            The paper misses out on some significant references:

Lines 110-120: Shiller (1991) GCA 55:3241

Lines 120-129: Brumsack (1983) Mar.Chem. 14:89 and Shaw (1990) GCA 54:1233

Lines 180-185: Elderfield (1970) EPSL 9:10 and Shiller (1987) GCA 51:3273

Line 275: Sherrell (1988) DSR 35:1319

And I would also add for lines 285-290, Arctic surface Cr is influenced by Fe(II) oxidation by reduced Fe released from organic-rich Arctic shelf sediments, with Cr(III) formation and Fe oxide scavenging.