In general, the manuscript is well written, and the topic is highly interesting. However, prior to publication I have some questions and remarks that should be clarified prior to publication.  For more details, please see below. 

Methods section

Experimental set up -section needs some clarification. Currently it is not clear how the sediment was homogenised? How was it ensured? The process is also likely to have had an impact on the porosity and water content of the sediment and hence the sediment structure may be different to what it is at the field situation. Could this have implications for the results and how the situation is in nature?

What size of foraminifera were picked and introduced into the experiment, also >125 um as sieved at the end of the experiment?

It is also not clear how often and how many times in total the oxygen profiling was conducted. Based on the Figure 3 profiles were conducted during 13 days? Was every control core (n=6) and foram core (n=6) measured on these days as a triplicate? Or were some cores measured more than others? In the section 2.6.2 it is stated that each core was divided into 5 areas to avoid multiple measurements in a same spot, implying that each core was measured max 5 times. The frequency and number or measurements per core should be clarified. I would also be interested to know if the authors have considered the impact of O2 profiling on the oxygen penetration depth in the sediment. In the end a considerable number of profiles were made, each producing a small vertical burrow down to several mm into sediment.

Please clarify in section 2.7 how the sampling was conducted, and from which depth intervals samples were taken. I assume that somehow the core was sliced, or was sediment scooped straight from the core? If latter, how was mixing of the sediment avoided, and contamination? Please clarify the text. 

Results section

Section 3.2 did you measure the TOC and N-content at the start of the experiment? What was it at the start prior to incubations and how did it differ from the control cores at the end of the experiment? The freezing of the sediment used for the experiments would have introduced some fresh OM to the sediment, consisting of meiofauna etc. Foraminiferal are known the feed on various prey (e.g. Chronopoulou et al. 2019  DOI: 10.3389/fmicb.2019.01169), so wondering how much of the OM was processed by foraminifera, and if the difference in the Corg content at the end is due to foraminiferal grazing or enhanced OM-processing by bacteria due to bioirrigation?  

Section 3.3. It seems that the O2 penetration depth was relatively shallow at 22 days. Could some of the differences in the OPD-measurements be due to different numbers of O2 profiles made in the cores? Or do authors have an idea why ODP was shallow at 22 days?

Section 3.4. It is a shame that the authors did not take a molecular sample at the start of the incubation, to observed how to prokaryote community  developed from T=0 to T=85. If such a sample was taken, it would be very valuable to analyse it and include the results here. 

Discussion section 

Section 4.1 line 275. Without the Corg and N-measurements for the T=0, the authors cannot for certain confirm that that the Corg declined during the experiment.

Section 4.1 and elsewhere: I suggest that authors do not call control cores undisturbed but always as control cores, as the control cores were disturbed during the experiment with O2 microprofiling. It would be interesting, for authors to discuss the influence of profiling on the results. I.e. if some control cores were measured more than others, is there a difference in Corg, N, or OPD or O2 uptake etc?

Line 293, and 303: how were the burrows measured? Is it possible there were longer burrows inside the sediment cores that are not visible from observing the core from outside? And also that burrows were made inside the cores after 40 days. 

Line 332-334 (and elsewhere related to bacterial richness estimate from OTUs) Bacterial richness estimation is based on number of OTUs. This, however, is somewhat problematic as OTUs do not directly translate to number of taxa (as some OTUs can be from same species). The more recent concept of an amplicon sequence variant (or ASV) could be a more appropriate means of estimating number of taxa present/bacterial richness. Authors could consider looking into ASVs, or at least they should be careful when interpreting bacterial richness based on OTUs and explain the limitations. Especially here, as the Shannon index is not showing the same trends.  

Line 358-360. Regarding the influence of foraminifera on sediment N-content. It is more likely that the sedimentary N-content here is a reflection of OM degradation than related to foraminiferal denitrification. During degradation of OM, typically N-containing molecules (e.g. amino acids) are broken down preferentially, hence causing a shift in C/N ratio of OM left in the sediment (e.g. Schneider et al 2003, https://doi.org/10.1029/2002GB001871)  

-End of commenst

General comments:

First of all, please keep in mind that the following comments come from a place of deep respect and recognition of your work and the energy required to build, perform, and analyze the data of work like this one. The aim of these comments is to improve my understanding of the manuscript with the hope that more people will be able to understand the importance of the presented results.

The paper by Langlet et al. untitled Single-celled bioturbators: benthic foraminifera mediate oxygen penetration and prokaryotic diversity in intertidal sediment is a valuable piece of work documenting for the first time a single-celled ecosystem engineer. This research highlights the importance of foraminifera in the functioning of ecosystems through their influence on sediment irrigation.

The MS is, overall, well written and understandable. However, since English is not my first language, I did not pay attention to the spelling and grammar.

I think the method need some work to clarify some points detailed in my specific comments. I have some issues with the models build for statistical analyses and the consideration of temporal and spatial dependency between sample. Also, no details are given on the precisions and limit of quantification of the analyses. This should be assessing each time especially in this kind of study where differences between treatment are very small. Without this information one could not tell if the differences detected by the author fall within the error of the analytical method. One of my major concerns is the shift in oxygen dynamics over the experimental period. The authors did not give explanation except for the reference of Bonaglia et al. 2020 where sediment characteristics were not comparable (very high OM content). Finally, I think the discussion is well written but not clearly sustained by the results given the information at my disposal. To conclude, I think the manuscript need major revisions to be considered for publication in biogeosciences.

Introduction

L36-37: sensu Kristensen et al. 2012 bioturbation is an umbrella term encompassing sediment reworking and burrow ventilation which causes bioirrigation.

L50-51: you already mentioned that meiofauna increase oxygen availability (L41). This sentence oversimplifies bioturbation process. Yes, it overall increase oxygen availability but also increase sediment heterogeneity by creating microenvironment and this may be the main cause of more diverse bacterial communities in bioturbated sediment

Material and methods

L82: Is 14 days enough time to reach a steady state?

L88-91: Did you assess OM content at the beginning of the experiment. Did you assume the OM content was similar between the sediment core at the beginning?  That is probably true but worth mentioning.

L96: Why 125 µm? it could be great to mention your definition of meiofauna and justify this mesh size in your context.

L111-112: What is your rational for the decreasing of microporosity over time? the sediment core did not reach a steady state after 14 days of acclimation? Why perform microporosity only on sediment without forams? In the introduction you mentioned the impact of meiofauna bioturbation on particle mixing. Do you assume that the particle mixing by meiofauna is negligible?

L116: How many samplings time during your experiment? 5? When?

L115-119: This is hard to follow even with the help of Fig. 1. The microprofiles were performed in the same 2 cores over the experiment? Above you mentioned n = 6.

L177: there is a dependence of data between two experiment time. Therefore, this factor should be random to account for this time dependency.

L178-180: Please expand on this shift in oxygen condition. Why does it change? Is this a result from a contamination of the control cores? An experimental artefact? This is a major concern since most of your data were collected after this shift.

L187-193: Do these bacterial orders perform only sulfate-reducing processes? Same question for the methanogens. Also, I have no data on the sediment characteristics, but do you think sulfate-reducing processes and methanogenesis occur in the first cm of the sediment column? With the OM concentration given below it would be very surprising. So why a focus on these processes?

L194-196: Did you account for the spatial dependency between sample collected at 0-5mm and 5-10 mm in the same core?

L198-199: Did the assumption met after log-transformation of the data? Did you try using glm instead of lm models?

Results

L210-213: Ok this is significant but does a difference of 0.2 % is significant for forams? Also, what are the (i) precision, (ii) limit of detection and (iii) limit of quantification of your analysis?

L213-215: Idem

L224-225: Again, I am really surprised by this shift in oxygen dynamic during the experiment. With the information at my disposal, I have major concern about the good execution of the experiment. If there is a natural shift in oxygen dynamic, you could use segmented analysis or multi change point analysis to correctly model the relationship and assess the time at which the change occurs.

L226-229: Please consider changing the unit of the DOUs to avoid using decimals. This is, in my opinion, hard to read.

L234-235: the correlation test is not described in the methods section.

L239: sensu stricto you did not test for an increase of Bacteroidetes with depth, you compared two groups.

L242: Ok, there is a significant difference but when we look at the figure, we can see that one sample with forams is in the range of the samples without forams. Do you think here you can detect a significant difference because of heterogeneity of variance? BTW, could you mention which data sets were transformed?

L249-251: I am not sure to understand what you mean here.

Discussion

L281-283: Here you refer to the .2 % (TOC) and .05 % (TN) differences? From Fig. 2 the variability in foram treatment within the 0-5mm layer is .1 % vs .2 % within the 5-10mm layer (TOC). In my opinion this is a too small difference to sustain your rational.

L290-291: Without assessing OM at the beginning of the experiment you cannot know if there was a decrease of OM content. Furthermore, in the study of Bonaglia et al., 2020, the TOC was 20 times higher than in your work which can explain the presence of H2S in the first cm of the sediment column.

L332-334: OM account for 46 % of the variability in you model. Could predatory pressure account for the remaining 54 %? Did you test it?

L340: by diversity of OM you mean diversity in the quality of OM? How a diversity could be limiting?

L370: Do you mean “reduced” instead of “oxygenated”?

L376: Did you measure iron and iron oxide? it could be linked to the presence of iron oxydes and the formation of AVS. Forams could also enhance the iron cycling of course.

L386-389: consider reformulate these sentences. We knew forams were bioturbators mixing sediment particles but now, with your work we know they also alter porewater distribution. I do not think organism should perform sediment reworking AND burrow ventilation to be considered as bioturbators. Some bioturbators do not burrow but crawl at the sediment surface.

L391: you should add a reference for the definition of EES.