Re: Review of se-2022-11

Hi Virginia,

Please find below my review of the manuscript by Lahiri et al entitled “Mechanical compaction mechanisms in the input sediments of the Sumatra Subduction Complex- insights from microstructural analysis of cores from IODP Expedition- 362”. This is a well written manuscript on a relevant and interesting topic for the journal. It does need some fairly major revisions to bring it up to publication standard, but nothing that should not be relatively easy for the authors to address. I have one main concern which I feel should be addressed in detail and then I’ll go through some questions and comments I have regarding the rest of the manuscript in the order I found them. I have also supplied a marked-up copy of the manuscript for edits, typos and referencing issues which is a bit messy, so I apologise for that.

1. My main concern with this manuscript is that the description of the clay mineralogy of the sediments is somewhat confusing and this requires significant clarification by the authors. The paragraph on page 4, line 113-121 notes a clay mineral assemblage of “illite with minor smectite and chlorite”, noting 33% smectite in unit 1 and 73% in unit III (Table 1), numbers which are not really minor (and are averages really useful in such thick units?). The illite percentages are then given followed by a discussion of illite-smectite (I-S) expandability. However, I-S is not documented at all in table 1, just presumably pure illite and smectite. How much mixed-layer clay is present, and is it being properly differentiated from the pure end members? What is the I-S composition? This is actually important for the whole manuscript since porosity corrections are required for smectite-rich sediments and this may impact significantly on results, especially as there seems to be 15-20% “missing porosity” as noted in the discussion between MAD and SEM porosity (although there are other possibilities for this). This lack of clarity over clay mineralogy also comes back in the discussion (p12, L385-388), where the numbers quoted in the text are not consistent with table 1. For me, I feel it would be really informative to have the clay compositions plotted up against depth and added to figure 2 for comparison. I’m sure IODP have the data and although the measurements wouldn’t be from exactly the same points as measured here, they would still provide clear trend information.
2. P2, L52-54, depositional environment is also a strong control on porosity evolution, compaction and diagenesis in mudrocks (e.g. Baruch et al., 2015, AAPG; Delle Piane et al 2015 MPG) as the initial clay and rigid grain compositions significantly affect both compaction (as this manuscript shows) and subsequent diagenetic alteration due to variations in composition.
3. Somewhere in the geological background and drilling section, there should probably be a mention of whether these sediments have been uplifted at all or are currently at maximum burial depth. I would assume the latter just from how things have been written but probably best to have a clear statement.
4. L137-147, somewhere in this paragraph there needs to be a description of how the samples were freeze dried as this is absolutely critical in very soft clay bearing sediments when trying to get quantitative data from pore characterisation techniques, whether that be via SEM or gas/mercury porosimetry. Need to know things like the method employed, temperatures achieved, how long they were frozen for before drying, how long the drying took etc etc. This goes to understanding the quality of results and I’m a little surprised there is no discussion of this. There was work done on this going back to the early 1980s looking at fabrics and microstructure of soft clays and the impact drying can have on porosity measurements and particle preferred orientation. There should be some consideration of this work and later work for Example in the Microstructure of Fine-Grained Sediments compilation quoted in the Bennett et al reference in the manuscript. Given the “missing porosity” between MAD and SEM quoted later in the paper, could the drying methods account for that? This is a really important topic where the authors need to convince the reader that the data quality is good. A few of the early references may prove helpful and these (plus younger ones that could be found through citations) could help formulate the discussion further. I’ll come back to this point below.

Delage P., Lefebvre G. (1984): Study of the structure of a sensitive Champlain clay and its evolution during consolidation. Canadian Geotechnical Journal, 21, 21-35.

Delage P., Tessier D., Marcel-Audigier M. (1982): Use of the Cryoscan apparatus for observation of freeze-fractured planes of a sensitive Quebec clay in scanning electron microscopy. Can. Geotech. J., 19, 111-114.

Griffiths F.J., Joshi R.C. (1989): Change in pore size distribution due to consolidation of clays. Geotechnique, 39, 159-167.

Griffiths, F.J., Joshi R.C. (1990): Clay fabric response to consolidation. Applied Clay Science, 5, 37-66.

Griffiths, F.J., and R.C. Joshi, (1991), Change in pore size distribution due to secondary consolidation of clays, Geotechnique, 40, 303-309.

5. P5, L169, The PPR and REA section needs some discussions around the assumptions, errors and limitations of determining porosity and pore sizes from 2D images. Pores are non-spherical 3D objects that are cut through in random orientations and at random points in a 2D thin section, so measurements taken on the images have associated assumptions and errors.
6. P6, L195, introduced the shipboard MAD measurements in the results section. Again, even though done elsewhere, a brief description of the methodology is warranted in the section above as to how this was performed and then point the reader to the appropriate reference for further details. These data are absolutely critical to the manuscript, so the details are sorely needed. In addition, Figure 2 shows some data in yellow points as measured at Aachen and BEG; were these done by the same method as shipboard or different? Again, a description is warranted.
7. L238-239, the correlation coefficient between the two porosity measurements should be shown either in the text or on Figure 3b.
8. P11, L341, in these plots it should be stated which points are used to calculate the D values. Is it all points or just the points that avoid the artificial tail offs from sampling bias/resolution issues at either end of the lines? Probably should be the latter but some of the trend lines do not lie through the straight-line section of the points.
9. For me, the whole discussion is a bit weak and lacking in detail. There are some interesting points made and comparisons, but they are not really built on very much and tend to be almost standalone points based on a couple of sentences. There also seems to be potential background discussion items that probably should be in there but are currently not referred to. For example, the soil mechanics community has done a lot of work around natural clays and remoulded clays (similar to the Mondol/Fawad type work) but this is not even discussed although thoroughly relevant to the behaviour noted here (e.g. classic papers by Burland, 1990, Geotechnique; I think also Lupini et al 1991, Geotechnique, which is currently quoted in the reference section but isn’t in the text). The references I note above in point 4, although old, also discuss topics alluded to here in terms of pore collapse with compaction, the very topic of the paper. Perhaps a bit more background in the introductory sections would provide a good lead into the discussion of the results obtained. Supplement 15 should also be in the discussion as it provides some details that are sorely missing at the moment.
10. P12, L400-405, it would be better to show the supplementary data 14 in the text rather than outside the manuscript. This is actually a pretty important result and consistent with the topic of the paper.
11. P13, L415-417, this is a good point here about the smectite content potentially resulting in over-estimated MAD porosity and one of the things I was getting at above in point 1. However, until there is some better clarity about the clay mineral composition, it is a moot point. I suspect this is a contributor for sure but again this needs a more detailed discussion as to why this is the case and appropriate supporting references quoted.
12. P15, L506, the authors discuss “intra-crystalline” slip as a mechanism for particle re-arrangements during compaction. This term has a very specific meaning, usually around things like dislocations and the like within crystals, calcite being a classic example at relatively low temperatures and pressures. One thing the authors haven’t really discussed is the flocculation and aggregation of clay minerals (although flocculation first makes an appearance in the conclusions section on P17, L545 and aggregate in the discussion). The results of these processes can be clearly seen in the SEM images the authors show, often with multiple clay platelets stuck together in aggregations, but this is not described in the images clearly. The intra-crystalline slip quoted here appears to be more like intra-aggregate slip of particles or inter-platelet slip and these would be better ways to describe these phenomena. Again, there is a lot of background work done on clay fabrics in compacting clays with well established terminology which isn’t referenced here, I think some in the references I have noted above for the early work and plenty in the Microstructure of Fine-Grained Sediments compilation as starting points. The manuscript would really benefit from a brief introduction to typical microstructural fabrics of clays which then naturally leads into the discussion.
13. As an overall comment, there is a lot of good data in the supplementary data that would really back up the observations made in the text. I have already mentioned 14 above, but 11 and 12 to me are also no brainers for supporting the hypothesis put forward by the authors. To be honest, I don’t think tables 2 and 3 bring anything to the paper, they are not things that can be quantified, and these for sure should be supplementary data as figure 8 covers their input adequately. If one was going to include a table of useful data in the paper, I would pick supplement 7, perhaps a slimmed down version with some columns removed and the full table still provided as a supplement. This would provide much more useful and quantifiable data to the reader.
14. Following on with the discussion on supplementary data, the large pore outlined in the image of the deepest sample in supplement 10 looks like a plucked grain to me. One can see the compacted clays around the edge of the hole outline and within the hole. I would use a better image than this if you have one. The other two images in this supplement for the shallower samples are clearly geologically realistic and relate to the interactions between rigid grains.
15. For figure 1b, the abbreviation HANP needs spelling out or removing if not relevant to the study.
16. In figure 2, b and c are the wrong way round. This plot would benefit from clay mineralogy data (e.g. illite vs depth, smectite vs depth, clay content vs depth, I-S vs depth etc) on a single plot (or added to the porosity plots).
17. Figure 3 would benefit from having a 1:1 line for easy reference for the reader. Also, the Boom Clay symbol colour on the graph doesn’t match the legend.
18. The references are a real mess, many not in the text (all the ones with red ticks on the marked manuscript were ones I could find, including in all the supplementary data), many dates are inconsistent between text and reference list (marked in manuscript) and many in the text but not in the reference list. To be honest, this is much more than a reviewer should have to deal with and tidy up!
19. There are a bunch of typos and suggested edits also in the marked-up manuscript supplied.

This is a really interesting manuscript with some fabulous SEM images of compacting clays and great insights into processes occurring during early mechanical compaction. At the moment, there are a few things around experimental techniques, clarity of results, especially clay mineralogy, as well as more background information which would significantly improve the manuscript and strengthen the discussion. I would be happy to review this again should that be required, especially if the references require less work next time 😊.

Best wishes

Dave Dewhurst

CSIRO Energy

Perth

Australia

Dear Virginia,

Please find below my review of the manuscript “Mechanical compaction mechanisms in the input sediments of the Sumatra Subduction Complex- insights from microstructural analysis of cores from IODP Expedition-362” submitted by Sivaji Lahiri, Kitty L. Milliken, Peter Vrolijk, Guillaume Desbois, Janos L. Urai. The manuscript is well written and addresses a topic relevant for Solid Earth. I suggest minor revisions and address my main points below. I also provide an annotated version of the manuscript highlighting typos, etc. Please also refer to this file to check on the references – they are a mess and many reference have either not been cited (properly) in the text or provided in the reference list.

Kind regards,

Bernhard

• Although quite prominently presented in the abstract (lines 17 – 21), the change of deformation mechanism from rotation and realignment of clay-platelets dominating in the shallower section (0 – 28 mbfs) to bending and subsequent sliding / fracturing of clay particles dominating in the deeper sections is not that evident from the observations presented. Maybe a figure presenting evidence for both mechanisms next to each other would support the statement. In addition, the manuscripts describes the change of deformation mechanism observed at 28mbsf but does not elaborate on the underlying reason. Maybe the authors could discuss this aspect.
• In addition to information on clay mineralogy (lines 113 – 121), additional information on bulk mineralogy should be presented, too. What is the bulk composition (not only the clay mineral assemblage)? In this context I have to admit that the information which should be transported by Supplementary data-1 is not clear to me – especially at the position where it is referred to (line 114).
• The fact that site U1480 consists of several holes, whose samples have been used for this study, should already be mentioned in line 94 instead of line 131.
• Section „sampling and methods“: based on the information given, I would assume that sample preparation and analysis in Aachen and Austin was done by Ar-ion cross-section polishing succeeded by SEM investigations (i.e. BIB-SEM) with the main difference between both universities having been the instruments used (BIB polishing device: JEOL vs. LEICA; SEM: Zeiss vs. FEI). Is this correct?
• MAD measurements already performed should be mentioned earlier than in the “results” section.
• Has the mineralogy presented in lines 227 – 231 been determined based on EDX element map / point analysis or is it the outcome of XRD bulk measurements (as presented in lines 113 – 121)? Qtz, Fsp, Cc, Mica und Ill are only five phases – what is the sixth phase mentioned in line 227? How was the detrital origin of the clay-size particles determined (l. 228)? Assuming that information in this paragraph are based on EDX analyses (i.e. no diffraction patterns), I wonder how it was possible to conclude that clay-size particles are dominantly illite, given that XRD analyses reveal the presence of smectite and illite/smectite.
• Given that segmentation of SE2 images only gives pores without additional information on the kind of pore, I wonder how the amount of intergranular pores (>99%) could have been determined and if this value can be considered to be representative or just a qualitative assessment (lines 241 – 242).
• Exchange (1) and (2) in line 254 to be consistent with Fig. 4 and lines 251/252 and 255, respectively.
• Contacts “EE”, “EF” and “FF” should be highlighted in Figures 5, 7 and Supplement 13.
• To be consistent with the information given in lines 281 – 283, a similar information should be given at some point in lines 259 – 267.
• Unfortunately, I was not able to extract the bulk density of the sediment from the reference McNeill et al. (2017) to calculate the vertical effective stress (cf. line 373). Maybe the authors could provide these values in the manuscript / the supplement?
• The statement given in lines 416/417 that “shallow depth samples are richer in smectite compared to the deeper samples” should be somewaht adjusted, because Table 1 shows that it is only true with respect to the difference between Unit I and Unit II, not with respect to Unit III.

Figure 1

• Please ensure that spelling of Ninetyeast ridge on the figure and in the text is identical
• (a) is not a geological map but a satellite image
• Please specify in the caption that (b) shows one of the holes (hole G) of site U1480.
• Please provide orientation and scale of (b) and indicate its extend/location in (a)
• Please explain the meaning oft he blue and red lines, respecively, as well as the meaning of „HANP“
• Please provide a scale for (d)

Figure 2

• Please adjust order of figures (a – b – c – d instead of a – c – b – d)
• Please add common title “U1480” and “U1481” for Figures a/b and c/d, respectively
• It appears that porosities at around 1300mbsf tend to increase – is this correct?

Figure 3b

• The straight line indicating the linear relationship should be plotted, too. In addition R2 should also be indicaed in the figure.

Figure 4

• Please ensure a consistent spelling of terms introduced in Figure 4 (e.g. silt-adjacent vs silt adjacent, etc.) in the text.

Figure 7

• To me it is not clear which additional information is provided by this figure compared to Figure 6. If you want to keep it, please adjust contrast and brightness, because the image quality is pretty poor compared to Figures 5 and 6. Irrespective of image enhancement, I suggest to move it either to the supplement or to completely skip it.

Figure 10

• Caption should state that porosity is MAD porosity.

Figure 11

• Please rearrange Figures a – f according to depth. This would help to support the statement of line 439.

Figure 12

• Caption states that there are white and black arrows. However, I cannot see black arrows, so I guess the caption refers to an earlier version of the figure.
• Figure g: while porosity reduction is evident in the four stages presented, I cannot see the described increase in preferred alignment of the long axes of pores, i.e. the pores’ long axes appear to already have a preferred alignment in stage one which does not change in succeeding stages. The figure should be modified to emphasize this more clearly.

Table 1: The table should not only present the clay mineralogy but the bulk mineralogy, too.

Table 2: I guess that this table is based on quantitative pore investigations. If so, I suggest replacing the qualitative terminology (i.e. “abundant”, “rare”, etc.) with the actual values, because an evolution with depth cannot be seen at the moment. In addition, one might consider to shift the table to the supplement.

Table 3: Basically the same as for Table 2. In addition, I suggest to add the meaning of “EE”, ”EF” and “FF” to the caption.

Supplement 2: More details (i.e. a legend) on the kind of information provided in column G („type“) should be provided.

Supplement 8: SN6: I would expect that in this kind of sample pyrite is mostly present as framboidal pyrite. Hence, I am somewhat surprised to see so many large pyrite grains in this figure. However, after comparing this figure with Fig. 5e, grains labelled pyrite appear to be micas. This should be verified.

Supplement 9: SN8 does not seem to correspond to Fig. 6a. SN33: mica is labelled in green instead of red. In addition, please verify that the large grain in the lower right corner is indeed mica.

Supplement 11: This figure should be part of the main manuscript, not of the supplement. Images (a) and (b) are a bit fuzzy, maybe some image processing could enhance that. Please modify the following captions: (b) “pore” instead of “pores”, (e) “pore” (as the label on the figure) instead of “fracture”.

Supplement 12: To better validate the data presented, the number (N) of segmented pores and grains, respectively, should be indicated, too.

Supplement 13: I suggest including supplement 13 in the manuscript and placing it before Figure 5.

Supplement 15: This should be included in the discussion of the manuscript.