General Comments

The authors present a study on the identification of soil properties which explain the differences in carbon mineralisation following a land use conversion. For this they apply classical experimental methods, the very promising X-ray CT method and statistical analyses. I see sound scientific work all over. Further, the manuscript is well-structured and written in a concise style. I made some suggestions with respect to the written English, however I am also not a native speaker.

Nevertheless, I identified two issues. I guess both could easily be resolved. I am not convinced that the PCA provides any new or fundamental findings that could not be determined via the extensive regression analysis presented in this study. Thus, i suggest to remove the PCA. The second issue is related to the conclusions section. It contains hardly any conclusion and is rather a summary. This should be entirely rewritten (see spec. comments).

Given this, I suggest major revisions.

Specific Comments

title: What about 'Land use impact on carbon mieralization is rather caused by variations of particulate organic matter than of soil structure'? Just a little shorter...

17 replace 'farming' with 'agriculture'

27 'to' is missing after 'due'

29 please add 'the topsoils of' before 'croplands'

35 well, the fact that microbial biomass is correlated with substrate-induced respiration is not much of a surprise ...

60 replace 'moisture' with 'water'

62 I guess you should be more precise about the 'initial flush' as a consequence of soil disturbance. Is this initial flush hours, days or weeks? And in contrast Herbst et al. 2016 actually detected signifiant differences in basal C mineralization between disturbed and undisturbed samples incubated at the same water content for five days following a conditioning period of seven days after disturbance.

85 better write '.. to assess how these microstructural'

99 better write 'Each land use treatment is replicated five times...' cause you do not make use of the climate treatments in this study. As written now this would rather confuse the reader.

178 eq. 1 In the end this is a matter of taste, but I would suggest to write the formula as 'p_si(t)=A+B*exp(ut)'  and replace the 'e' with 'exp' cause otherwise the reader might search for the variable 'Be'

194 The statistical analysis procedure is sound. I would just point to the fact that the error probability level of 0.05 is an absolutely arbitrary choice. You are doing a lot of sophisticated statistical analysis based on an arbitrary choice. I know it has been done for ages like this, which could also serve you as a justification ... Just be aware.

203 I would refer the reader to table 1 at this point, otherwise all the symbols might be a little overwhelming

217 well, 'manually selected' actually means something like 'variables selected based on expert knowledge', which may be a more precise statement

227 'no difference in average bulk density' don't you think this is related to the time since conversion? I would assume that it will actually take a long time. May be you find some literature data for the same type of land use conversion.

235 table 1 Units! There are some flaws in the units, please check. POM is probably mg g^-1 (not g^-3), there is a '1' too much for the basal respiration unit and the '-1' for TMP and AMB should be upper case.

254 remove one '.' at the end of the sentence

262 Well, I would not refer to a R2 of 0.53 as a 'strong' correlation. Just a little more than half of the variation is explained ... I suggest to moderate this statement. And strictly spoken: A coefficient of determination is not a correlation coefficient.

264 'independent of initial water saturation' well, there is actually not too much variation in the initial effective saturation. And the differences that may be resulting from differences in water stauration are probably counteracted by POM content: lower saturation but higher POM...

271 Generally, I do not think that SIR is very reperesentative of what is happening under real-world conditions. The C source is totally liquid-phase and soil is almost entirely water saturated, which rarely occurs under field conditions.

276-278 so, this is all a measurement error?

286-287 I think that p_0 and the t_exc are very redundant. A higher SIR will inevitably lead to an earlier excess of respirometer uptake capacity. Basically, the R2 should be 1...

296 For the C:N ratio this might be explained by the fact that the C:N ratio of the biomass of crops and grassland is rather similar?!

312 I guess this should be Fig. 6a and not Fig. 7a

320 I guess this should be Fig. 6b and not Fig. 7b

325 i do not think that the PCA contributes anything new or viable. I suggest to remove the PCA. 

337 I guess this should be Fig. 7a and not Fig. 8a

347 I guess this should be Fig. 7b and not Fig. 8b

395 I do not think that those statments can really be validated with an R2 of 0.39. I strongly suggest to moderate the statements. 

405 replace 'overlooked' with 'additional'

436-450 The conclusions section is rather a summary than any real conclusion. Basically, only lines 447-450 contain some conclusions and even those are rather common statements. I suggest to entirely rewrite the conclusions sections, simply refer to the objetives stated at the end of the introduction.  

This is an original approach to study drivers of decomposition and microbial respiration in soils, in particular soil microstructure. The work is based on clear objectives and a strong sampling design and overall clearly written. I therefore believe this work is worthwhile publishing. The authors find that soil micropore structure as determined with XR Tomography does not explain variation in microbial respiration or growth among different land use types, at least not in well aerated soil (NW European spring conditions). They also conclude that POM (particulate organic matter) mainly causes the differences in carbon mineralization among soil types (see title of the paper). My general concerns about this paper mainly relate to this latter conclusion, and in my opinion those concerns need to be addressed through major revision in order to make sure that the results are correctly interpreted and the limitations of the study are understood. Those concerns include:

• The interpretation of results based on correlations as proof of causal relationships (see for example the title).
• The authors make quite strong conclusions on the role of POM in explaining variation in respiration. However it is clear that the methodology used for POM fractionation did not separate root biomass from POM. So it should be very clear that all data and conclusions on the role of POM , in fact correspond with POM + ROOT biomass. This is an artefact of the method because these roots died as a result of sampling. Important to acknowledge this when interpreting your results and the possible implications of this for the interpretation of the results need to be explained clearly.
• Important aspect for interpretation of the results is that “in none of the soil cores the initial water saturation was high enough to induce a deficiency of soil aeration”. This is a very important observation that requires attention when drawing general conclusions on the importance of structure for respiration (again, see title). The implications of this for the generalization of the results need to be discussed. At the moment this is only brought up at the end of the conclusions section.
• The authors should discuss the implications of the fact that the XRTomography has its limitations in terms of resolution (minimum pore size) and also discuss in a bit more depth the potential drivers that could explain mineralization but that were not measured (OM quality, nutrient availability, total SOM, microbial community composition, efficiency). Total SOC data seem to be available but this parameter not included in the regressions as a bulk soil property? Why not?

Other comments

Title:

The word "caused" is too strong here  I believe, was a causal relation really tested? The word "explained" would reflect better the relationships obtained. Also this statement cannot be generalized beyond well-aerated soils. I suggest to change to: “Variation in carbon mineralization across land use types in well aerated soils is mainly explained by particulate organic matter content rather than soil microstructure”.

Abstract:

Ln 15-16: Please be more specific: This paper is about respiration and microbial activity. Not about the other functions

The organization of the abstract can be improved from Ln 22 onwards. First, present the main results that were obtained, and then interpretation, followed by clear conclusions. At the moment results, assumptions and interpretations are all mixed and it is not clear what was actually measured and what was assumed to explain these results. Also be careful with the word “cause” where causal relationships were not established.

Ln 24: The word “absent” is too strong here. The effect of moisture cannot be excluded but in the current study these effects may be dominated by other factors such as total SOC? whether this effect is absent should be assessed by manipulating the moisture content of the same treatment and measure respiration. This was (unfortunately) not part of the current methodology.

Ln 25-26:  How was this concluded from the data presented? No measurements were done in microbial hotspots. This is an assumption not a finding.

Ln 33: I don’t think we can exclude that there would be an effect of land use through micropores although the smaller the structure the less likely it is affected by land use.

Ln 38-39: Again, mineralization in microbial hotspots was not measured; this explanation is assumed. Make sure this is understood as the abstract should be understandable as a stand alone text. Also make clear that POM is not only POM, but POM + root biomass

Introduction:

Ln 43- 51: This description does not provide a complete picture of the most recent theories on C storage and mineralization - biochemical and physical protection through interactions between organic molecules and reactive soil mineral surfaces is missing.

Ln 70 and other places in the manuscript: The authors tend to mix up R and R2. Normally the small letter r is used for correlation coefficient. R2 for goodness of fit of regressions. See also Ln 262-263

Ln 80-82. The objectives are very clear. But it is not clear to me why you chose to measure basal respiration at field moisture content. To address the objectives wouldn’t it have made more sense to test respiration at different moisture contents for the same structure? This would have permitted you to separate microstructural effects from environmental conditions (like soil moisture content).

Ln 85: Can you explain what kind of bulk properties you had in mind for this?

Materials and methods:

Ln 95: No need to mention the part of the experiment with future climate conditions. This is not relevant for the current study.

Ln 115: Rather that “initial water saturation” isn’t “water filled pore space” a term readers may be more familiar with? I suggest to replace throughout the text and figures

Ln 121-125: The method used to measure POM weight needs better explanation. POM data are quite crucial for the main conclusions of the paper. Make sure it is understood how POM fractionation was done, how the soil was dispersed and if dispersion was complete. Also make sure  it is understood that what is measured here as the POM fraction includes root biomass present at the time of sampling. Do you have any idea what was the contribution of root biomass to POM in the different land uses? Did you not measure POM-C? This could provide a more accurate estimate of the amount of POM+rootbiomass?

Ln 146: Did the visual method also include root biomass in the POM fraction?

Ln 191-200: The PCA analysis is not mentioned in the statistical methods. Also please explain how you dealt with the psuedo-replicates in your models (2 and 3 cores per plot were analysed)?

Ln 201-203: Where are all these symbols clearly explained?

Results

Presentation of results is confusing; please make it easier for the readers to follow. I suggest:

• Present the results according to the research objectives in a coherent way.
• What is not directly related to the objective put that in supplement and vice versa.
• Too many symbols are used in text and graphs. Try to limit them or explain them more clearly.

Ln 230: Is the volumetric air content after glucose saturation a bulk soil property? What does it tell us? other than the conditions during which SIR was measured?

Ln 251: Did both POM measurement methods that were compared include the root biomass?

Ln 252-Ln 255: This belongs to the discussion or to the materials and methods

Fig 4: Please explain the codes. Graphs should be self-explanatory.

Ln 295-301: This fits better the discussion or conclusions section

Fig. 6 + 7: PCA model: It is not clear why the PCA was not used to inform the variables to be used in the multiple regression. Why not pick the strongest correlating parameters from PC1, and the strongest correlating parameter from PC2 as test model?

Fig. 6a: can the variation in respiration in cropland soils be explained at all? The model with only POM has a R2 of +/- 0.1, so can you say that POM is such a good predictor for respiration across land uses? or only in grassland?

Fig 6 and 7 do not match with in-text references (fig 7 and 8).

It would be nice to have the rational of the different model (e.g. all parameters minus microbial parameters) also explained in captions. Now hard to understand the different models and the ideas of the different models

Discussion

The discussion is well organized and generally clearly written, with some exceptions:

Ln 371: “ambient conditions”. Please be more specific. Rather than ambient conditions it is better to call this field moisture level. For example temperature was controlled, not ambient.

Ln 376-377: Fertilization/nutrient availability could also be an important explanatory factor, or the C/N of POM. 

Ln 386: “..showed that soil compaction reduced soil respiration (Liebig et al., 1995)” -> At what scale? And what moisture levels? In the current study rings were only 100 cm3 and quite dry.

Ln 391-395: Good observation about POM including root biomass. But this is a methodological artefact. What are the implications in terms of interpretation? Please discuss.

Ln 401-404: This is not a surprising finding, when considering structural effects. However you did not look at interactions with reactive surfaces so be careful you cannot make any conclusions on this not being a driver of carbon mineralization.

Ln 405: parameters overlooked could include nutrient availability, POM composition.

Conclusions

Ln 438: Be careful to suggest causal relationships where they were not tested.

Ln 446-450: Move this part to the Discussion. This is not a good way to end your conclusions section. Finalize with your main conclusion about your objectives, not bringing up a new issue for discussion/reflection.