The manuscript aims to determine the effects of soil parent material and landscape position on soil organic carbon contents and retained by the mineral matrix in tropical soils, which is a very relevant issue to be considered in understanding soil carbon turnover and biogeochemical processes, especially in areas with limited available information as in the studied area. While the manuscript is well organized and written, I identified the following shortcomings that need to be addressed to improve the clarity of the paper:

1) The authors use the term “organic carbon stabilization” in the title and in many other parts of the manuscript. It could be stabilization against decomposition, temperature, erosion, dispersion, oxidation, or all these together. However, the authors do not provide a context in which the word "stability" is used.

2) In the introduction (section 1.2 Environmental and geochemical controls on SOC dynamics in tropical forests), you never talk about very important environmental controls such as temperature, moisture, pH, redox potential, and oxygen diffusion. You should explore the optimum conditions (e.g., temperature and moisture) for enzymatic activity in the tropics, which ultimately will determine organic matter decomposition rates.

3) It is not clear in your hypothesis ii, the proxy for the amount of “stabilized” SOC. Is it the carbon content adsorbed on clay minerals? Does it also consider the carbon occluded in microaggregates that are physically stable against dispersion?

4) How many soil profiles were sampled? What is the classification of these soil profiles that you sampled? It needs to be shown in the methods section “2.2 Study design and soil sampling”.

L22: I think you meant “changes in hydraulic conditions”.

L23/24: What is “fossil organic carbon (FOC)”? Is it defined by a specific organic compound (e.g., polycyclic aromatic hydrocarbon) or by a certain “age” (e.g., >10000 years)? The authors should define “fossil organic carbon” when they first mentioned it in the abstract.

L29: What depth interval are you considering for this observation about SOC stocks? 0-20 cm, 0-50 cm, 0-70 cm, 0-100 cm?

L66: What do you mean for “stabilize C in the soil”? stabilize against decomposition? For how long? 100, 1000, 10000 years?

L66-68: Please, provide the reasoning for why pyrogenic carbon may remain in the soil for centuries. Does it have to do with the following? The degradation of condensed aromatic carbon compounds is an energy-demanding process. For degradation to occur, the biological capacity for specific degradation pathways must exist in the soil [Baldock et al., 2004]. It is likely that certain microorganisms can produce the (costly) enzymes required for pyrogenic carbon degradation and their presence or absence from soils is thus a crucial control on pyrogenic carbon accumulation. However, images of fungal hyphae encasing charcoal particles in soils [Hockaday et al., 2007] indicate that fungi can very well decompose pyrogenic carbon.

L69: Are you sure that the word “recalcitrant” is the best one to describe what you meant here? Maybe using terms like “long turnover times” may help you to overcome this issue. I highly recommend the authors to read the following publications for further ideas and suggestions:

Mikutta, R., Kleber, M., Torn, M.S. and Jahn, R., 2006. Stabilization of soil organic matter: association with minerals or chemical recalcitrance?. Biogeochemistry, 77(1), pp.25-56.

Kleber, M., Nico, P.S., Plante, A., Filley, T., Kramer, M., Swanston, C. and Sollins, P., 2011. Old and stable soil organic matter is not necessarily chemically recalcitrant: implications for modeling concepts and temperature sensitivity. Global change biology, 17(2), pp.1097-1107.

L116: Define “geogenic carbon”? Is it the same as “fossil carbon”?

L117: You said “more active microbial communities”. However, you didn’t conclude the comparison. Is it more active than what or where?

L140: Check the concordance of the preposition "on" and the verb “affect” in this sentence: “our current understanding on how geochemistry and topography in highly weathered tropical soils affects SOC stocks and stabilization mechanisms is still limited”

L225: What method did you use for “texture” determination? Hydrometer, pipette?

L248: Did you subtract the rock fragment to calculate SOC stocks?

L329: Please, check if this statement is necessary: “For non-valley positions, pyrophosphate extractable oxides (0.02 to 1.93 mass%) and oxalate extractable oxides (0.32 to 2.33 mass%) were low compared to DCB extractable oxides”. Assuming that DCB extracts most of the oxides (including oxalate and pyrophosphate extractable oxides), then, DCB extractable oxides will always be the highest regardless of the landscape position.

Figure 3. It would be helpful if we know the soil types. Are they all ferralsols? Or the soils derived from “mafic” parent material are lixisol and nitisol as well? Also, it would be nice to see in another figure and table the average clay content as a function of soil depth in each group: mafic, felsic, mixed.

L539: Iron concretions are found, but I am not sure about aluminon concretions. Check whether all these soils contain Fe-concretion or nodules because these concretions are not always found even in microaggregates.

L540: The citation Martinez and Souza, 2019 is actually from 2020. Check it in other parts of the manuscript.

L543: Check the following publication to obtain more information about the effects of soil depth and texture on carbon retention in Ferralsols. They may help you with new insights and allow comparisons with your data. Soil depth and texture seem to regulate the content of carbon retained by the mineral matrix in Ferralsols. Make sure you addressed it in the discussion.

Souza, I.F., Almeida, L.F.J., Jesus, G.L., Kleber, M. and Silva, I.R., 2017. The mechanisms of organic carbon protection and dynamics of C‐saturation in Oxisols vary with particle‐size distribution. European Journal of Soil Science, 68(5), pp.726-739.

Souza, I.F., Almeida, L.F., Jesus, G.L., Pett-Ridge, J., Nico, P.S., Kleber, M. and Silva, I.R., 2018. Carbon sink strength of subsurface horizons in Brazilian oxisols. Soil Science Society of America Journal, 82(1), pp.76-86.

Review Comments

1. I) General comments

I consider the submitted manuscript merits publication due to novel and interesting findings are reported covering tropical soil which has been so far neglected. In general, the manuscript is well written and structured.

However, the most important concern for me is the research design, why the slope plots are 6 while valley and plateau are 3 each?

How could you justify the absence of soil erosion at the sampling plots as mentioned in your results section while sediment soils were collected in the valley plots? Kindly elaborate on this in your discussion.

Though in forest, there is no or little surface soil erosion but landslides occur most of the time in Tropical forest

I understand that one of the main focuses was Geochemistry of the three sites, while you also recorded soil fertility parameters especially chemical. but anywhere biological parameter such enzymes related to carbon cycle such as b-glucosidase which decompose carbon source is important for the decomposition were recorded? or mentioned in relation to the study.

1. II) Specific comments

1) could you elaborate why the slope of you compared sites of different slopes eg., Kibale site (3-55%) while Kahuzi-Biega and Nyungwe have similar slopes (1-60%).

2) Soil development is dependent to several factors including environment, which are different from the three. What was your reference of standardization to justify the study comparison between the sites.

3) in material and methods you are referring to Fick and Hijmans, 2017. is his work conducted in all your study regions?

4) How could you compared results from three different slope length beside variability sites elevation. e.g., Nyungwe and Kibale sites are more variable than Kahuzi biega.

5) How can you explain the absence of soil erosion in plots, while the sampling was conducted during rainy season (March to June)

6) on the paragraph 340 “Not significant correlation was found with the included climate variables (data not shown); For the reader to realize that they were not significant difference a figure or table is require. can you add that?

7) on 360; You mentioned that total P was high in Malfic region as compared to both mixed sedimentary rocks and felsic regions. How could you justify your finding with the known situation of low available P in that region?

8) At 365; How the reader could know that the study region is highly weathered without displaying soil data of all the three depths?

9) at 375; How could you explain the high depletion of P (72 to 14 %) in parent material under natural conditions? (without any agricultural activity that could contribute to P removal)

Dear editor and authors,

I really appreciate the opportunity to read and comment on the manuscript Soil-2020-92 entitled “The role of geochemistry in organic carbon stabilization in tropical rainforest soils”. Overall, the manuscript is very interesting and covers a topic of global interest, and as such, it would fulfill the requirements for publication in your Journal. I would recommend accepting, although a few minor issues should be dealt with first. I acknowledge the fact that other reviewers participated in the discussion and gave detailed suggestions to improve the original version of the manuscript. For the version of the text I have read, there are no major issues that would require extensive reworking. So, only minor issues will be pointed out in the following text and I hope my comments can still be useful.

Abstract

Lines 22-23. I could not find strong evidence to support the claim that “fluvial dynamics and changed hydrological conditions had a secondary control on SOC dynamics in valley positions, leading to higher SOC stocks there than at the non-valley positions”. This should be better explained. How can the reader agree that “fluvial dynamics and changed hydrological conditions” can be inferred from the results reported and help to explain higher SOC stocks in valleys than in non-valley positions?

Lines 23-24. I believe the term “Fossil organic carbon” could be more precisely described and referred to as “geogenic organic carbon” in the whole manuscript. In fact, geogenic organic carbon was used by the authors themselves elsewhere in their text (e.g., page 3, line 116).

Introduction, Section 1.2. Environmental and geochemical controls on SOC dynamics in tropical forests

Lines 69-70. To what extent geogenic carbon (FOC) is preserved owing to its inherent chemical properties (e.g., recalcitrance) or the specific conditions (e.g., burial of organic carbon mixed with mineral particles) under sedimentary environments?

Line 86. Please provide a reference in which the authors have reported pedogenic oxides contents above 50% in the clay fraction of tropical soils. I agree that some tropical soils can exhibit more than 50% of pedogenic oxides, but such soils are not the norm as implied in the text. Please check.

Introduction, Section 1.4 Study aims

Line 140. “composition” or “decomposition”?

Hypotheses

All hypotheses proposed should be rephrased and put into simpler functional relationships e.g., y = f(x). This would reduce verbosity and give the reader a glimpse on how each hypothesis would be effectively tested. I believe the hypotheses can be better used to guide the reader through the Discussion section as well.

Hypothesis (i): what parameters would be used/measured to determine the control of topography on lateral fluxes of water and mineral mass? This is not clear to me.

Hypothesis (ii): I understood the context, but verbosity can be reduced.

Hypothesis (iii): I found the third hypothesis particularly confusing as it includes a reference to “priming effects”, which were not measured in this study.

Example to rephrase the third hypothesis:

iii) Geogenic soil carbon stocks vary more consistently as a function of soil depth than landscape position or soil parent material.

Results, Section 3.1 Climate and topography

I believe the supplement 1 could benefit the reader if kept in the main text, all results reported therein are very nice. Besides, as far as I understood hypothesis (i), the observation of higher soil C stocks in valleys than in non-valley positions is important for this research.

Lines 350-353. The inference that “Even though valley positions are of the same geochemistry as the non-valley positions, geochemical soil properties in valleys were significantly different than at non valley positions, as fluvial activity and sedimentation unrelated to hillslope processes were dominant”, seems quite speculative to explain higher C stocks in valleys relative to non-valley positions. In my opinion, a predominant effect of “fluvial and sedimentation” rather than “hillslope processes” would make sense only if the geochemistry in the valleys were significantly different from that observed in non-valley positions.

Results, Section 3.3 Variation in SOC properties with geochemistry

Lines 410-411. In the sentence “Note that while SOCbulk decreased strongly with depth in the mafic and felsic region, only a weak decrease of SOCbulk with depth was observed in the mixed sedimentary region (Fig. 4)”, can we infer that SOC buildup followed the accumulation of sediments over time to a greater extent than C inputs from the local vegetation? How does the ¹⁴C depth-trend compare to that observed in valley positions as shown in Fig. S2?

Results, Section 3.5 Explained variability and relative importance of predictors (non-valley soils)

Lines 471-472. Based on the observation that “Soil depth and rPC4_nv explained 73 % of variability (R²) in SOCbulk (p < 0.01). Soil depth contributed 82 % to the explanatory power of the model”, how (in)sensitive tropical C pools may be to changes in climate or land use?

Results, Section 3.6 Partial correlations controlled for soil depth

It looks quite amazing that when the effect of depth is controlled, the explanatory power of the other variables included in the model does not increase substantially (except silt content). How does this trend compare to temperate ecosystems? What can be inferred about the relationship between pedogenesis and soil C accumulation in the tropics? What is the mineralogy of the silt fraction? Given the data shown in Table 3 and Figure 5, such information is very important for this research and would facilitate the discussion (lines 570-578).

Discussion, Section 4.1 Soil C stabilization driven by soil chemistry and parent material

To what extent the inference that “In contrast to our initial hypothesis that topography affects C stabilization in tropical forest soils through lateral material movements, we found no indication of this in our analysis (Supplementary results and short discussion therein)” can be reconciled with the observation of higher SOC stocks in valley positions, despite exhibiting similar geochemistry to non-valley positions?

Conclusions and outlook

Lines 410-412. “Differences in Δ¹⁴C were best explained with soil depth and the presence of FOC, which appears to be decomposable by microbial communities under more fertile, topsoil conditions.” There is an apparent redundancy here since Δ¹⁴C would co-variate with FOC and factors limiting microbial respiration at depth should be more important than soil fertility.