Overall the suggested changes would appear to accommodate most comments raised by both referees and lead to an improved manuscript. There are some remaining points of attention for the submission of a revised discussion text, enlisted here below.

for Referee 1: The suggested revised text in response of comment 5 and 29 and certainly comment 9and 23 is too long and should be further condensed. The authors’ solution to accommodate comment 10 seems too creative. At present it is not clear just how based on soil fractionation we could get a reliable notion on whether new C has been stored and if old carbon has been lost. Without radiocarbon dating or use of a transition from C3 to C4 vegetation this seems impossible. To some extent insight in the molecular-level biochemical composition of the SOM could indicate if such shifts occurred, but this is not what the authors intended to suggest in the revised version. It would be better to just recognize that the SOC stocks under barren land were probably not be the best estimate of SOC at time 0: also explain this in the revised discussion. Instead, do continue with the t0 estimate based on the exponential model that you introduce in your response to reviewer 2 and change the discussion accordingly. Further presentation of outcomes by statistical tests as requested needs to be taken along, use of medians is ok. Comment 25 still needs attention: Correlation coefficients always need to be accompanied by a significance value) freeware statistical software exists and also in excel linear regression analysis will provide a p-value.

for Referee 2: The suggested changes would appear to accommodate major comments raised by referee 2. Do pay attention to better account for referee 2’s suggestion to compare more to other relevant studies on SOC in chronosequences (i.e. Lillifein et al., contact perhaps the authors). Clarification of the distribution of samples over sites/categories/pits also requires some changes to figure 1, but the authors’ response was unclear here. The size of the presented maps also seems too small.

Overall, the authors have well taken action according to points raised by both referees. There are, however, still one major and some minor points of attention requiring the submission of a revised manuscript, enlisted here below.
Remove comma after ‘et al’ when the literature reference is part of the sentence: needs to be done corrected on many occasions throughout the manuscript.
When presenting correlations give r, not r²
Fig. instead of Figure – change everywhere
P9L13 ‘report’ instead of ‘reported’ + change last part of sentence into ‘report the initial SOC stock before the ecosystem change began’.
P10L28 The newly added part ‘Such pedologic conditions can ask for soil C decay studies, as it was introduced by Barré et al., (2010). However, this was not the objective of the present study’ is not clear with additional information. I suggest to again remove this reference to bare fallow plot studies – I do not see a straightforward link with this study.
P10 ‘3.3.1 Vertical resolution of unfractionated SOC stocks’ reads awkwardly, remove ‘unfractionated’
The present Conclusion section is really a summary of the present study with even report of p-values of statistical tests etc. This section is not to the point and strongly repeats parts of the introduction and discussion. For example: “Nevertheless, afforestation with mountain birch leads to a significant (p = 0.05) increase of the SOC stock (+15 t C ha-1; +48 %) for birch 25 stands between 15 and 50 years. During this time, the POM mass (+131 mg g-1 soil; +300 %) and POM-C concentrations (+35 mg g-1 soil; +285 %) increase during the succession of the mountain birch ecosystem. These increases were mainly observed in the top 10 cm of the mineral soil. Further, at least 56 % of the total SOC stock (0-30 cm) was found in the HF- and ‘< 63 μm’ fractions and at all tested categories most of the carbon was stored in the < 63 μm fraction.”
This whole part mainly just reports results once more, with no conclusion drawn on key messages, implications for management, new insight etc. Other parts as well are not generic enough and too focused on the conducted study. This entire conclusion needs to be rewritten:
-The first 6 lines of the conclusion are a repetition of preceding parts and need to go
- ‘ The results of the present study also clearly show that undertaking research on soil organic carbon patterns on severely degraded soils within this area is challenging, owing to the high SOC stocks (0-30 cm) of these degraded soils.’ is not a crucial nor generic conclusion and rather – if anywhere – belongs in the Results & Discussion
-Only form p15L29 more or less a more appropriate style is used -> use this as a guideline but severely un-vegetated soils may on the other hand need to be specified a bit more
Check carefully also for typos with spaces commas and periods in the final part of the conclusion

Dear Matthias

Many thanks for submitting your article to SOIL and engaging positively with the review process. The Topical Editor has now recommended publication of the article and it will now pass to the production team.

With best wishes

John