Dear Dr He,

Many thanks for uploading you responses to both reviewers. Both reviewers thought that your study was a meaningful advancement for our understanding of forestry on peatlands, but pointed out some important issues with your analysis. I am satisfied with your clarifications on these points and in particular welcome the re-analysis of C stock changes using carbon contents of peats up to 50 cm, rather than 10 cm. Please implement the edits you outline in our responses into a revised manuscript, which I look forward to receiving.

With best wishes,
Jens-Arne Subke

Dear Dr He,

Many thanks for the revised manuscript. You have addressed the majority of concerns raised by referees well, and I’m generally minded to accept this manuscript for publication, subject to some further (minor) revisions. Below, I outline a number of detailed points (formatting, or editing) as well as a general comment that I would like you to address in the manuscript, which has not happened to a meaningful extent. My apologies for this becoming a two-stage reviewing process, but my hope is tat you can address these final edits relatively easily.

My main concern is the treatment of N2O in this study. The COUP model simulates some very high fluxes in individual years, which you accept at face value. Your sensitivity analysis shows that both the water table depth and the initial C/N ratios are parameters that influence this flux in particular. Looking at Figure 5, there are three years, which account for about half of the rotation’s emissions (1969, 1973 and 1977). Drivers as well as other simulated dynamics do not indicate anything unusual about these years. In your discussion of results, you speculate about possible influence of the water table, when you should be in a position to investigate more closely what the exact episodes of N2O emissions are. (As the annual averages of other parameters are not unusual, I suspect that there must have been extreme episodes leading to these estimates). It should therefore be possible to identify if these high fluxes are reasonable (i.e. whether there were indeed conditions for which experimental studies indicate high N2O production), or not. This should be investigated more closely, and also discussed, as the N2O emissions are critical to the discussion of whether or not the site is an overall GHG source or sink.

Detailed comments:
P. 2, l. 2/3: Delete “GHG”, and change “has” to “have”; I suggest “… which means a net emission of GHGs” for the end of the sentence.

P. 2, l. 14: “positive is an ambiguous word here, as in atmospheric science, the term generally refers to fluxes from the soil/ecosystem to the atmosphere. Pleaser avoid (e.g. by saying “Overall, the balance of photosynthesis and respiration in peatlands means that these systems act as C sinks, acting to mitigate climatic warming (e.g. Gorham, 1991).” (Note that this refers to CO2, rather than GHGs, but your original sentence also focused on photosynthesis and respiration, which excludes CH4 and N2O).

P. 2, l. 22/23: “3% of the Earth’s surface, of which in turn 10-20% have been…”

P. 2, l. 24/25: Make it clear whether this sentence refers to all peatlands, or drained peatlands in particular.

2/28: delete “e.g.”

3/24-25: Delete web-address here (can be added in methods).

3/29-32: Delete final sentence of this paragraph.

3/33-4/12: This description of the modelling approach should be integrated into the methods section.

4/29: Change title to “Modelling approach”, and incorporate former section 1.1 here.

5/13: comma after “decays”.

5/14: “added” rather than “adds”.

5/27-31: This is not written clearly. Please change to: “First, the soil layers are assumed the same over the 60 years simulated, and the soil physical characteristics in 1951 are assumed the same as measured in 2006. Whilst this assumption may not hold in detail, we consider any changes minor as 1) this site has been drained for many years (starting in the 19th century), and 2) soil properties…”

6/4-6: Rephrase to: “The higher drainage level of 0.8 m was set according to general forest management practices, also taking into consideration the maximum simulated soil depth 1 m.”

6/9: Table 2 does not provide such a C loss figure. Do you mean Table 3? If so, where is the actual figure in that?

6/7-17: Why are there references to Table 2 throughout this paragraph? Make sure that (1) the information you refer to is in that table, and (2) that you refer to tables and figures in the appropriate order!

6/23: Delete “similar to the previous calibration study”

7/3 (and throughout the document) spell “spruce” with a lower-case “s”.

7/23: delete “a” before “ca.”

8/22-23: Why is this offered as a speculation? You should have the model outputs to establish growth rates of understorey and tree biomass.

8/28: “1970” rather than “1970s”.

8/29-30: Explain what you mean by “the model’s difficulties to distribute the importance of trees and understorey layers”. Is this a parameterisation issue, or matter of programming/modelling concept?

9/26: “added” rather than “adds”.

10/3: “The linear correlation coefficient between…”. Is this R, or R2?

10/11-13: I don’t think this statement describes your results well. You have very high emissions peaks in your time series, which bear no resemblance of the magnitude reported in your other paper. I really think you need to look into the conditions (i.e. values or variations of drivers) that have led to your extreme annual fluxes, and either highlight conditions that lead to these, or (in case you think they are unrealistic), qualify these findings.

11/9: “previous cumulative”, rather than “hitherto”.

11/25-26: “Also, magnitudes of NEE and N2O fluxes are very sensitive…”

12/10: better: “The GHG balance over a rotational period for forestry on drained peatland is mainly determined by two large fluxes viz. plant C assimilation and peat decomposition.”

13/20-21: No – Figure 5 shows no variation in GWL that could explain the observed N2O fluxes!

13/21-23: Again, do you need to speculate, when your model should allow you to test whether N uptake by understorey was significant? I’m puzzled why you treat your own model as if it’s a black box that produces values you interpret by speculation.

15/30-31: “strong” rather than “a large”.

16/19: Delete the rhetorical question. Also: replace “tells” by “shows”.

16/21-23: This is poorly phrased. Also, what is the reasoning for your statement that an older forest will become a C source. There is evidence that also old forest stands can continue to sequester C. Use literature to back up any claims of a declining C sink.

17/5: Rephrase: “However, this is equivalent to releasing C from peat and storing it in agricultural soils…”

17/21: LULUCF has not been defined as an acronym.

18/5: “… peat soil over a full forest rotation.”
Figure 2 (in caption as well as axis title): Should be “absorbed” radiation, rather than “adsorbed”. (As the radiation can not be desorbed…”

Dear Dr. He,

Many thanks for the comprehensive revision. I think that your treatment of soil N status and modelled N2O emissions is much more insightful.

I’m happy for you to keep the very high emission pulses, along with the discussion of potential over-estimates due to the model’s exclusion of reductive processes during transport in the soil profile. It would be best o include this caveat also in the abstract, where at present you still have the 0.7 g N m-2 yr-1 statement. I suggest changing this to “up to 0.7 g N m-2 yr-1”.

I’m happy for this paper to go to print, pending only a very small n umber of typographic/grammatical changes (below).

Many thanks,
Jens-Arne Subke


Remaining edits:
p. 1, l. 29: add “up to” before “0.7”.

p. 8, l. 22: “Trees” or “Spruce trees”, rather than “The spruce’s”.

p. 8, l. 29: “stages”, rather tan “stage”.

p. 13, l. 24: Replace “the spruces” by either “spruce trees” or simply “trees”.

p. 13, l. 25: Delete “the” before “total”.

p. 13, l. 26: Place brackets around “Supplementary Fig. 1”

p. 13, l. 26/27: Rephrase. Suggestion: “During this period, the GWL was also deep (Fig. 5), resulting in high peat mineralisation rates and a net release of inorganic N.

p. 13, l. 29/30: Rephrase. Suggestion: “Besides, a high soil N availability also accounts for the ‘vegetation fitted’ model consistently overestimating spruce growth during this period (Fig. 2c).”