BG-2014-374 Helfter et al R1 comments
This manuscript presents important and useful information about interannual variability of carbon exchange by a temperate peat bog, quantifies environmental driver relationships with net and gross fluxes, and provides comparisons with other similar studies. It was reviewed favourably by two expert referees, however there are some important concerns that need to be addressed before it can be considered suitable for publication.

Abstract: As written, it lacks a concluding statement describing the implications of the study, and does not link well to the stated hypotheses. Please condense some of the details in order to provide one or two sentences placing this study into a broader context.
Introduction: Much of the writing could be explained more succinctly. For instance, remove extra wordiness:
P3, Lines 16-19, shorten to: In contrast to the variability exhibited by these sites, a sub-arctic permafrost mire in Northern Sweden was relatively stable over the period 2001-08 (-18 46 g C m-2 yr-1) (Christensen et al 2012).

Please re-cast the hypotheses to make them more useful and insightful, and then re-structure the discussion to address the hypotheses. Final acceptance of the manuscript will be based partly on appropriately improved hypotheses and discussion.

1) Peatlands release more CO2 to the atmosphere under drier and warmer conditions.
AE: Drier and warmer than what? Is seasonality implied? Was this really tested? It is more of a prediction than hypothesis.
2) The length of the growing season has a significant impact on annual NEE.
AE: What sort of impact? At least give an expected direction
3) Plant productivity during the growing season is affected by the preceding winter’s meteorological conditions and can exhibit significant inter-annual variability.
AE: affected in what way? What is considered to be “significant” inter-annual variability? Why is this important?
4) Autotrophic processes dominate heterotrophic processes at well-watered, vegetated sites.
AE: What is meant by “well-watered” and “vegetated”? Were non-vegetated sites investigated? The inter-site comparison does not distinguish these categories, nor does it distinguish autotrophic vs. heterotrophic processes.

Methods section: You seem to be missing a section explaining the statistics involved in testing your hypotheses and describing the curve fits, etc. In particular, please justify the use of (weak) non-parametric statistics (Spearman rank correlations) compared to stronger parametric statistics (ANOVA). Additional details regarding the modelling of the Reco and GPP data are required as outlined below.
P.5, line 31, do you have a citation for the drainage history?
P.7, line 26-29, please provide the model fit statistics, and include a plot of predicted vs. observed (PvO) Reco as a supplementary figure.
P.8, line 4, what were the reference temperatures you used? Provide the model fit statistics and PvO plot.
P.8, line 5, please clarify that GPP was estimated as the difference between measured NEE and modelled Reco (derived from the Arrhenius function in Eq 1). Also again, provide the fit statistics and PvO plot.

Results:
The paragraph starting on P.10, L11 and going to P.11, L18 is much too long and should be broken down into at least three separate paragraphs. Further, the details of the parabolic relationships between WTD and Reco are not compelling, and should be shortened. I agree with the referee 2 (comment 14) that the sign of the WT level would be easier to understand if reversed, but will leave it up to the authors to decide which direction to take.

Discussion:
This section could be improved by starting it with a short paragraph summarizing the main findings of the study with reference to the hypotheses (after revising them). This paragraph can serve as an outline for discussing the most important results in the following sections. Overall, this section could be improved by organizing it around the hypotheses. Each paragraph should discuss one main idea, starting out with a topic sentence. For instance, the first paragraph moves from PAR, LAI in soybeans to a warming experiment and then freeze-thaw cycles. It would be improved by focusing on the most relevant comparisons with the literature. The paragraph on P.12, L29 to P.13, L19 is an example of a paragraph that is well written, with a more focused message.
P. 13 L4, Please replace “feedback” with “effect,” unless you can demonstrate a specific cause-effect relationship between photosynthesis and WTD that works in both directions. Likewise, on P.12, L.29, and throughout the manuscript. On P.17, L4, the idea of an internal feedback mechanism is unclear as well; rather, it seems like you are simply trying to say that the C balance of the different peatland ecosystems is similar when all flux pathways are considered.
P.14, L3, what does species richness have to do with your hypotheses? The relevance of this paragraph is not clear, and it mainly reiterates results.
P. 14, L.17-19: “The decrease of the sensitivity of Reco with respect to Tair (Q10) with deepening water table further supports the idea that the contribution of heterotrophic to total ecosystem respiration was enhanced under drier conditions.” Please provide a justification and citations for this idea. Overall, the speculations about autotrophic and heterotrophic processes are poorly supported by the available data and should be removed or supported with additional data.
P.16, L29-31, make sure the sign for NEE (sink strength) is consistent.

Summary: Please revise in light of your new hypotheses and discussion.

Thank you for your careful revisions. The manuscript has been improved through the review process and is now ready for publication.