Dear Authors,

Thank you for submitting your manuscript to Biogeosciences. Your manuscript has been reviewed by two reviewers with expertise on GHG dynamics in inland waters. While the reviewers agreed that your manuscript presents comprehensive, invaluable data sets collected in a “globally important, yet understudied” river system for multiple years, their evaluations of the paper quality were starkly contrasting. Irrespective of the overall evaluation, both reviewers raised a number of concerns about not only the structure but also interpretations of some critical data. Therefore, I had to reread the manuscript carefully and double checked on which factual grounds the two reviewers had based their evaluations. Your responses to reviewer comments were thoughtful and detailed, but some important points remain inadequately addressed, as described below.
Given the many raised issues and several points inadequately addressed, I might need to reconsider your manuscript for publication in Biogeosciences after major revisions. I would like to ask you to make all the changes easily identifiable in a marked-up manuscript and a point-by-point reply to reviewer comments. I would also suggest you to specify the line numbers of the revised parts when you respond to the reviewer comments and my own suggestions as follows:

1. Manuscript structure and focusing
In my view, the way you structured data presentations and interpretation lies at the heart of many concerns raised by both reviewers and the second reviewer’s negative evaluation. First, I agree with the first reviewer saying that “we should be grateful for the authors’ persistence and long-term dedication for studying this very important river basin”. I also agree to your statement that “long papers presenting large and broad data-sets…are not automatically of poor quality”. However, both reviewers and myself, who all appear keenly interested in the topics addressed in this manuscript, found it difficult to follow many parts due to long and often redundant descriptions. As suggested by the first reviewer, you don’t need to remove any of all the important results, but some efficient editing throughout the manuscript would make it a lot easier for readers to follow up the key points you want to highlight. I don’t think that your plan (a paragraph at the start of the results-discussion section that explains the “road-map” of the paper) would address the concerns raised by the reviewers and myself.
Please first consider how to streamline the logical structure of your hypothesis (presumably small carbon export from terra firme – contrasting spatial patterns of CO2/CH4 vs N2O primarily determined by wetlands in CCC – relatively small in-stream respiration compared to riverine CO2 and CH4 emissions). The first reading of the abstract and three R&D sections did not allow me to understand your major point. Furthermore, the Conclusions section is solely focused on your hypothesis regarding the dominant role of terra firme, even though you wanted to present the full spectrum of large data sets as the rational for a long paper. As suggested by the second reviewer, if you want to weave multiple pieces of information into a single coherent storyline around wetland contributions, you might need to consider refocusing and even splitting R&D. Otherwise, please clarify your goals (and hypotheses) in the abstract and introduction and accordingly provide summary paragraphs in each relevant R&D sections that could be understood as stepping stones leading to conclusions. In the current version, key messages are hidden, scattered in separate R&D sections and conclusions focused on single hypothesis spring up suddenly.
Given the vast volume of data, the long abstract itself is not a problem at all. However, it could be more precise by rearranging some long and redundant sentences. For example, the lines starting from L 12 to 34 on P 3 can be shortened by focusing on key spatial patterns (please refer to the first reviewer’s comment on long “explanations of correlations”). In addition, the next page repeats the same spatial patterns, but this time reiterates spatial variations in fluxes. The key message on the role of CCC appears late after just describing variations in concentration in the previous page. Again, streamlining the story based on your key questions would also reduce redundancy here in the abstract.

2. Alternative explanations
The two reviewers raised several issues about balancing alternative explanations for the observed patterns. In responding to the first reviewer’s question about “what extent patterns observed in the river depend on in-river processes or on what is brought to the river from the soils”, you said that “We have modified the text accordingly”, without detailing how you would change in which parts. The second reviewer also raised a critical question about the contribution of wetland-derived OM and CO2 to downstream CO2 emissions (river respiration and retention time). I agree with the reviewers in that your approach is too straightforward focusing only on your hypothesis, not balancing alternative explanations. Please clarify the following uncertainties in relevant Discussion parts and draw your conclusion more carefully considering all these uncertainties (in Conclusions):
- Estimates of NEE and the proportion of the hydrologic C export (including DOC, POC, and DIC) in NEE (you just assumed that C export would comprise a small portion, without referring to any measurements or estimates)
- The relative contributions of OM and CO2 derived from wetlands in CCC (no data available on measurements in source sediments and waters?)
- Estimates of PP and CR: I am afraid that many reader would not be convinced by your explanation (“it seems unrealistic to envisage an under-estimation of CR by an order of magnitude that would allow reconciling the CR (and NCP) estimates with those of FCO2”).

3. Methodological details
Some of your responses to technical questions are inadequate in terms of providing details. For instance, in responding to the first reviewer’s questions about k estimation, you only mentioned your plan for error analysis. In another example, your reply to the second reviewer’s questions about pump depth and boat speed could not address the technical issue about how to avoid aeration problems arising from the equilibrator employment at high speed. In addition to citing your previous papers, you could briefly describe technical details in the relevant Methods section. The same goes to your statistical analyses.

Sincerely,

Ji-Hyung Park
Associate Editor, Biogeosciences