The study by Keskitalo et al. aims to decipher carbon sources and their drivers in an Arctic river and its tributaries. The authors focus on particulate organic carbon (POC) and are particularly interested in differences between two seasons, spring freshet and summer, as well as differences among different sizes of streams/rivers and the mainstem of the Kolyma River in Siberia. The authors found differences in POC sources between seasons and in small tributaries compared to larger streams/rivers. With this study, the authors want to contribute to a better understanding of the carbon dynamics of the often overlooked smaller tributaries in the Arctic and their sensitivity to climatic stressors. The study provides important results for a better understanding of streams/rivers of different sizes, their carbon dynamics, and their sensitivity to climate warming. Nevertheless, I have questions about the data and the statistical analyses of the data that I would like the authors to address. Please find my comments in the order they appear in the manuscript below.

Title: I like that the authors formulated subheadings in the discussion that summarize the main findings. However, the title of the manuscript is very descriptive stating that the study is about “Seasonal carbon dynamics of the Kolyma River tributaries, Siberia”. I would like to ask the authors to think about rephrasing the title and summarize the major result(s) and better highlight the focus of the study, the POC. Furthermore, the authors do not only report seasonal dynamics but also look at the role of the smaller sized streams compared to the mainstem of the river. A well-summarized title might also encourage more readers to look at the article.

Last sentences of abstract (lines 20/21): “As Arctic warming and hydrologic changes may increase OC transfer from smaller waterways through river networks this may intensify inland water carbon outgassing.” I do not see the link between carbon transfer and outgassing. I miss the conversion of carbon, i.e. decomposition in the system that leads to higher CO₂ production through respiration, which can then cause a higher outgassing rate than at other times. I also do not believe that an earlier onset of primary production in tributaries compared to the mainstem during freshet necessarily leads to higher evasion rates. This process in itself actually leads to a reduction in CO₂ levels. I therefore suggest rewording the last sentence to fit the reported results or adding one or two more results from this study to make the connection here.

Statistical approach: I wonder if the Welch’s t-test for testing differences between the two seasons is correct here for this study design. The assumption of this test is that the two groups are independent. At the same time, the authors want to test whether spatial characteristics in the watershed influence carbon dynamics at a sampling point. This implies, in my opinion, that the authors assume that the location of the sampling point in the landscape influences water quality and carbon pools. Hence, the two samples collected in two seasons at the same site might be more similar than the others and should be paired for the statistical test. Can the authors please explain why they use the independent Welch’s t-test or change their statistics if there is no justification for choosing the statistical test. I have one additional comment about the statistics. The authors use simple linear regressions to investigate how environmental factors influence carbon dynamics (Fig. 2). They state that they want to “examine how watershed characteristics control carbon concentrations.” (lines 18/19). This could be done by running multiple linear regressions or linear mixed models with sampling site as a random factor (to account for their dependency) to see which factors are “most relevant” for controlling carbon dynamics. In this way, they could incorporate several independent variables.

The authors measured POC and particulate nitrogen (PN) and also show ratios of POC to PN in table 1. The C:N ratio can also be an indicator of algal or terrestrial material, with ratios around 8 being of algal origin and with increasing ratios being more terrestrial. Please see Figure 1 in Meyer 1994 (Meyers, Philip A. "Preservation of elemental and isotopic source identification of sedimentary organic matter." Chemical geology 114.3-4 (1994): 289-302.). Perhaps this could be included in this manuscript and highlighted in the discussions. For example in line 238.

At the beginning of the discussion before the first subheading (after line 229): It would be nice to insert here a summary of the main findings in relation to the main objectives formulated in the abstract (lines 17-19) and at the end of the introduction (lines 36-38).

Lines 347-352: “While POC concentrations did not significantly differ between large and small/midsized rivers during freshet, composition of POC showed clear differences: the δ13C-POC was lower and POC-% higher in small and midsized streams/rivers than in large ones, indicating an early onset of primary production in these lower order streams. This may fuel CO₂ evasion via degradation of autochthonous POC that is likely partly comprised of permafrost OC and/or prime degradation of allochthonous OC, however, further studies are needed to discern implications on CO₂ emissions in a system level.” I like the conclusions the authors draw here. They highlight very nicely the most important results and implications here that I think are worth publishing. However, when primary production is higher, the authors usually also conclude that there is higher CO₂ evasion. I am not sure I follow this interpretation. Also in the discussion, the authors interpret their data in a similar way. Although one cannot rule out the possibility that more CO₂ is emitted when primary production is high, the direct consequence is that more inorganic carbon is taken up. Demars and colleagues nicely discuss the balance between primary production and respiration in streams as temperatures rise. They conclude that warming will not lead to an increase in CO₂ emissions in streams and rivers. See Demars et al. 2016 for a discussion on this topic (Demars, Benoît OL, et al. "Impact of warming on CO₂ emissions from streams countered by aquatic photosynthesis." Nature Geoscience 9.10 (2016): 758-761.). This comment is related to the one about the last sentence of the abstract. 

See attached pdf