the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 12 Mar 2026
The carbon balance of a pristine subarctic river in northern Fennoscandia
Abstract. Carbon emission estimates from riverine ecosystems at high northern latitudes are rarely assessed, resulting in significant uncertainty in the carbon budgets of this region. To close this gap, we determined chamber-derived carbon dioxide (CO2) and methane (CH4) fluxes alongside surface partial pressure measurements within the Teno river catchment. The Teno river represents an undisturbed, natural ecosystem at the border between northern Finland and Norway. Chamber measurements were conducted on the water and at river banks (i.e. vegetation-free parafluvial and partially vegetated areas) and yielded predominantly low fluxes. Water-air CO2 fluxes amounted to 0.25 ± 0.27 μmol m−2 s−1 (mean ± standard deviation), and water-air CH4 fluxes to 1.77 ± 3.19 nmol m−2 s−1. Soil-air CO2 fluxes from banks reached 1.45 ± 1.38 μmol m−2 s−1 for ecosystem respiration (Reco), and 0.55 ± 1.34 μmol m−2 s−1 for net ecosystem exchange (NEE), while corresponding CH4 fluxes were 1.03 ± 3.14 nmol m−2 s−1. Using hydrographic datasets, we upscaled our observations for water-air fluxes from the river channel and for soil-air fluxes from the banks. Accordingly, the total vertical carbon emissions amounted to 3,017 tC yr−1 for CO2 and 3.3 tC yr−1 for CH4, while banks overall were estimated to emit more than four times the amount of CO2 than the river channel. Furthermore, we determined the lateral carbon export of the river towards the ocean and estimated a total annual export of roughly 70,000 tC yr−1 from Teno river, mostly in the form of dissolved organic carbon and bicarbonates. As a result, the river network acted as a small carbon source to the atmosphere and the ocean. Carbon dioxide dominated the net carbon balance of the river, while CH4 emissions were minimal. Although the measured fluxes and the carbon balance of the river network were small in comparison with the landscape-scale carbon balance, the study provides important baseline data for pristine, subarctic rivers, a data-poor ecosystem in current literature. As previous flux measurement sites may be biased towards high-emission ecosystems, it is important to highlight the significance of carbon dynamics for low-emitting sites in order to achieve a balanced understanding. In addition, this study provides insight into the land-water-atmosphere carbon dynamics of northern rivers, and can be used as benchmark for Earth System and process-based models.
- Preprint
(24963 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2026-1260', Anonymous Referee #1, 17 Apr 2026
-
AC1: 'Reply on RC1', Judith Vogt, 06 Aug 2026
We thank the reviewer for their comments and appreciate their critical feedback on our manuscript. Below, we respond to the reviewer’s comments (restated in italics).
1: The quantification of an annual carbon balance is the central claim of this study, which rests on a data foundation that is actually not fit for this purpose. The analysis combines data from three campaigns conducted across two different calendar years (August 2023, June 2024, September 2024). This approach treats the system as if it were in a steady-state "average year," which is a highly questionable assumption given the hydrological extremes observed.
We appreciate the reviewer’s thoughtful critique regarding the representativeness of our annual carbon budget estimate. The revised manuscript will highlight key components and processes that shape riverine carbon export in the studied region.
We acknowledge the observed variation in hydrological conditions with August 2023 reflecting medium discharge, June 2024 high discharge and September 2024 low discharge. However, we argue that the very heterogeneity of our sampling - spanning from high-flow to low-flow conditions is a strength and captures various states of hydrology in the river.
To deepen our analysis, we will add a sensitivity analysis showing how the annual net carbon budget estimate of the river changes under different scenarios (e.g., only considering fluxes associated with low-, medium- or high-flow, and neglecting seasonality). This analysis will demonstrate that our initial estimate was conservative, but also shows considerable variation, e.g. for scenarios where less importance is given to seasonality. Therefore, rather than attempting to infer a single "typical" year, our revised approach will capture a range of hydrological conditions that may influence carbon fluxes across seasons and throughout the year, especially under consideration of river states affected by climatic extremes. We will compare and discuss the results of the different scenarios in the revised manuscript.
We will clarify that the previously estimated annual budget is not presented as a "representative" average year, but as a best-estimate composite based on observed hydrological conditions, and we will emphasize that this approach reflects the non-stationarity of the system. The annual riverine net budget estimates from the different scenarios will provide the reader useful information, especially given the limited availability of carbon flux data from comparable subarctic rivers.2: August 2023 was characterized by 54% above-average precipitation and a specific storm event (peak intensity 15 mm h⁻¹), while September 2024 was an extreme low-flow period with tributary desiccation. By stitching these together, the authors cannot distinguish between seasonal succession (summer vs. autumn) and interannual hydrological forcing (wet year vs. dry year). The resulting annual flux (e.g., lateral export of 70,000 t C yr⁻¹) is therefore not a robust estimate. It is a composite of two distinct hydrological states that may not represent any single real year.
In the revised manuscript, we will embed a sensitivity analysis showing water-air carbon emission estimates of the river under different scenarios of hydrological states. The analysis shows that the initial budget estimate represents a possible yearly budget that may not be representative for any single real year, but gives a conservative estimate of carbon fluxes resulting from different discharge levels throughout a year. Therefore, the results of this analysis will provide additional insight into the carbon budget of the river.
Furthermore, in the revised manuscript, we will more clearly bring forward the limitations of the lateral carbon export estimate, which is based on measurements from one campaign only.3: Beyond presenting the data, the manuscript offers limited analytical insight. The findings are largely descriptive: "X is low," "Y is higher than Z." The paper does not sufficiently grapple with the why in a way that generates transferable knowledge.
In our study, we found that vertical fluxes of the subarctic river were dominated by CO2 rather than CH4, and that lateral fluxes dominated the overall carbon release of the ecosystem. Besides, we found that tributaries showed lower fluxes than the higher-order sections and that banks emitted more carbon than the river. All these insights are important for studying aquatic carbon budgets in Teno river, but also in other systems. We will shorten very descriptive parts of the manuscript, and dive deeper into the analysis of these findings.
We will look more into the reasons for the low-order stream paradox, explore how silicate weathering of the Lapland Granulite Belt locks the catchment into a carbon balance dominated by lateral bicarbonate export rather than vertical outgassing, and further examine different scenarios and outline limitations of the annual carbon emissions of the river.4: The introduction emphasizes the importance of hydrological connectivity, yet the analysis of the data relies on rudimentary correlations (e.g., Spearman with catchment size). There is no attempt to model or conceptualize how varying water sources (groundwater vs. soil water vs. direct precipitation) dictate the observed pCO₂ dynamics.
While we tried to implement machine learning models (random forest), the gaps in our dataset did not allow for an establishment of a feasible model. Therefore, we rely on rather simple statistical tools to determine pairwise relationships between variables. In addition to the presented data, we also measured carbon isotope ratios (d13 C-CO2 and d13 C-CH4) and ion concentrations at selected sites. However, ion concentrations were mostly found to be below the detection limit. Based on our data, we are therefore not able to derive the sources of the water between groundwater, precipitation or runoff. However, we will add results of the isotopic signatures in the revised manuscript, suggesting that carbon around tributaries likely originated from wetlands.
The observation that low-order streams had lower emissions than the main stem contradicts established paradigms (Hotchkiss et al., 2015), but this is merely noted in passing rather than explored as a potential mechanistic feature of this specific lithology or landscape.
In the revised manuscript, we will more closely tie the findings to the lithology and landscape of the study area, following this general line of argumentation: The observation that low-order streams in the Teno river exhibited lower CH4 emissions than the main stem contrasts with the widely reported headwater-dominated emission paradigm (e.g., Hotchkiss et al., 2015). However, this reversal is consistent with the unique geological and landscape characteristics of our subarctic study area.
The Teno river basin is underlain by Precambrian bedrock (granite, granulite, gneiss, and schist) and is characterized by glacially rounded fells with limited soil development (Mehr et al., 2022; https://doi.org/10.1016/j.ecolind.2022.109203). These lithologies are typically low in organic matter and nutrient content, and the cold, subarctic climate further limits biological productivity and organic carbon accumulation in headwater catchments. As a result, low-order streams – which are primarily fed by snowmelt and shallow groundwater from these weathered bedrock zones – receive limited labile organic carbon inputs, constraining CH4 production.
In contrast, the high hydrological connectivity in larger reaches enables longer water residence times and enhanced in-stream processing of carbon, as supported by Catalán et al. (2016; https://doi.org/10.1038/ngeo2720). This promotes microbial activity and CH4 production, particularly in slow-flowing, low-gradient sections (like LL) where carbon can settle and react.
Therefore, the reversal in emission patterns is not a general anomaly, but a landscape-specific response driven by the interplay between low carbon availability in headwaters (due to bedrock lithology and cold climate) and enhanced carbon processing in larger, more connected reaches. This underscores the importance of local geology, and hydrological connectivity in shaping riverine carbon dynamics – and cautions against overgeneralizing emission patterns from temperate or forested systems to subarctic, bedrock-dominated landscapes.5: The manuscript concludes that vertical fluxes are negligible (<1% of catchment uptake). While numerically correct based on a literature-derived catchment uptake value, this framing undermines the study's own justification. If the process is negligible, the paper reads as a data dump rather than a critical piece of the puzzle. The discussion would be strengthened by shifting focus from "it's small" to "under what hydrological conditions does this pristine river transition from a passive pipe to an active source?"
We will replace “negligible” with “orders of magnitude smaller”. We believe that measuring small fluxes is a valuable finding as well and should not affect the justification of the study in itself.
Crucially, our data show that the Teno river was a source of both CO2 and CH4 to the atmosphere, even in its smallest reaches. While the absolute fluxes were small, they were not negligible in ecological terms, but represented a component of the river’s biogeochemical function. The key insight is not that the river is “passive,” but that even in low-carbon, high-connectivity systems, rivers can be active carbon sources, albeit at low rates.
Therefore, rather than framing the river as a “passive pipe” or a system on the verge of becoming active, we will emphasize that this pristine subarctic river is a low but persistent carbon source to the atmosphere, with spatial variation. This highlights the importance of measuring small fluxes to understand the full spectrum of riverine carbon dynamics.
We believe that quantifying these small emissions is essential for building accurate regional and global budgets, especially in understudied subarctic systems.6: The authors position Teno as a benchmark for pristine systems against which disturbed sites are compared. However, the lithology (Lapland Granulite Belt) and specific land cover (broadleaf forest/shrubland) are highly specific to this region. Without a stronger analysis of how these landscape drivers (weathering rates, low wetland connectivity) cause the low fluxes, the utility of this benchmark is limited to "rivers that look like Teno." The paper does not provide a framework for predicting whether another pristine river (e.g., in the Siberian lowlands) would behave similarly or differently.
We will emphasize that this study is not intended to generalize across all high-latitude rivers, but rather to provide a detailed characterization of pristine, low-carbon river systems in northern Europe – specifically those with gentle topography, intermittent connectivity in low-order streams, thin organic-rich soils, and neutralizing bedrock. Given the limited availability of data from such systems, this work contributes an essential baseline for understanding biogeochemical dynamics in underrepresented subarctic environments. We do not claim to derive a universal framework for boreal rivers, but rather to highlight the unique functional traits of this specific landscape, which may differ significantly from rivers in other regions.
Specific comments:
Ln 6-9: What is the averaging period of these estimates?The values refer to the overall mean of all measurements from all campaigns in 2023 and 2024, and restate the numbers given in Tab. 2. We will clarify this.
Ln 11: The previous sentence says bank Reco and NEE are 5.8 and 2.2 times the water-air CO2 flux, respectively. As you have Reco and NEE, you can also report GPP data for the river banks both in the abstract and figure.
Bank fluxes were measured with two different systems, while one of them used exclusively opaque chambers, the other one used both transparent and opaque chambers. Reco fluxes were always measured and were pooled between both systems, while NEE was given based on measurements from one system. Therefore, the difference NEE-Reco would not represent GPP as it was measured. To avoid confusion, and given that it is not a central finding of the study, we decided to remove numbers for NEE and refrain from stating GPP values in the abstract.
Ln 14: How should I understand the magnitude here, ranging from 3.3 t C yr-1 to 70000 t C yr-1? Why is it a small carbon source? Could you provide the area-normalized flux values for these components?
We will remove “small” to avoid confusion at this point. For clarification, vertical emissions of the river and banks were small in comparison with the net emissions of the river (including lateral export). Taking together the vertical emission and lateral export, the carbon source is rather modest and we will clarify that. We will state the values of the net carbon balance of the river and its catchment in the following sentence, which will also clarify the magnitude.
Ln 15: Please include the NECB value in the abstract. I don’t see the dominant role of CO2 in NECB. Instead, DOC and bicarbonates were greater.
We realized that this was poorly phrased in the manuscript and will revise it. For clarification, CO2 plays a dominant role for the vertical river emissions, and DOC and bicarbonates in the net carbon budget of the river. We will include values of the net carbon balance of the river and its catchment.
Ln 87-91: The authors are trying to answer seven research questions in the study. However, these questions are verbal and redundant. Please summarize and reorganize them.
We will condense the research questions and revise the last paragraph of the introduction to:
“To provide emission estimates for a natural river, we conducted spatially distributed in situ measurements of vertical CO2 and CH4 fluxes and water chemistry across three seasons at Teno river (Finnish: Tenojoki, Norwegian: Tana, Northern Sámi: Deatnu; hereafter referred to as Teno river) and its adjacent banks. We characterized the spatial and seasonal variability in CO2 and CH4 fluxes and partial pressures, assessed the contribution of bank emissions to total riverine fluxes, and quantified the relative magnitude of emissions to the atmosphere, the lateral carbon export, and estimated a first-order catchment-scale carbon balance.
This study advances our understanding of how riverine and riparian components interact to shape landscape-scale carbon dynamics.”Ln 105: A parenthesis is missing here.
We will reduce the number of parentheses in this sentence to make it visually clearer.
Ln 226-227: The difference between NEE and Reco represents total CO2 uptake, not NET CO2 uptake.
We will revise this.
Ln 299-305: It is not clear to me how the catchment-level fluxes were upscaled from the site-level measurements. More detailed method description is required.
This section serves as an introduction of the following sections further describing how fluxes were upscaled. We will clarify this.
Ln 308: Please include the unit for each variable listed in the equation.
All variables have the unit tC yr-1. We will add this information in the revised manuscript.
Ln 393: You haven’t reported the CO2 fluxes of this study so far, thus making no sense to make this comparison to the Tibetan plateau yet.
We will remove the CO2 fluxes from this sentence.
Ln 383-410: This part repeats the information provided in Table 2, which reads a bit redundant and merely a comparison to the literature rather than an analytical discussion. It could be shortened.
We will shorten this part.
Ln 410-413: Discussion on how much CH4 fluxes attributed to bubbling should be added based on previous similar studies, if any.
There are few studies to date that give quantitative estimates of CH4 ebullition fluxes in boreal/subarctic rivers and streams, limited to a northeastern Siberian headwater catchment with high CH4 ebullition fluxes (Spawn et al., 2015; https://doi.org/10.5268/IW-5.3.845) and a regulated German river, where contributions of up to 80% were determined (McGinnis et al., 2016; https://doi.org/10.1021/acs.est.6b03268). However, these systems are likely not comparable to Teno river in terms of catchment activity and carbon content. We will add a discussion about this.
Fig 1: Please add the scale and north arrow on each map.
The first map showing Europe is displayed in a lat/lon projection, which distorts distances and directions. As such, a scale bar or north arrow would be inaccurate. For the second map however, we will reproject the data to a UTM projection which preserves local distances and directions, and we will add a scale bar and a north arrow to this map.
Table 3: How robust was it to use August measurements to estimate the annual export? References are required in the relevant method part.
Given limitations of our data, we chose a pragmatic approach estimating the first-order annual export based on measurements from only one campaign in August 2023. As the different carbon pools yielded different magnitudes, with DOC accounting for the largest component of the river carbon export in August 2023, we would expect similarly high magnitudes for the other seasons, although the absolute estimates would differ. Therefore, we will add error margins to the numbers in the revised manuscript to clarify temporal differences. Besides, we will revise the methods section of the lateral carbon export and add more references.
Citation: https://doi.org/10.5194/egusphere-2026-1260-AC1
-
AC1: 'Reply on RC1', Judith Vogt, 06 Aug 2026
-
RC2: 'Comment on egusphere-2026-1260', Anonymous Referee #2, 05 May 2026
Review report for Manuscript egusphere-2026-1260 by Vogt et al., entitled:
The carbon balance of a pristine subarctic river in northern Fennoscandia
The manuscript by Vogt et al. provides a valuable dataset on CO2 and CH4 fluxes and estimates the carbon balance for the pristine Teno river in northern Fennoscandia. Given that high-latitude, undisturbed riverine ecosystems are underrepresented in current literature, these baseline data are of clear interest. However, the narrative is overly descriptive, and there are critical methodological and conceptual issues that must be resolved.
Major Comments
- The manuscript attempts to quantify vertical fluxes, lateral carbon transport, and catchment-scale budgets simultaneously. Consequently, the analysis of each component remains superficial and observational. A rigorous scientific paper must move beyond merely reporting discrete numbers to elucidating the underlying environmental controls. I strongly suggest narrowing the scope of the manuscript. The authors should focus deeply on one core aspect and drastically expand the mechanistic analysis of that chosen component.
- Introduction: The literature review is too broad and disconnected from the specific research questions. Streamline this into a focused, hypothesis-driven narrative.
- Discussion: The discussion is overly descriptive. It needs a clear storyline that uses the data to directly address the research questions and extract core mechanistic insights.
- Using anchored chambers in flowing tributaries induces artificial turbulence, which artificially inflates gas exchange processes. The authors must rigorously justify this method, quantify the bias, or declare it as a major limitation.
- The authors state carbonates were neglected due to near-neutral pH, but later claim a high bicarbonate/carbonate load from silicate weathering. This contradiction needs to be resolved. If there is high alkalinity, CO2 results from the headspace equilibrium method should be explicitly corrected for the carbonate buffering effect (see Koschorreck et al., 2021).
- Relying on the HYDRORIVER dataset to delineate river lengths for stream orders 1-4 introduces high uncertainty. I suggest using direct remote sensing imagery to obtain accurate, concurrent river width and length data.
Minor Comments
- The Methods section reads like a field diary. Remove unnecessary logistical details and focus strictly on the procedures and parameters necessary for scientific reproducibility.
- Section 2.1.2 is excessively detailed. Condense this text into a single summary table detailing coordinates, Strahler order, dominant substrate, and surrounding land cover.
- For the lake-like system, the authors state that higher fluxes are driven by higher C mineralization. The authors can confirm this by tracing concentration data to assess the relative importance of gas exchange versus C concentrations.
Citation: https://doi.org/10.5194/egusphere-2026-1260-RC2 -
AC2: 'Reply on RC2', Judith Vogt, 06 Aug 2026
We thank the reviewer for their comments and appreciate their critical feedback on our manuscript. Below, we respond to the reviewer’s comments (restated in italics).
1: The manuscript attempts to quantify vertical fluxes, lateral carbon transport, and catchment-scale budgets simultaneously. Consequently, the analysis of each component remains superficial and observational. A rigorous scientific paper must move beyond merely reporting discrete numbers to elucidating the underlying environmental controls. I strongly suggest narrowing the scope of the manuscript. The authors should focus deeply on one core aspect and drastically expand the mechanistic analysis of that chosen component.
We appreciate the thoughtful comment regarding the breadth of our manuscript, and understand how incorporating vertical fluxes, lateral transport and catchment scale budgets leads to an overly descriptive manuscript that seems to gloss over the deeper processes of the ecosystem.
However, we respectfully argue for keeping all of these aspects in our manuscript since the key findings of this study are relationships between component fluxes, and how they contribute to the net carbon budget. The primary value of our study is the observation-driven attempt to highlight the key players and their relative magnitudes in the catchment-scale carbon budgets for a pristine high-latitude system. Omitting, for example, the lateral or catchment-scale part of the study would highlight the vertical chamber fluxes, which would fundamentally misrepresent how carbon moves through this subarctic landscape.2: Introduction: The literature review is too broad and disconnected from the specific research questions. Streamline this into a focused, hypothesis-driven narrative.
As detailed in our response to Reviewer 1 (regarding the research questions and the last paragraph of the introduction), we will condense the research questions and streamline the literature review accordingly.
3: Discussion: The discussion is overly descriptive. It needs a clear storyline that uses the data to directly address the research questions and extract core mechanistic insights.
As stated in our response to comment 3 of Reviewer 1, we will condense the research questions and revise the Discussion section by removing descriptive, site-by-site reporting. Instead, we will revisit the Discussion to make sure that the storyline is cohesive and focuses on the main aspects of this study outlined above.
4: Using anchored chambers in flowing tributaries induces artificial turbulence, which artificially inflates gas exchange processes. The authors must rigorously justify this method, quantify the bias, or declare it as a major limitation.
The anchored chambers were predominantly used to determine fluxes in tributaries. In these highly turbulent, rocky and fast moving waters of the tributaries, it was impossible to enter with a boat - as was done in (most of) the main channel in August 2023. We acknowledge that anchored chambers were shown previously to overestimate fluxes due to increased turbulence (Lorke et al., 2015; https://doi.org/10.5194/bg-12-7013-2015). The anchored chambers were therefore a compromise, and we can assure that measurements were conducted with caution to minimize disturbance to the water boundary layer. We will add an explanation about this in the manuscript to clearly state this limitation.
5: The authors state carbonates were neglected due to near-neutral pH, but later claim a high bicarbonate/carbonate load from silicate weathering. This contradiction needs to be resolved. If there is high alkalinity, CO2 results from the headspace equilibrium method should be explicitly corrected for the carbonate buffering effect (see Koschorreck et al., 2021).
Not applying the Koschorreck et al., 2021 correction was an oversight on our part, and for manually sampled pCO2 concentrations we will make the corrections in the revised manuscript. As the FaRAGE equilibrator is a continuous-loop system, this correction is not applicable for the continuous measurements of partial pressures, and the results from this approach will be left as-is.
6: Relying on the HYDRORIVER dataset to delineate river lengths for stream orders 1-4 introduces high uncertainty. I suggest using direct remote sensing imagery to obtain accurate, concurrent river width and length data.
We acknowledge that HydroRIVERS suffers from inaccuracy at latitudes >60°N and agree that HydroRIVERS can introduce localized errors in precise channel routing or length extraction. We will clarify this in the revised manuscript. However, we argue that relying exclusively on direct remote sensing imagery (e.g., Landsat or Sentinel-2) to delineate stream orders 1-4 may introduce even greater, systemic errors for two primary reasons:
- Lower-order streams (Strahler orders 1-3) of Teno river can be quite narrow and are frequently covered by riparian vegetation or topography. At a 10 or 30 m resolution, these vital headwater channels may be invisible to satellites, leading to an underestimation of the total water surface.
- Subarctic rivers, and especially unregulated ones, experience extreme hydrological volatility (e.g., the observed heavy rain event in August 2023 or the tributary desiccation in September 2024). Satellite image availability is severely limited by cloud cover at high latitudes. Calculating the area from arbitrary cloud-free summer snapshots would capture a single, biased water stage that may not represent the shifting surface areas active during high-flow or low-flow regimes.
- We used several geospatial datasets embedded in HydroSHEDS (HydroBASINS, HydroRIVERS and GloRiC) which are conveniently connected through river section identifiers. If we were to move away from using HydroRIVERS, and subdivide river reaches according to remote sensing products and determine their widths and lengths separately, we would not be able to reliably connect these river sections with the other products used for upscaling carbon fluxes.
Rather than swapping to an equally uncertain remote sensing snapshot, our study accounted for the inherent geometric limitations of HydroRIVERS by executing a propagated uncertainty framework (Monte Carlo simulations). In our upscaling equations, we did not treat the HydroRIVERS surface areas as fixed facts; instead, we perturbed the stream lengths and widths across a 10% randomized error distribution over 10,000 iterations. This directly integrated the geometric uncertainty into our final carbon flux confidence intervals.
Minor Comments
1: The Methods section reads like a field diary. Remove unnecessary logistical details and focus strictly on the procedures and parameters necessary for scientific reproducibility.
We will streamline the Methods section.
2: Section 2.1.2 is excessively detailed. Condense this text into a single summary table detailing coordinates, Strahler order, dominant substrate, and surrounding land cover.
We will condense the text into a table with the necessary information.
3: For the lake-like system, the authors state that higher fluxes are driven by higher C mineralization. The authors can confirm this by tracing concentration data to assess the relative importance of gas exchange versus C concentrations.
To assess whether elevated CO2 and CH4 fluxes at the lake-like river section were driven by increased carbon mineralization rather than enhanced gas exchange, we will add further discussion to support these results. We will evaluate the relationship between measured fluxes and dissolved gas concentrations using Spearman’s rank correlation. A significant positive relationship was observed between CO2 flux and pCO2 (ρ = 0.47, p < 0.01) and between CH4 flux and pCH4 (ρ = 0.57, p < 0.01). Therefore, the lake-like river section, despite lower flow velocities and higher water depth (indicating reduced gas transfer), exhibited the highest median pCO2 and pCH4 values and the highest water-air fluxes. These results indicate that elevated emissions are primarily driven by high dissolved carbon concentrations – likely resulting from enhanced mineralization and methanogenesis – rather than by increased gas exchange. We will also add a figure to visualize these findings.
Citation: https://doi.org/10.5194/egusphere-2026-1260-AC2
Data sets
Surface greenhouse gas fluxes, partial pressures and hydrochemistry at Teno river in northern Fennoscandia Judith Vogt, Joonatan Ala-Könni, Clara Mendoza-Lera, Taija Saarela, Niko Kinnunen, Wasi Hashmi, Ivan Mammarella, Anne Ojala, Carlos Palacin-Lizarbe, Jukka Pumpanen, Janne Rinne, Huizhong Zhang-Turpeinen, and Mathias Göckede https://doi.org/10.5281/zenodo.18632913
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 904 | 520 | 93 | 1,517 | 71 | 112 |
- HTML: 904
- PDF: 520
- XML: 93
- Total: 1,517
- BibTeX: 71
- EndNote: 112
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
The manuscript by Vogt et al. assesses the carbon balance of a pristine subarctic river in northern Fennoscandia based on three campaigns conducted in 2023 and 2024. The dataset collected from this pristine subarctic catchment is undoubtedly valuable. However, I have several major concerns about the current form of the manuscript. Please see below.
Specific comments:
Ln 6-9: What is the averaging period of these estimates?
Ln 11: The previous sentence says bank Reco and NEE are 5.8 and 2.2 times the water-air CO2 flux, respectively. As you have Reco and NEE, you can also report GPP data for the river banks both in the abstract and figure.
Ln 14: How should I understand the magnitude here, ranging from 3.3 t C yr-1 to 70000 t C yr-1? Why is it a small carbon source? Could you provide the area-normalized flux values for these components?
Ln 15: Please include the NECB value in the abstract. I don’t see the dominant role of CO2 in NECB. Instead, DOC and bicarbonates were greater.
Ln 87-91: The authors are trying to answer seven research questions in the study. However, these questions are verbal and redundant. Please summarize and reorganize them.
Ln 105: A parenthesis is missing here.
Ln 226-227: The difference between NEE and Reco represents total CO2 uptake, not NET CO2 uptake.
Ln 299-305: It is not clear to me how the catchment-level fluxes were upscaled from the site-level measurements. More detailed method description is required.
Ln 308: Please include the unit for each variable listed in the equation.
Ln 393: You haven’t reported the CO2 fluxes of this study so far, thus making no sense to make this comparison to the Tibetan plateau yet.
Ln 383-410: This part repeats the information provided in Table 2, which reads a bit redundant and merely a comparison to the literature rather than an analytical discussion. It could be shortened.
Ln 410-413: Discussion on how much CH4 fluxes attributed to bubbling should be added based on previous similar studies, if any.
Fig 1: Please add the scale and north arrow on each map.
Table 3: How robust was it to use August measurements to estimate the annual export? References are required in the relevant method part.