the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 12 Mar 2026
TundraFlux: A database of ecosystem respiration with biotic and abiotic metadata from Arctic and alpine tundra warming experiments
Abstract. Empirical in-situ measurements of ecosystem carbon dioxide respiration (Reco) in high-latitude ecosystems remain limited, yet they are essential for understanding how tundra carbon cycling responds to climate warming across different environmental contexts and for reducing uncertainties in upscaled carbon budgets and carbon–climate feedbacks. Here, we present the TundraFlux Database, which to date is the most comprehensive synthesis of tundra Reco responses to experimental warming. The database compiles over 24,000 daily-aggregated in-situ Reco measurements from control and plots warmed with open-top chambers at 64 Arctic and alpine tundra sites across 12 countries. By coupling Reco measurements with extensive metadata on climate, vegetation, and soil characteristics, the TundraFlux Database enables the integration of field-scale ecological processes into large-scale models, offering new opportunities to refine global carbon budgets and test predictions of tundra ecosystem responses to warming. Open access to the TundraFlux Database empowers the research community to better quantify and predict how warming alters carbon cycling in Arctic and alpine tundra ecosystems.
- Preprint
(1529 KB) - Metadata XML
-
Supplement
(866 KB) - BibTeX
- EndNote
Status: final response (author comments only)
- RC1: 'Comment on essd-2025-837', Anonymous Referee #1, 13 May 2026
-
RC2: 'Comment on essd-2025-837', Anonymous Referee #2, 19 May 2026
The authors present a commendable collaborative effort in integrating daily aggregated in-situ Reco measurements from air warming experiments consisting of control plots and warmed plots with open top chambers at 64 Arctic and alpine tundra sites across 12 countries. The dataset spans from 2000 to 2024 and has decent coverage of circumpolar Arctic ecosystems, making it a robust resource for spatial and temporal studies. The dataset has strong potential to contribute to collaboration across nations and research teams in current and emerging studies, allowing for reliable model comparison and leading to a scalable synthesis of Arctic warming and responses in Reco, and becoming a useful resource to further our understanding and identify knowledge gaps of Arctic ecosystems and biogeochemical cycling. Therefore, I think this dataset is fitting for Earth System Science Data and is a meaningful contribution to the literature. However, I found that the future directions of the manuscript were lacking, and I think the authors could be more explicit in mentioning their next steps, as well as mentioning how they will streamline the process of maintaining the database for future data integration and cross-collaboration.
Overall, the manuscript was easy to follow and is well-written. I found that I was able to have a solid understanding of what the database is and its utility. Accessing the database took me some time, but once I had downloaded all the files from Zenodo and had tinkered with them in R, I was able to view them. I appreciate the authors’ organization of the datasets and files, as well as the detailed information provided in the supplementary materials. I think TundraFlux is an impressive and valuable resource, and I appreciate it being an open-access dataset.
Below, I have made some line-by-line suggestions for the authors to consider:
Supplementary:
S1: I really liked the inclusion of the start year for each of the sites for the warming experiments. I think it would be helpful to include a column for the end year as well here, or within the same column for the start year, and change it to provide the beginning and end year (e.g., Start – End warming)
ITEX_biomass_method_protocol: The inclusion of this document is a nice addition and a helpful resource for the database. However, it looks like the links included in the document do not work for reviewer access and received a “404 file not found” error.
Main Text
- Introduction
Line 119: Might want to consider including a definition for permafrost (e.g., soil that remains consecutively frozen for at least 2 years)
- Description and structure
Lines 158-164: adding a line or two in this paragraph to mention that warming alters tundra ecosystems in the increasing frequency of thermokarst activity (geomorphology impacts), which may also contribute to Reco. Some references below:
https://www.nature.com/articles/ncomms13043
https://www.nature.com/articles/s41467-019-09314-7
https://ecoss.nau.edu/wp-content/uploads/2016/05/Vogel_et_al-2009-Journal_of_Geophysical_Research-_Solid_Earth_1978-2012.pdf
https://www.researchgate.net/publication/281261197_Permafrost_collapse_alters_soil_carbon_stocks_respiration_CH4_and_N2O_in_upland_tundra
Line 210: This could be a user error on my end, but when opening the Tundra_flux_daily_v1 in my version of R, I see 76 variables instead of 74. Could it be a typo here?
Lines 258-259: You may want to double-check to make sure the DOI works for others to access Zenodo. Additionally, having a direct link here to the dataset would be helpful to readers. Maybe this can be included in the DOI display name?
- Applications of the TundraFlux database
Lines 274-275: I appreciate the authors' efforts in including shoulder season observations in the database.
- Future Directions
Lines 369-379: Would including other variables be considered for future directions in addition to what is already included in the database? For example, the tundra type or landform (thaw slumps, peatlands, coastal, etc.) for each of the sites would be interesting and helpful for future studies. Additionally, mentioning other potential methods that can be incorporated into, for example, the integration of field-based observations and remotely sensed observations?
-
EC1: 'Comment on essd-2025-837', Tobias Gerken, 30 May 2026
Dear authors.
Based on the helpful and constructive reviewer comments, I am inviting you to provide a response. I feel comments about the database accessibility and network integration are particularly important to ensure usability and accessibility of the manuscript. Please make sure to provide careful consideration to these.
Thank you very much.
Tobias Gerken
Citation: https://doi.org/10.5194/essd-2025-837-EC1
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 456 | 263 | 24 | 743 | 37 | 15 | 29 |
- HTML: 456
- PDF: 263
- XML: 24
- Total: 743
- Supplement: 37
- BibTeX: 15
- EndNote: 29
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
This manuscript describes the aggregation of ecosystem respiration data from air warming experiments across arctic and alpine tundra warming manipulations. Efforts to aggregate data from similar groups of experiments run by PIs distributed across institutions and nations is extremely valuable to facilitate synthesis, model comparison, and identify gaps. Thus, I think this database is a valuable contribution to the literature. There are several other chamber-based flux aggregation products which the authors mention in their next steps as a goal to facilitate integration. I believe it would be worth explicitly mentioning and distinguishing these in the introduction or description and structure sections. A few aggregated datasets of which I am aware are ABCFlux (ambient only), COSORE (global, focused on continuous, not necessarily aggregated for experimental manipulations), and the ITEX Ecosystem carbon flux dataset. Additonally, are all the sites in TrundraFlux listed in Vogt et al 2025. ARGO: ARctic greenhouse Gas Observation metadata version 1? If not, adding entries there would be a good way to build on existing resources and reduce diverging aggregation efforts.
Overall, this paper is well written with informative graphics and descriptions of the data structure. I feel I have a good grasp of what I expect to find in the dataset and how it is organized.
Some detailed, line-by-line comments
Data sources:
Table S1: do some of these come from databases (eg: LTER for CiPEHR and Toolik sites)? It would be helpful to add these links to the table as well. I think it would be additionally helpful to identify which of these sites originates from the establishment of ITEX.
Table S2: not sure how useful Nobs is other than to show generally large vs small datasets. Instead perhaps adding the frequency/interval of measurement or mode of measurement such as manual vs automated.
Table S4: Is the Daily_% the percent of days in a year with measurements? So 0.5% would be and average of 1.8 days of measurement over the years measured?
Use country codes that match a standard. I believe Fluxnet currently uses: ISO-3611-2: https://fluxnet.org/data/fluxnet2015-dataset/known-issues/. I think using standardized names reduces guessing and makes it easier to link with other products.
Reviewer access URL for code, readme, data dictionary did not work (page not found error from Zenodo)
Text:
Line 128: Suggestion not starting the paragraph with ‘its’. Instead something like: ‘Ecosystem respiration plays a central role… making it important to predict… ‘
Line 135: It would be useful to briefly describe the OTC method.
Line 169-171: Are these metadata time-varying if the experiment is multiple years long? Or are they site-specific and temporally fixed?
Line 232-235: Restricting the NEE fluxes to night-time only is a reasonable approach. However, I wonder what the data source was for the CiPEHR data because Reco (extrapolated from night to day using a temperature-response function) is included in the datasets in the BNZ LTER data catalog: https://www.lter.uaf.edu/data/data-detail/id/481.
There are also some respiration measurements during and from the end of the growing season. Not sure if they are included?
https://www.lter.uaf.edu/data/data-detail/id/611
https://www.lter.uaf.edu/data/data-detail/id/572
https://www.lter.uaf.edu/data/data-detail/id/652
Line 262: I think this sentence has a word missing
Line 288: perhaps cite Schadel et al 2018 Divergent patterns of experimental and model-derived permafrost ecosystem carbon dynamics in response to Arctic warming as an example?
Line 295: Remove ‘while’ to start with The TundraFlux Database? I think it’s important to point out, as you do, that the imbalance is a reflection of the field rather than a limitation of the TundraFlux data aggregation effort. In a sense, these aggregation efforts allow us to see the imbalance more clearly.
Line 372: and/or air vs soil warming as the CiPEHR experiments show that the effects of snow-fences carry over into the summer because the warming soil profile is cumulative. And air warming in summer can alter winter conditions as shifts in vegetation can alter soil conditions (eg: Heather Kropp et al 2021 Environ. Res. Lett.)