the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 19 Jun 2023
Nitrous oxide emissions from pan-Arctic terrestrial ecosystems: A process-based biogeochemistry model analysis from 1969 to 2019
Abstract. Nitrous oxide (N2O) is a potent greenhouse gas with radiative forcing 265–298 times stronger than that of carbon dioxide (CO2). Increasing atmospheric N2O burden also contributes to stratospheric ozone depletion. Recent field studies show N2O emissions from the Arctic ecosystems have increased due to warming. To date, the emissions across space and time have not been adequately quantified. Here we revised an extant process-based biogeochemistry model, the Terrestrial Ecosystem Model (TEM) to incorporate more detailed processes of soil biogeochemical nitrogen (N) cycle, permafrost thawing effects, and atmospheric N2O uptake in soils. The model is then used to analyze N2O emissions from pan-Arctic terrestrial ecosystems. We find that both regional N2O production and net emissions increased from 1969 to 2019, with production ranging from 1.2–1.3 Tg N yr-1 and net emissions from 1.1–1.2 Tg N yr-1 considering the permafrost thaw effects. Soil N2O uptake from the atmosphere was 0.1 Tg N yr-1 with a small interannual variability. Atmospheric N deposition significantly increased N2O emission by 31.5 ± 3.1 %. Spatially, terrestrial ecosystems act as net sources or sinks ranging from -12 to 700 mg N m-2 yr-1 depending on temperature, precipitation, soil characteristics, and vegetation types in the region.
- Preprint
(2245 KB) - Metadata XML
- BibTeX
- EndNote
Status: closed
-
RC1: 'Comment on egusphere-2023-1047', Anonymous Referee #1, 21 Jul 2023
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-1047/egusphere-2023-1047-RC1-supplement.pdf
- AC1: 'Reply on RC1', Ye Yuan, 10 Sep 2023
-
RC2: 'Comment on egusphere-2023-1047', Anonymous Referee #2, 23 Aug 2023
General comments:
This is a very coarse estimation of N2O in the pan-arctic region with a lot of issues. The effort is laudable and I also want to highlight that it is one of the first attempts to model N2O budget of the Arctic, but some of the issues really need to be addressed before this MS can be published.
- The author should clearly define the extent of the pan-arctic. In Table 1, the authors showed the sites for N2O emission observations that are used for model parameterization and verification. Which sites belong to the pan-arctic? Does France belong to the pan-arctic? Boreal Forest? Tibetan Plateau? As scientists, the authors should be careful here. Please clearly define pan arctic and remove the data that do not belong to pan arctic. I also suggest to use more data from real arctic sites (currently only 4) which are available in the literature for model calibration.
- The authors should please provide detailed description of how they collected and cleaned the site observation data in Table 1 that are used for model parameterization and verification and make the data available, so that the reviewers and readers can check if everything that the authors did were correct.
- Model parameterization and verification are two steps. What part of the site data were used for model parameterization and what part were used for verification? Looks the authors do not separate them (Figure 2).
- I would like that the authors further test the performance of the model in simulating N2O. The co-author here Narasinha Shurpali has N2O data from eddy covariance. Please use this continuous eddy covariance data to further test the performance of your model. If the model works well, the reviewers and readers will then have more confidence in your model.
- Peatlands are an important part of the high-latitude region and the highest emitters of N2O among natural arctic ecosystems. But, peat is not considered in the model. How does your model represent N2O fluxes from peatlands?
- the discussion is not well organised. while the main point, modelled N2O emission, receives far too little attention and is poorly referenced, the authors ramble and repeat themselves about N2O uptake and N deposition, which is interesting, but has relatively minor importance here. please re-organize and elaborate on the N2O budget and emissions and the microbial pathways (nitrification, denitrification). also, some assumptions can be made for future emissions based on the sensitivity analysis made with variable temp and preciptation.
- Spin up is not described in the method. Please describe it. This is just one case. The authors should describe all the steps that you did in the method in detail. Please check if anything else is missing and describe the missing steps.
- all N2O uptake data should have a negative sign.
Specific comments:
Lines 44-45: double check if all citations are listed in the reference section. I could not find e.g. the reference for Vigot et al. 2017
Lines 55-56: this is incorrect. Please rephrase to make this sentence read “Soil moisture controls the oxygen availability for microbes and thus nitrification and denitrification rates.” Add to the next sentence that denitrification operates under high water content and nitrification under low water content, and that generally denitrification produces much more N2O.
Lines 70-76: what about the model QUNICY? this model has been calibrated for permafrost regions and N2O emissions have been simulated. Make sure to appropriately cite this reference in your MS
Lacroix, S. Zaehle, S. Caldararu, J. Schaller, P. Stimmler, D. Holl, et al.
Glob Chang Biol 2022 Vol. 28 Issue 20 Pages 5973-5990
Line 89: ‘Here we revised the N cycling algorithms in TEM by incorporating the loss of nitrogen through gas emissions’. Looks this is not correct. Yu, 2016 and Yu and Zhuang, 2019 already considered the loss of nitrogen through gas emissions (Quantifying global N2O emissions from natural ecosystem soils using trait-based biogeochemistry models).
Line 91: ‘and additional inputs of organic nitrogen and carbon resulting from permafrost thawing (Fig. 1)’. This point is not shown in Fig. 1.
Figure 1: ‘the difference between mineralization (organic N mineralized to inorganic N) and mobilization (inorganic N to organic N)’. First, it should be immobilization instead of mobilization. In addition, in the figure, you just show the immobilization of NH4+. I would like to know, if the immobilization of NO3- is considered in the model? or only the immobilization of NH4- is considered? In addition, I would like to know, if the model can consider the DNRA process and the heterotrophic nitrification process?
Lines 104-106: where did the authors get data on soil gas concentration of N2O from artic sites to estimate N2O uptake? i do not find these data in the MS. please display.
Line 113: What kind of observational data is used?
Line 114: what is Nip?
Line 135: consider using the word “depth” instead of “deep”.
Lines 135-138: Two bold assumptions are evident here. Firstly, it's widely acknowledged that carbon (C) and nitrogen (N) content tends to decrease with increasing profile depth, a fact well-documented in the literature. Secondly, the established understanding is that the main reservoir of C and N is relatively stable, indicating its slow decomposition rate. Consequently, the assumption that half of the nitrogen pool undergoes mineralization within a few years appears, from my perspective, to be an overestimation. Nevertheless, if better data cannot be acquired, I suggest that the authors, at the very least, account for the uncertainties linked to these assumptions in their model (lines 177-180).
Line 140-143: as mentioned in the general comments, I am curious about the rationale behind incorporating numerous boreal forests and sites from France in model calibration and parameterization, given that these sites are situated well beyond the the pan-Arctic zone. Several of these forests lack permafrost entirely and, as such, should ideally be excluded from the analysis. Furthermore, to align with the manuscript's title appropriately, alpine sites should also be omitted. This revision would result in merely four wet tundra sites available for model calibration and parameterization, a quantity that is evidently insufficient. It would be more appropriate for the authors to use additional published N2O emissions data from Arctic sites for model calibration and parameterization. I point towards the work of Voigt et al. (2020) for these data.
- Voigt, L. van Delden, M. E. Marushchak, C. Biasi, B. W. Abbott, B. Elberling, et al.
Distributor: PANGAEA 2020
DOI: 10.1594/PANGAEA.919217
Table 1: I am not aware of all the sites, but double check the classification. At least the Russian site in Voigt et al. (2017) is not a wet tundra. It is a permafrost peatland (dry, raised peatland) and an upland mineral soil. Maybe use simply the term "Arctic Tundra”? Again, please clearly define pan-arctic and remove the data that do not belong to pan arctic from the Table, model analysis and runs and MS as such.
Table 2: please provide the full name of each parameter
Line/Chaper 167: Again: the author should clearly define the extent of the pan-arctic. is the boreal region inlcuded? alpine? Tibetan Plateau only? the permafrost region of the northern latitudes? if the latter is true, the title needs to be changed.
Line 197: there is a typo in “pan-arctic”
Line 227/Chapter 3.3.
can the authors elaborate a bit on why you did not find large effects of precipitation changes on N2O emissions? that is a bit unexpected, given that soil moisture is primary control of N2O production and consumption. could it be that precipitation changes do not translate into soil moisture changes? the authors mention that there were large regional changes in nitrification rates (what about denitrification, by the way?), but i do not get how the total N2O production from the different microbial pathways then does not change.
Line 240/Discussion
the discussion is absolutely not in line with the results. the authors discuss N2O uptake, N deposition while the modelling work focuses on N2O emissions and the effect of permafrost thaw.
First off, the discussion should start with a chapter on N2O emissions (4.1.) and not with the role of N2O uptake in the net emissions. this process has relatively minor importance here (about 10% of the emissions, as the authors state themselves), and this discussion on N2o uptake should be considerably toned down (e.g. completely delete lines 277-288, not focus of the MS). the discussion on the result of N2O emissions is generally missing or far too short (lines 335-347), and needs to be stronger in the MS (again at the beginning of the discussion, 4.1., at least one page, while reduce the discussion about N2O reduction by more than half). also, compare your results much more with all the data available from the literature. It should be stressed that the estimate of the pan-arctic N2O budget the authors provide is one of the largest currently reported in the literature.
Second, the whole discussion about N deposition comes out of a blue. this topic should be part of the introduction, methods and results, only then it can be discussed.
Citation: https://doi.org/10.5194/egusphere-2023-1047-RC2 - AC2: 'Reply on RC2', Ye Yuan, 10 Sep 2023
Status: closed
-
RC1: 'Comment on egusphere-2023-1047', Anonymous Referee #1, 21 Jul 2023
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-1047/egusphere-2023-1047-RC1-supplement.pdf
- AC1: 'Reply on RC1', Ye Yuan, 10 Sep 2023
-
RC2: 'Comment on egusphere-2023-1047', Anonymous Referee #2, 23 Aug 2023
General comments:
This is a very coarse estimation of N2O in the pan-arctic region with a lot of issues. The effort is laudable and I also want to highlight that it is one of the first attempts to model N2O budget of the Arctic, but some of the issues really need to be addressed before this MS can be published.
- The author should clearly define the extent of the pan-arctic. In Table 1, the authors showed the sites for N2O emission observations that are used for model parameterization and verification. Which sites belong to the pan-arctic? Does France belong to the pan-arctic? Boreal Forest? Tibetan Plateau? As scientists, the authors should be careful here. Please clearly define pan arctic and remove the data that do not belong to pan arctic. I also suggest to use more data from real arctic sites (currently only 4) which are available in the literature for model calibration.
- The authors should please provide detailed description of how they collected and cleaned the site observation data in Table 1 that are used for model parameterization and verification and make the data available, so that the reviewers and readers can check if everything that the authors did were correct.
- Model parameterization and verification are two steps. What part of the site data were used for model parameterization and what part were used for verification? Looks the authors do not separate them (Figure 2).
- I would like that the authors further test the performance of the model in simulating N2O. The co-author here Narasinha Shurpali has N2O data from eddy covariance. Please use this continuous eddy covariance data to further test the performance of your model. If the model works well, the reviewers and readers will then have more confidence in your model.
- Peatlands are an important part of the high-latitude region and the highest emitters of N2O among natural arctic ecosystems. But, peat is not considered in the model. How does your model represent N2O fluxes from peatlands?
- the discussion is not well organised. while the main point, modelled N2O emission, receives far too little attention and is poorly referenced, the authors ramble and repeat themselves about N2O uptake and N deposition, which is interesting, but has relatively minor importance here. please re-organize and elaborate on the N2O budget and emissions and the microbial pathways (nitrification, denitrification). also, some assumptions can be made for future emissions based on the sensitivity analysis made with variable temp and preciptation.
- Spin up is not described in the method. Please describe it. This is just one case. The authors should describe all the steps that you did in the method in detail. Please check if anything else is missing and describe the missing steps.
- all N2O uptake data should have a negative sign.
Specific comments:
Lines 44-45: double check if all citations are listed in the reference section. I could not find e.g. the reference for Vigot et al. 2017
Lines 55-56: this is incorrect. Please rephrase to make this sentence read “Soil moisture controls the oxygen availability for microbes and thus nitrification and denitrification rates.” Add to the next sentence that denitrification operates under high water content and nitrification under low water content, and that generally denitrification produces much more N2O.
Lines 70-76: what about the model QUNICY? this model has been calibrated for permafrost regions and N2O emissions have been simulated. Make sure to appropriately cite this reference in your MS
Lacroix, S. Zaehle, S. Caldararu, J. Schaller, P. Stimmler, D. Holl, et al.
Glob Chang Biol 2022 Vol. 28 Issue 20 Pages 5973-5990
Line 89: ‘Here we revised the N cycling algorithms in TEM by incorporating the loss of nitrogen through gas emissions’. Looks this is not correct. Yu, 2016 and Yu and Zhuang, 2019 already considered the loss of nitrogen through gas emissions (Quantifying global N2O emissions from natural ecosystem soils using trait-based biogeochemistry models).
Line 91: ‘and additional inputs of organic nitrogen and carbon resulting from permafrost thawing (Fig. 1)’. This point is not shown in Fig. 1.
Figure 1: ‘the difference between mineralization (organic N mineralized to inorganic N) and mobilization (inorganic N to organic N)’. First, it should be immobilization instead of mobilization. In addition, in the figure, you just show the immobilization of NH4+. I would like to know, if the immobilization of NO3- is considered in the model? or only the immobilization of NH4- is considered? In addition, I would like to know, if the model can consider the DNRA process and the heterotrophic nitrification process?
Lines 104-106: where did the authors get data on soil gas concentration of N2O from artic sites to estimate N2O uptake? i do not find these data in the MS. please display.
Line 113: What kind of observational data is used?
Line 114: what is Nip?
Line 135: consider using the word “depth” instead of “deep”.
Lines 135-138: Two bold assumptions are evident here. Firstly, it's widely acknowledged that carbon (C) and nitrogen (N) content tends to decrease with increasing profile depth, a fact well-documented in the literature. Secondly, the established understanding is that the main reservoir of C and N is relatively stable, indicating its slow decomposition rate. Consequently, the assumption that half of the nitrogen pool undergoes mineralization within a few years appears, from my perspective, to be an overestimation. Nevertheless, if better data cannot be acquired, I suggest that the authors, at the very least, account for the uncertainties linked to these assumptions in their model (lines 177-180).
Line 140-143: as mentioned in the general comments, I am curious about the rationale behind incorporating numerous boreal forests and sites from France in model calibration and parameterization, given that these sites are situated well beyond the the pan-Arctic zone. Several of these forests lack permafrost entirely and, as such, should ideally be excluded from the analysis. Furthermore, to align with the manuscript's title appropriately, alpine sites should also be omitted. This revision would result in merely four wet tundra sites available for model calibration and parameterization, a quantity that is evidently insufficient. It would be more appropriate for the authors to use additional published N2O emissions data from Arctic sites for model calibration and parameterization. I point towards the work of Voigt et al. (2020) for these data.
- Voigt, L. van Delden, M. E. Marushchak, C. Biasi, B. W. Abbott, B. Elberling, et al.
Distributor: PANGAEA 2020
DOI: 10.1594/PANGAEA.919217
Table 1: I am not aware of all the sites, but double check the classification. At least the Russian site in Voigt et al. (2017) is not a wet tundra. It is a permafrost peatland (dry, raised peatland) and an upland mineral soil. Maybe use simply the term "Arctic Tundra”? Again, please clearly define pan-arctic and remove the data that do not belong to pan arctic from the Table, model analysis and runs and MS as such.
Table 2: please provide the full name of each parameter
Line/Chaper 167: Again: the author should clearly define the extent of the pan-arctic. is the boreal region inlcuded? alpine? Tibetan Plateau only? the permafrost region of the northern latitudes? if the latter is true, the title needs to be changed.
Line 197: there is a typo in “pan-arctic”
Line 227/Chapter 3.3.
can the authors elaborate a bit on why you did not find large effects of precipitation changes on N2O emissions? that is a bit unexpected, given that soil moisture is primary control of N2O production and consumption. could it be that precipitation changes do not translate into soil moisture changes? the authors mention that there were large regional changes in nitrification rates (what about denitrification, by the way?), but i do not get how the total N2O production from the different microbial pathways then does not change.
Line 240/Discussion
the discussion is absolutely not in line with the results. the authors discuss N2O uptake, N deposition while the modelling work focuses on N2O emissions and the effect of permafrost thaw.
First off, the discussion should start with a chapter on N2O emissions (4.1.) and not with the role of N2O uptake in the net emissions. this process has relatively minor importance here (about 10% of the emissions, as the authors state themselves), and this discussion on N2o uptake should be considerably toned down (e.g. completely delete lines 277-288, not focus of the MS). the discussion on the result of N2O emissions is generally missing or far too short (lines 335-347), and needs to be stronger in the MS (again at the beginning of the discussion, 4.1., at least one page, while reduce the discussion about N2O reduction by more than half). also, compare your results much more with all the data available from the literature. It should be stressed that the estimate of the pan-arctic N2O budget the authors provide is one of the largest currently reported in the literature.
Second, the whole discussion about N deposition comes out of a blue. this topic should be part of the introduction, methods and results, only then it can be discussed.
Citation: https://doi.org/10.5194/egusphere-2023-1047-RC2 - AC2: 'Reply on RC2', Ye Yuan, 10 Sep 2023
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 505 | 225 | 36 | 766 | 38 | 35 |
- HTML: 505
- PDF: 225
- XML: 36
- Total: 766
- BibTeX: 38
- EndNote: 35
Viewed (geographical distribution)
| Country | # | Views | % |
|---|---|---|---|
| United States of America | 1 | 268 | 35 |
| China | 2 | 69 | 9 |
| Germany | 3 | 67 | 8 |
| United Kingdom | 4 | 63 | 8 |
| Finland | 5 | 31 | 4 |
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
- 268