the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 04 Mar 2025
Assessing Carbon Flux Variability in an Alpine Steppe: Insights from Dual-Height Measurements
Abstract. Future projections of climate warming on the Tibetan plateau (TP) imply a 4 °C warming in the next 100 years, the largest in the middle of the troposphere. Climatic variabilities of this magnitude are likely to trigger a cascade of climate-carbon (C) feedbacks within the TP ecosystems. However, a robust consensus of the feedback mechanisms and their drivers is scarce due to a lack of observations and unaccounted spatial heterogeneity. In the present study, we investigated how coarse-scale heterogeneity impacts the CO2 fluxes using a dual eddy covariance tower system (3 m and 19 m) over an alpine steppe ecosystem near the Nam Co Station for Multi-sphere Observation and Research (NAMORS) on the central TP. The source area of the 3 m height is relatively homogenous. On the other hand, the source area of the 19 m height covers the steppe and part of the neighboring lake. The steppe acted as carbon neutral over the 10-month measurement period (August 2018–May 2019) at the 19 m footprint as opposed to the 10-month long-term (2006–2018) average (-78 g C m-2) observed at the 3 m footprint. We found that the difference in the magnitude of CO2 fluxes observed from the two towers was attributed to the combined effects of winter snow cover and lake-land interactions. The extreme snow accumulation over the period increased the ecosystem respiration thus elevating the emissions in winter, highlighting the role of extreme snow events in regulating carbon dynamics in high-altitude ecosystems. Additionally, the neighboring lake substantially influenced carbon fluxes over larger spatial footprints, serving as a natural buffer that mitigates land carbon emissions during critical periods. Fluxes measured from land-dominated areas at both tower heights were largely consistent, demonstrating the reliability of steppe-derived flux measurements across 3 m and 19 m footprints. The findings emphasize the critical need for adopting a landscape-scale perspective to better capture flux variability in heterogeneous environments.
- Preprint
(1196 KB) - Metadata XML
- BibTeX
- EndNote
Status: closed
-
RC1: 'Comment on egusphere-2025-530', Anonymous Referee #1, 19 Mar 2025
-
AC1: 'Reply on RC1', Nithin dinesan Pillai, 11 Jul 2025
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-530/egusphere-2025-530-AC1-supplement.pdf
-
AC1: 'Reply on RC1', Nithin dinesan Pillai, 11 Jul 2025
-
RC2: 'Comment on egusphere-2025-530', Anonymous Referee #2, 20 Jun 2025
Within the manuscript titled “Assessing Carbon Flux Variability in an Alpine Steppe: Insights from Dual-Height Measurements” the authors compare flux measurements of CO2 from two eddy covariance (EC) systems mounted at different heights, at Nam Co Station for Multisphere Observation and Research (NAMORS) on the Tibetan Plateau (TP).
Their goal of this approach is to investigate the impact of coarse-scale heterogeneity on the CO2 fluxes measured at NAMORS. For a more detailed analysis of the influence by the underlying surface the authors used a footprint model to estimate the source areas of each EC system and accounted the contribution of different land-use classes to the measured fluxes. Further they used remote sensing products to distinguish difference in snow cover within the footprint area of the two systems during winter month.
The systems were deployed at 3m, the standard observation height for the long-term monitoring system deployed by the Institute of Tibetan Plateau Research, and 19m on the boundary layer tower. While for the long-term monitoring station data is available for the period 2006-2019, the 19m system only provided data from August 2018 till End of May 2019.
Considering the heterogeneity of vegetation cover and the scarcity of reliable measurement stations on the Tibetan Plateau this study sets out to investigate a very important aspect of CO2 exchange estimates for the region. By choosing a height of 19m above ground the footprint of the higher system includes alpine steppe with different vegetation cover and the nearby shallow lake adjacent to Nam Co Lake. The footprint area of the 3m system is composed of alpine steppe.
They see difference in the net ecosystem exchange (NEE) between the fluxes derived from the different systems. The conclude that these differences originate from differences in the vegetation and the lake contribute as well as differences in snow cover during winter month.
Overall, the conducted study shows some potential by addressing an important topic. In general, the background information given in the introduction show that the authors understand the theory behind the applied methods, however the presentation of the results and the structure of the manuscript make it hard to follow the aim of the study and the reasoning of the authors. Due to the restrictions of the data availability, as pointed out by the authors themself, the study falls somewhat short of reaching the aim the authors set out. While the language is clear and the miss match of information in different sections as well as the order of presented results make it rather cumbersome to read. Due to some doubts about the presented method and unclear presentation results, I can not recommend this manuscript to be published in its current state. Overall, I think this is an interesting topic and the authors have presented in parts some interesting results, however in my opinion this manuscript needs a major revision, and potentially additional data should be included.
Dear Authors,
thank you for this manuscript about EC flux measurements on the TP. I think that you are looking at a very interesting topic. While I find the topic to be important and the idea of this study intriguing, I have quite a lot of questions and comments after reading your manuscript. In general I find the manuscript well written and easy to read from a language point of view, figures are all understandable and figure captions sufficient. However I find the structure of the manuscript at times hard to follow.
One of my major points about this manuscript is that the information about available and used data is not consistent. There are miss matches in the description of the data availability of the two systems. The data availability of the 3m system is stated as 2006-2019, with gaps 2012-2015 and 2018-2019, while in Figure 2 it is stated that data from 2006-2007 and 2016-2017 is used for the comparison. This makes it unnecessarily confusing for the reader. Could you please look through the manuscript and fix those mismatches. An easier understandable presentation of the data availability would also help the reader to follow. Further you state that the overlapping time of the two systems was from May 13th to June 3rd 2019 when talking about the processing of the data from the 3m system. However, when talking about data from the 19m system you state that data was analysed from August 01st 2018 to May 31st 2019. This would only leave an overlap of data from May 13th till 31st. While you treat the long-term data set more as a reference data set, I still find the comparison period of less than 3 weeks too short. This is particularly true when comparing the fluxes monthly, especially at the early stage of the vegetation period, before onset of monsoon when fluxes are rather small and uncertainties high. I see a vast improvement if the dataset could be extended to include more overlapping data from both systems.
I like the idea of a more ecosystem and landscape flux approach, which would make it easier to compare the resulting fluxes to satellite products. However, I have some questions and concerns about the method you used.
You explain the treatment of data with contribution from the lake and how a land only and land lake flux dataset was derived from your overall dataset. Given the situation this is interesting and in principle a valid concept. So, from a general data handling perspective I have no issues with this approach, but this leads to some questions that arise about the methods applied.
Why did you choose this the specific height and was the lake contribution was specifically targeted? This does not become clear when reading the manuscript.
Do I understand it correctly that the lake contribution was the difference of the total NEE and the pure land NEE? Considering that it is very common to have a strong land-sea breeze situation at Nam Co with quite distinct times when the EC system sees a lake contribution in the fluxes, does this lead to a large uncertainty the lake contribution?
The estimation of the monthly NEE budget of the 19 m footprint area for land-only and land-lake systems also raises some questions. It might be my misunderstanding of the described method, so could you please explain how this was calculated in more detail? How were lake contributions to the total NEE considered for periods the EC system measured only land fluxes?
I think I can follow the classification into land-land and lake-land due to the contribution derived from the footprint analysis. You describe that you created the land-land and land-lake system dataset by including fluxes form the specific surfaces according to the overall contribution of the flux from this surface to the overall flux. In my understanding this was done for each of the 8 wind sectors you defined and then a composite flux was estimated according to the land use contribution derived from the footprint model. Then monthly budgets were calculated. This is the point I must admit that I can’t follow how this was done. Could you please elaborate more on how it is done? How were fluxes gap filled/modelled for specific land cover for periods when there were no contributions from this direction. In line 240 you describe that that the portioning was performed after gapfilling. Does this mean lake fluxes were gapfilled as if they were land fluxes?
You go on to state that the datasets might have missing half hour value. Can you please explain? Does this mean each dataset was only sorted according to the classification but no continuous dataset for each classification was derived? Are the monthly budgets, growing season and non growing season then calculated on unequal amount of data? Or did you include the relative contribution of the lake fluxes into each half hour based on your footprint calculations? This is a point that needs to be clarified for future publication. If it is rather my lack of understanding, please consider the way how you describe this rather complex but very important part of the data handling.
The subsequent comparison of the two datasets is based on a dataset of an averaged ten-month period resembling the month the 19m system was deployed. While I see some value in this approach, I don’t see how this can lead to the rather in detail comparison the authors present. Considering the rather rapid trend in warming on the TP, coinciding with shifts in precipitations pattern, available soil moisture and potential vegetation changes, I would have expected to see the in-detail comparison of the overlapping date before discussing the differences of the reference period with the 19m system.
The purpose of the paragraph of wind sector discussion in Section 3.6 is not entirely clear to me since it on one hand talks about NDVI but quantify the differences in NDVI only briefly and on the other hand it discussed the different contributing wind directions, namely the land sea (lake) breeze system found at NAMORS due to Nam Co Lake. Considering that the main aim of the study was to relate CO2 fluxes to landscape heterogeneity I would suggest including an overview of the differences found in NDVI also in a map.
Detailed remarks:
Ll 21: Here in the abstract, you mention a 10-month long average of -78 gC m-2 derived from data collected 2006-2018. This does not fit to the data-availability described in the Method section. Did I misunderstand something? Could you please elaborate how this number was calculated? If it is indeed derived from a non continuous dataset, I suggest you mention this here. Otherwise, it confuses the reader later.
L 24: Please define extreme in this context.
L 25: for which period? 2006-2018 or 2019?
Ll28: If one of your main conclusions is that the 3m system is representative for alpine steppe, but that the lake has a rather large influence on CO2 exchange on a landscape level, wouldn’t it make more sense to propose a lake flux system to have continuous fluxes from both surfaces?
Introduction:
Covers the most important background both for the TP and EC, well written and easy to read. Leads into the problem of the spatial differences between in situ observations and larger scale model domains.
Ll 82: I like this paragraph as introduction to the measurements done in this study; however, I am not sure if it is the correct place. It already includes quite a lot of details regarding the 19m system and differences in the properties of the fetch. However, I think that for most readers some more details are missing to fully understand what is going on. I suggest generalizing a bit more here and move some of this paragraph into the method section.
Methods:
2.1 gives a good overview over the site setup, location and used equipment. I think that some of the information from the paragraph ll82 would fit well here. Potentially as a new sub chapter.
ll 230: You describe that you calculated the land cover contribution for each of their 8 wind sectors, but the result of this is not presented adequately in my opinion. This information would help a lot to interpret the overall results.
Results:
Ll250: are these the results of land-land or overall NEE? In my opinion a figure showing the NEE data would make it easier for the reader to follow than just numbers in the text.
3.3 When comparing the single monthly values to the long-term mean from previous years. It would be good to show also how the meteorological conditions compare and how vegetation cover was developing over the time period.
Figure 3: Ther seems to be a large shift in the magnitude of NEE for all seasons from 2010 on at the 3m system. Is there an explanation for this?
Where was the snow depth measured at? Or is this from a remote sensing product?
Ll 350: How large was the difference? What do you mean with practical importance?
This whole section discusses a lot of statistics of comparison of the fluxes measured from the different levels but does only briefly discuss the results of the comparison on a flux level.
NDVI data: Also here a figure or a table would make it easier for the reader to follow and the manuscript much more attractive. What were the actual NDVI values?
Discussion:
Your discussion provides a lot of interesting facts and puts your findings in context. However, I would have liked to see some numbers a various places eg. Ll386 (slightly more release), ll387 (percentage contribution). Overall, the discussion touches a lot of topics, that have not been explored earlier in this study. I think by looking more into detail of the differences in vegetation cover, composition etc in the wind sectors and the different footprint areas, and showing this, some of this discussion points would be more rooted in the manuscript. The figure 6 and snow cover result description should in my opinion be rather moved up into the results.
Ll409: this should have been shown in detail before.
Ll420: this is information missing in the result part.
Ll 425: From here on this is more results than discussion as you discuss only your own results and not with other literature. And foremost this is information that would have helped to follow the described results.
Citation: https://doi.org/10.5194/egusphere-2025-530-RC2 -
AC2: 'Reply on RC2', Nithin dinesan Pillai, 11 Jul 2025
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-530/egusphere-2025-530-AC2-supplement.pdf
-
AC2: 'Reply on RC2', Nithin dinesan Pillai, 11 Jul 2025
Status: closed
-
RC1: 'Comment on egusphere-2025-530', Anonymous Referee #1, 19 Mar 2025
The authors aim to explore the variability in carbon dioxide fluxes across different spatial scales. They attempt doing so using two eddy covariance systems mounted at different heights above ground, 2 and 19 m, at a grassland site in the Tibetan plateau. In addition to the eddy covariance flux data they use footprint modelling to provide the spatial context of the flux measurements and use some remote sensing data.
While the combined analysis of eddy covariance flux data from different heights (and thus different source areas) in combination with footprint modelling and remote sensing data analyses is a potentially powerful approach for gaining insights into the spatial variability of flux source areas, the manuscript falls short of achieving anything meaningful.
First and foremost, the period for which concurrent 2 and 19 m data are available is confined to merely 3 weeks. Given day-to-day variability in weather conditions and the short-term random uncertainty of eddy covariance flux measurements, this is way too short to achieve anything meaningful. The authors attempt to navigate around this key problem by analysing the 2m fluxes which are available for multiple years (and published in a previous study) in comparison to 10 months of flux measurements from 19 m, but this approach is flawed as it ignores the influence of differences in biotic and abiotic conditions (at least in the way the authors attempt this analysis).
Second, while the authors use some footprint model, the way the resulting data are used is highly non-transparent and confusing. It is unclear (in text and figures) when the authors address the 19 m fluxes as measured, i.e. inclusive of the lake contribution, or those restricted to times when almost all flux originates from land only. In addition, the authors appear to use the results of the footprint analysis in a quite simplistic fashion instead of as a powerful tool for exploring the spatial variability of the source area and the resulting effects on the measured fluxes. In the discussion the authors state that they did not do so because of “the inherent variability and potential noise of the flux measurements”. This is quite odd as it would suggest doing so is not possible, which would invalidate the entire point of this study. As a consequence, we do not learn much apart from that the fluxes at 2 and 19 m are different at various temporal scales. Here I should add that I think there is tremendous potential if such an analysis would be done in a proper fashion that exploits the information content that the combination of footprint modeling and remote sensing data analysis offers.
Third, many of the conclusions are not well supported by the results, for example the reasoning about the influence of snow cover is fully anecdotal.
Fourth and finally, the manuscript is poorly structured with entire sections (e.g. 3.6) not being supported by any display items (table or figure), which makes it near impossible to follow the text.
Taken together, the manuscript at present is way too premature to be published.
Citation: https://doi.org/10.5194/egusphere-2025-530-RC1 -
AC1: 'Reply on RC1', Nithin dinesan Pillai, 11 Jul 2025
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-530/egusphere-2025-530-AC1-supplement.pdf
-
AC1: 'Reply on RC1', Nithin dinesan Pillai, 11 Jul 2025
-
RC2: 'Comment on egusphere-2025-530', Anonymous Referee #2, 20 Jun 2025
Within the manuscript titled “Assessing Carbon Flux Variability in an Alpine Steppe: Insights from Dual-Height Measurements” the authors compare flux measurements of CO2 from two eddy covariance (EC) systems mounted at different heights, at Nam Co Station for Multisphere Observation and Research (NAMORS) on the Tibetan Plateau (TP).
Their goal of this approach is to investigate the impact of coarse-scale heterogeneity on the CO2 fluxes measured at NAMORS. For a more detailed analysis of the influence by the underlying surface the authors used a footprint model to estimate the source areas of each EC system and accounted the contribution of different land-use classes to the measured fluxes. Further they used remote sensing products to distinguish difference in snow cover within the footprint area of the two systems during winter month.
The systems were deployed at 3m, the standard observation height for the long-term monitoring system deployed by the Institute of Tibetan Plateau Research, and 19m on the boundary layer tower. While for the long-term monitoring station data is available for the period 2006-2019, the 19m system only provided data from August 2018 till End of May 2019.
Considering the heterogeneity of vegetation cover and the scarcity of reliable measurement stations on the Tibetan Plateau this study sets out to investigate a very important aspect of CO2 exchange estimates for the region. By choosing a height of 19m above ground the footprint of the higher system includes alpine steppe with different vegetation cover and the nearby shallow lake adjacent to Nam Co Lake. The footprint area of the 3m system is composed of alpine steppe.
They see difference in the net ecosystem exchange (NEE) between the fluxes derived from the different systems. The conclude that these differences originate from differences in the vegetation and the lake contribute as well as differences in snow cover during winter month.
Overall, the conducted study shows some potential by addressing an important topic. In general, the background information given in the introduction show that the authors understand the theory behind the applied methods, however the presentation of the results and the structure of the manuscript make it hard to follow the aim of the study and the reasoning of the authors. Due to the restrictions of the data availability, as pointed out by the authors themself, the study falls somewhat short of reaching the aim the authors set out. While the language is clear and the miss match of information in different sections as well as the order of presented results make it rather cumbersome to read. Due to some doubts about the presented method and unclear presentation results, I can not recommend this manuscript to be published in its current state. Overall, I think this is an interesting topic and the authors have presented in parts some interesting results, however in my opinion this manuscript needs a major revision, and potentially additional data should be included.
Dear Authors,
thank you for this manuscript about EC flux measurements on the TP. I think that you are looking at a very interesting topic. While I find the topic to be important and the idea of this study intriguing, I have quite a lot of questions and comments after reading your manuscript. In general I find the manuscript well written and easy to read from a language point of view, figures are all understandable and figure captions sufficient. However I find the structure of the manuscript at times hard to follow.
One of my major points about this manuscript is that the information about available and used data is not consistent. There are miss matches in the description of the data availability of the two systems. The data availability of the 3m system is stated as 2006-2019, with gaps 2012-2015 and 2018-2019, while in Figure 2 it is stated that data from 2006-2007 and 2016-2017 is used for the comparison. This makes it unnecessarily confusing for the reader. Could you please look through the manuscript and fix those mismatches. An easier understandable presentation of the data availability would also help the reader to follow. Further you state that the overlapping time of the two systems was from May 13th to June 3rd 2019 when talking about the processing of the data from the 3m system. However, when talking about data from the 19m system you state that data was analysed from August 01st 2018 to May 31st 2019. This would only leave an overlap of data from May 13th till 31st. While you treat the long-term data set more as a reference data set, I still find the comparison period of less than 3 weeks too short. This is particularly true when comparing the fluxes monthly, especially at the early stage of the vegetation period, before onset of monsoon when fluxes are rather small and uncertainties high. I see a vast improvement if the dataset could be extended to include more overlapping data from both systems.
I like the idea of a more ecosystem and landscape flux approach, which would make it easier to compare the resulting fluxes to satellite products. However, I have some questions and concerns about the method you used.
You explain the treatment of data with contribution from the lake and how a land only and land lake flux dataset was derived from your overall dataset. Given the situation this is interesting and in principle a valid concept. So, from a general data handling perspective I have no issues with this approach, but this leads to some questions that arise about the methods applied.
Why did you choose this the specific height and was the lake contribution was specifically targeted? This does not become clear when reading the manuscript.
Do I understand it correctly that the lake contribution was the difference of the total NEE and the pure land NEE? Considering that it is very common to have a strong land-sea breeze situation at Nam Co with quite distinct times when the EC system sees a lake contribution in the fluxes, does this lead to a large uncertainty the lake contribution?
The estimation of the monthly NEE budget of the 19 m footprint area for land-only and land-lake systems also raises some questions. It might be my misunderstanding of the described method, so could you please explain how this was calculated in more detail? How were lake contributions to the total NEE considered for periods the EC system measured only land fluxes?
I think I can follow the classification into land-land and lake-land due to the contribution derived from the footprint analysis. You describe that you created the land-land and land-lake system dataset by including fluxes form the specific surfaces according to the overall contribution of the flux from this surface to the overall flux. In my understanding this was done for each of the 8 wind sectors you defined and then a composite flux was estimated according to the land use contribution derived from the footprint model. Then monthly budgets were calculated. This is the point I must admit that I can’t follow how this was done. Could you please elaborate more on how it is done? How were fluxes gap filled/modelled for specific land cover for periods when there were no contributions from this direction. In line 240 you describe that that the portioning was performed after gapfilling. Does this mean lake fluxes were gapfilled as if they were land fluxes?
You go on to state that the datasets might have missing half hour value. Can you please explain? Does this mean each dataset was only sorted according to the classification but no continuous dataset for each classification was derived? Are the monthly budgets, growing season and non growing season then calculated on unequal amount of data? Or did you include the relative contribution of the lake fluxes into each half hour based on your footprint calculations? This is a point that needs to be clarified for future publication. If it is rather my lack of understanding, please consider the way how you describe this rather complex but very important part of the data handling.
The subsequent comparison of the two datasets is based on a dataset of an averaged ten-month period resembling the month the 19m system was deployed. While I see some value in this approach, I don’t see how this can lead to the rather in detail comparison the authors present. Considering the rather rapid trend in warming on the TP, coinciding with shifts in precipitations pattern, available soil moisture and potential vegetation changes, I would have expected to see the in-detail comparison of the overlapping date before discussing the differences of the reference period with the 19m system.
The purpose of the paragraph of wind sector discussion in Section 3.6 is not entirely clear to me since it on one hand talks about NDVI but quantify the differences in NDVI only briefly and on the other hand it discussed the different contributing wind directions, namely the land sea (lake) breeze system found at NAMORS due to Nam Co Lake. Considering that the main aim of the study was to relate CO2 fluxes to landscape heterogeneity I would suggest including an overview of the differences found in NDVI also in a map.
Detailed remarks:
Ll 21: Here in the abstract, you mention a 10-month long average of -78 gC m-2 derived from data collected 2006-2018. This does not fit to the data-availability described in the Method section. Did I misunderstand something? Could you please elaborate how this number was calculated? If it is indeed derived from a non continuous dataset, I suggest you mention this here. Otherwise, it confuses the reader later.
L 24: Please define extreme in this context.
L 25: for which period? 2006-2018 or 2019?
Ll28: If one of your main conclusions is that the 3m system is representative for alpine steppe, but that the lake has a rather large influence on CO2 exchange on a landscape level, wouldn’t it make more sense to propose a lake flux system to have continuous fluxes from both surfaces?
Introduction:
Covers the most important background both for the TP and EC, well written and easy to read. Leads into the problem of the spatial differences between in situ observations and larger scale model domains.
Ll 82: I like this paragraph as introduction to the measurements done in this study; however, I am not sure if it is the correct place. It already includes quite a lot of details regarding the 19m system and differences in the properties of the fetch. However, I think that for most readers some more details are missing to fully understand what is going on. I suggest generalizing a bit more here and move some of this paragraph into the method section.
Methods:
2.1 gives a good overview over the site setup, location and used equipment. I think that some of the information from the paragraph ll82 would fit well here. Potentially as a new sub chapter.
ll 230: You describe that you calculated the land cover contribution for each of their 8 wind sectors, but the result of this is not presented adequately in my opinion. This information would help a lot to interpret the overall results.
Results:
Ll250: are these the results of land-land or overall NEE? In my opinion a figure showing the NEE data would make it easier for the reader to follow than just numbers in the text.
3.3 When comparing the single monthly values to the long-term mean from previous years. It would be good to show also how the meteorological conditions compare and how vegetation cover was developing over the time period.
Figure 3: Ther seems to be a large shift in the magnitude of NEE for all seasons from 2010 on at the 3m system. Is there an explanation for this?
Where was the snow depth measured at? Or is this from a remote sensing product?
Ll 350: How large was the difference? What do you mean with practical importance?
This whole section discusses a lot of statistics of comparison of the fluxes measured from the different levels but does only briefly discuss the results of the comparison on a flux level.
NDVI data: Also here a figure or a table would make it easier for the reader to follow and the manuscript much more attractive. What were the actual NDVI values?
Discussion:
Your discussion provides a lot of interesting facts and puts your findings in context. However, I would have liked to see some numbers a various places eg. Ll386 (slightly more release), ll387 (percentage contribution). Overall, the discussion touches a lot of topics, that have not been explored earlier in this study. I think by looking more into detail of the differences in vegetation cover, composition etc in the wind sectors and the different footprint areas, and showing this, some of this discussion points would be more rooted in the manuscript. The figure 6 and snow cover result description should in my opinion be rather moved up into the results.
Ll409: this should have been shown in detail before.
Ll420: this is information missing in the result part.
Ll 425: From here on this is more results than discussion as you discuss only your own results and not with other literature. And foremost this is information that would have helped to follow the described results.
Citation: https://doi.org/10.5194/egusphere-2025-530-RC2 -
AC2: 'Reply on RC2', Nithin dinesan Pillai, 11 Jul 2025
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-530/egusphere-2025-530-AC2-supplement.pdf
-
AC2: 'Reply on RC2', Nithin dinesan Pillai, 11 Jul 2025
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 288 | 60 | 22 | 370 | 24 | 44 |
- HTML: 288
- PDF: 60
- XML: 22
- Total: 370
- BibTeX: 24
- EndNote: 44
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
The authors aim to explore the variability in carbon dioxide fluxes across different spatial scales. They attempt doing so using two eddy covariance systems mounted at different heights above ground, 2 and 19 m, at a grassland site in the Tibetan plateau. In addition to the eddy covariance flux data they use footprint modelling to provide the spatial context of the flux measurements and use some remote sensing data.
While the combined analysis of eddy covariance flux data from different heights (and thus different source areas) in combination with footprint modelling and remote sensing data analyses is a potentially powerful approach for gaining insights into the spatial variability of flux source areas, the manuscript falls short of achieving anything meaningful.
First and foremost, the period for which concurrent 2 and 19 m data are available is confined to merely 3 weeks. Given day-to-day variability in weather conditions and the short-term random uncertainty of eddy covariance flux measurements, this is way too short to achieve anything meaningful. The authors attempt to navigate around this key problem by analysing the 2m fluxes which are available for multiple years (and published in a previous study) in comparison to 10 months of flux measurements from 19 m, but this approach is flawed as it ignores the influence of differences in biotic and abiotic conditions (at least in the way the authors attempt this analysis).
Second, while the authors use some footprint model, the way the resulting data are used is highly non-transparent and confusing. It is unclear (in text and figures) when the authors address the 19 m fluxes as measured, i.e. inclusive of the lake contribution, or those restricted to times when almost all flux originates from land only. In addition, the authors appear to use the results of the footprint analysis in a quite simplistic fashion instead of as a powerful tool for exploring the spatial variability of the source area and the resulting effects on the measured fluxes. In the discussion the authors state that they did not do so because of “the inherent variability and potential noise of the flux measurements”. This is quite odd as it would suggest doing so is not possible, which would invalidate the entire point of this study. As a consequence, we do not learn much apart from that the fluxes at 2 and 19 m are different at various temporal scales. Here I should add that I think there is tremendous potential if such an analysis would be done in a proper fashion that exploits the information content that the combination of footprint modeling and remote sensing data analysis offers.
Third, many of the conclusions are not well supported by the results, for example the reasoning about the influence of snow cover is fully anecdotal.
Fourth and finally, the manuscript is poorly structured with entire sections (e.g. 3.6) not being supported by any display items (table or figure), which makes it near impossible to follow the text.
Taken together, the manuscript at present is way too premature to be published.