the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 19 Apr 2024
How long does carbon stay in a near-pristine central Amazon forest? An empirical estimate with radiocarbon
Abstract. Amazon forests play a significant role in the global C cycle by assimilating large amounts of CO2 through photo- synthesis and storing C largely as biomass and soil organic matter. To evaluate the net budget of C in the Amazon, we must also consider the amplitude and timing of losses of C back to the atmosphere through respiration and biomass burning. One useful timescale metric that integrates such information in terrestrial ecosystems is the transit time of C, defined as the time elapsed between C entering and leaving the ecosystem; transit time is equivalent to the age of C exiting the ecosystem, which occurs mostly through respiration. We estimated the mean transit time of C for a central Amazon forest based on the C age in ecosystem respiration (ER), taking advantage of the large variations in CO2 in the atmosphere below the forest canopy to estimate the radiocarbon signature of mean ER (∆14CER) using Keeling and Miller-Tans mixing models. To evaluate changes in the isotopic signature of the main ER sources, the δ13CER was estimated through Keeling plots using the same samples. We collected air samples in vertical profiles in October 2019 and December 2021 at the Amazon Tall Tower Observatory (ATTO) in the central Amazon. Air samples were collected in a diel cycle from two heights below and one above the canopy (4, 24, and 79 m agl, respectively). For the campaign of October 2019, the ∆14CER was 33.9 ± 7.7 ‰ using the Keeling plot method, and 31.6 ± 7.5 ‰ with the Miller-Tans method. In December 2021, ∆14CER was 77.0 ± 28.3 ‰ using the Keeling plot method, and 77.9 ± 24.0 ‰ with the Miller-Tans method. The δ13CER showed a smaller variation, being -27.8 ± 0.3 ‰ in October 2019 and -29.0 ± 0.5 ‰ in December 2021. Combining the ∆14CER estimates with the record of atmospheric radiocarbon from the bomb period, we obtained estimates of the mean transit time of 6 ± 2 years for 2019 and 18 ± 5 years for 2021. In contrast to steady-state carbon balance models that predict constant mean transit times, these results suggest an important level of variation in mean transit times. Nevertheless, new carbon fixed in this tropical forest is respired, on average, in one or two decades, which means that only a fraction of the assimilated C can act as a sink for decades or longer.
- Preprint
(585 KB) - Metadata XML
-
Supplement
(3943 KB) - BibTeX
- EndNote
Status: closed
-
RC1: 'Comment on egusphere-2024-883', Karis McFarlane, 10 Jun 2024
In this paper, the authors present 14C values for ecosystem respiration in a tropical rainforest in the Amazon, which they derive from a series of 14C measurements of CO2 from 2 field campaigns with samples collected at multiple heights above the ground surface. They estimate the 14C and 13C end-members of ecosystem respiration using both the Keeling Plot and Miller-Trans approach and compare the results from these two approaches to one another. They further compare their 14C values for ecosystem respiration to atmospheric values to derive mean transit times and compare these mean transit times to similar variables (mean turnover time) from the literature determined using different approaches. The approach used in this paper is sound and the data novel and interesting. The manuscript itself could use some additional editing prior to publication. Suggestions for a revised manuscript are outlined below with specific locations of these issues marked in the manuscript pdf with highlight or strikethrough.
In some cases, it seems like things are misplaced or not consistently called out in the appropriate places. It reads as if perhaps it was first written as a short-format paper with the Methods last so some things that should be in the Methods are instead in the Results and Discussion.
At the end of the introduction, you state that you discuss your results in comparison to model predictions. How interesting – why not include this in the Abstract? The section of the Results that describes this comparison needs some reworking. It is quite confusing and is not followed up with sufficient discussion of what the reader should take away from these comparisons.
The use of “14C/C” is confusing, since we do not discuss 14C/C data and 14C is such a small fraction of total C (1 out of 1 trillion C atoms!) You’re also reporting Delta values, which are an isotopic ratio not 14C/C ratios. Either use Delta 14C values or 14C isotopic signatures or something similar throughout.
I found the methods section a bit awkward in that the introduction to the Keeling and Miller-Trans methods came first, but then there are not very clear connections to how the data you collected are used in these approaches (which would justify explaining how the approaches work before discussing how you collected samples and generated the data). Consider describing the study site and field approaches first, then analytical methods, then introduce the approaches for deriving the end-member 14C and 13C for ecosystem respiration. You might add more in-text citations to the Miller-Trans 2003 paper (not just the Phillips et al paper). It’s a little awkward that you don’t cite the older papers when discussing the approach in the methods.
It is not clear what exact approach was used to solve for the 14C of ER. There are different ways to fit linear regressions, which one did you use in R? e.g., Did you use lmodel2 or lm() or something else? See Pataki et al 2003 methods section for discussion and recommendation on dealing with minimizing both x and y errors (model II regression). It’s cited elsewhere in the paper. Whatever you used, this should be in the methods. How are the reported errors, 95% confidence intervals, and ranges derived? This should also be in the methods. It was not clear until one looks at Figure 3 that the Keeling plot method was applied across the 4 tower heights to achieve the range in CO2 concentration. This should be clear from the methods as the Keeling plot approach can also be applied over time at night when CO2 from respiration accumulates. The height-based approach is indeed more sensitive to varying backgrounds across the 4 heights – this would be easier to follow if it were clear that this how the Keeling plot approach is applied here. Also, it is not clear from the methods that the two approaches are used for 13C also until you see the results. This should be clarified in the methods.
How were Delta 14C-ER values converted to mean transit time? The abstract and methods (L206) and results (L273) are vague, please be more specific.
Figure 2 is not very helpful. Can you replot in a way that distinguishes between day and night? The text states that day and night values differed and that they differed between sampling dates, but it is not very easy to compare in this plot. If it can’t be improved to highlight your results more effectively, move it to the supplement.
Section 3.3 is difficult to follow. The header should be something more like “Estimates of mean transit time and comparison to other values from the literature”. I am not convinced this should be in the results, especially considering there is no explanation of how the 14C data were used with the atmospheric data to determine the mean ages or meant transit times. Was this just comparing the end-member values to the atmospheric data and matching them to years in the atmospheric data set or did you use a one pool model? I do not understand where the ranges reported here for the Miller-Trans approach come from. The reported ranges in the text and those in the table are not the same – but appear to be from this study. The paragraph does not describe the results from the Keeling plot approach, but they are in the table. These results are reported as single values in the section 3.2 then as very large ranges in 3.3. All of this needs to be better explained. I do not understand why there are decimal places for everything when the ranges and uncertainties are magnitudes higher. It makes this section more difficult to read and conveys a false level of precision and accuracy. This section needs to be rewritten. I think the comparison to the other methods is useful, but perhaps in the discussion is better. Table 1 also needs footnotes or other clarification of where these values came from (with the references in the table).
There are numerous sentences and phrases that are awkward and could benefit from editing. In some cases, these are word choice, in others they are long sentences that could be easier to read if split into multiple statements. There are several single sentence paragraphs – these need to be restructured into cohesive paragraphs. Some of these are highlighted in the PDF preprint. I have also marked unnecessary text with a strikethrough in the PDF preprint. Specifics are below:
L15: what do you mean by “combining”? Please be more specific. If you are up against the word count, you can reduce the text in the lines above reporting the Keeling plot and Miller-Trans methods – they’re close enough you can provide ranges. And no need to provide decimal places when the errors are greater than 1.
L16-19: The final sentence of the Abstract is problematic. The relatively young mean transit time of ecosystem respiration does not suggest anything about the size of the fraction that is assimilated for decades or longer. Arguably the fact that respired C is so young is a good sign that these systems are not losing older C. Revise this. I understand why the authors might not want to conclude that the presumed increase in mean transit time in 2019 from 6 years to 18 years in 2019 suggests this forest is losing older C but it seems that a stronger statement could be made about what the observed variation might mean and the possible implications should this be a real trend. At minimum it points to something we might want to watch as these forests become increasingly impacted by climate change.
L29: “in land photosynthesis” is awkward phrasing. Consider “among terrestrial ecosystems” or something else.
L30: “might compensate most” is missing a word but I suggest more specific, if not quantitative, language here.
L49-50: This is very awkward and confusing because you’ve not yet explained that the 14C in the atmosphere is oxidized to CO2 – you might specify this in the previous sentence.
L-53-54: Revise this it’s very awkwardly written. The first part of the sentence is about respiration but the end is about organic matter ages. Maybe the age of the organic matter is several years should come first.
L55: integrates rather than reflects seems like a better word choice.
L58: Do not use 14C/C – you are not deriving the 14C/C ratio you are deriving the isotopic ratio and they are not the same thing.
L68: use “and” instead of “/”
L116: I do not understand what “also counts with two more towers” means.
L135: flasks come up here but not earlier – when would flasks from the tall tower not be available? Perhaps it would be better to describe the tall tower data in the section above where the other towers are introduced? Otherwise rather than starting with the tall tower start with the lower altitude samples, then say you used the tall tower as the background (for Miller-Trans approach, I assume – you could specify that here too) and that when you needed to you filled in missing tall tower data with measurements from the 80m walk up tower.
L153: Clarify why then you bothered to remove the water or omit the statement before that you don’t need to remove the water – or both.
L185: just “graphite” the underscore and italics are unnecessary and make it harder to read.
L189: It is difficult to tell for sure if all of the 14C data were corrected for 13C measured on the AMS or if this only happened at MPI. It is also confusing that 13C measurements via IRMS are described between two paragraphs describing 14C measurements. Can you reorganize this section and make this more clear?
L195: Yes, we are lacking in monitoring atmospheric 14C data for the tropics but I think it’s better to state what you did and point out later (perhaps in the Discussion) if you think this lack of data is important for your interpretation of your results.
L233-235: Cut, this type of statement belongs in the discussion, but probably not worth discussing. Be sure to put the previous sentence with the preceding paragraph.
L245-247 and elsewhere: it is a little odd to provide both F14C and Delta. You might explain why you do this in the methods and/or use one throughout but provide the other in parenthesis or in a data table. It’s also not clear if you applied the Keeling plot to F14C and to Delta 14C or if you applied it to F14C and then converted the ecosystem respiration end members to Delta 14C using the year of sampling (2019 and 2021). This should be specified or just use one 14C notation throughout. Figure 3 shows F14C but the figure caption starts with Delta 14C. If you want to show the F14C figure, the caption should read: “Keeling plot of F14C….” Also here, no need to provide decimal places for the Delta values.
L252: It is odd to reference the small variations that have not been presented in the results. Also not clear what you mean by variation at 79 m – between dates? Cut this. I would rephrase that the delta 13C of ER are similar between the Keeling and Miller-Trans approaches, despite the explicit incorporation of background variation in the later method.
L255: “qualify as a violation” is a bit awkward. Perhaps you mean of the assumption of a stable background implicit in the Keeling plot method? I’d omit this statement here and save it for discussion, but you could rephrase that this suggests the small variations in CO2 and 13C at 79 m were small enough not to violate the assumption of a stable background made in the Keeling plot approach. Where are these results shown?
L266: This and the next paragraph should come first seem to be about how you quantified the background for the Miller-Trans approach. This should come first in this section before the first paragraph that reports the results. There is also quite a lot of detail here that arguably could be in the methods or supplement.
L264: “we selected” sounds like you arbitrarily picked one. Better to say “we used” or you should explain how you selected this specifically and what implications this has vs choosing something else.
L287-268: This is unnecessary if all you did was convert Delta to back to Fraction. You must have rearranged Eq 4 to do so and so this is confusing as well as unnecessary.
L270: Why do you provide the dates here (days of month) but not elsewhere? I don’t see why it is necessary.
L275: Where did this large range come from? In L270-271 you report 32 permil for October 2019 from the Miller-Trans method. Same comment for December in the lines that follow. You can’t even see the ribbon for the 95% confidence interval for October – how can it be so large? If these are the correct numbers they should be provided in the section above.
L290: “We were able to obtain” sounds like someone gave them to you. Try “We estimated the mean transit time…”
L299: Why isn’t 2 years included in the range in Table 1? I see 4-24 in Table 1, 2-28 in the text in Section 3.3 and 2-30 here. Why aren’t these values consistent across text, table, and sections?
L301: Delete “may” – “suggests” already hedges your statement. Elsewhere you make the point that this may be variation rather than a [linear trend] change from 2019 to 2021 – I think that it is worth pointing that out directly here when you discuss that variation is observed.
L305: Use difference rather than variation. With 2 data points it is difficult to call this “variation” you just know the 2 values differ. Do you mean variability in mean transit times or do you mean in 13C? This paragraph is mostly about 13C so it seems out of place. Perhaps discuss 13C first and then 14C together or in separate paragraphs.
L306-310: this is very hard to follow with some grammatical issues. Please break this down into multiple sentences. It seems to say that you need to know the background conditions because you use them for the Miller-Trans estimate and because you need them for estimating mean transit times. This is confusing because the first part is about concentrations and isotopes but the second is really just about the 14C. Conflating the requirements of the approaches for the end-members with the mean transit time estimation is confusing. Start with one paragraph to discuss point i and a second paragraph for point ii.
L313: Delete “nevertheless” as you have overcome this limitation – I don’t know that you need to spend so much space on this. You did a good job filling in data gaps appropriately and you can address this in the methods and not belabor the point in the results and discussion. You make the point nicely in the next paragraph that these atmospheric background and calibration data are generally lacking for the tropics.
L330: I think you mean “steady annual decline”. In the next sentence, be clear that you mean the end-member from the Keeling plot vs Miller-Trans approach. You still report a large range (e.g. in Table 1 and the results and beginning of the Discussion) and again I don’t follow where this 2-30 year range (or 15-45 permil range ) comes from when you provide single values from the Keeling and Miller-Trans approaches that are so close to one another.
L333: Are you suggesting that the analytical errors on the raw 14C values are larger than the intercept errors from your linear fits? This is where it is very important to know how you fit the linear regressions as some approaches take into account the uncertainties on the data points, in both x and y, and others do not.
L335: Yes this has already been demonstrated, so it is promising. It is a nice result that it worked here but I don’t understand why you say “especially for the tropical regions”. You might rephrase that you demonstrate that it works for tropical regions, not just temperate ones as demonstrated previously. In the next sentence, it would be more compelling to call for more work to better quantify spatial and temporal variation for the tropics as it is difficult to explain why you have different results for 2019 vs 2021 and whether these differences capture interannual variation or a trend in shifting mean transit times of respired CO2 in the tropics.
L345: This is the first I’ve seen the difference in seasonality pointed to as an explanation for why October 2019 and December 2021 might provide different mean transit times (or 13C). This seems like something that could be mentioned earlier – then you can include seasonal variation in your statement that the method is useful and data is interesting but more is needed to understand what is happening in the tropics (see comment about L335). Perhaps this could be included in the paragraph that starts on L298.
L350: remove the decimal places from this paragraph and if you aren’t going to describe the difference from 2019 to 2021 provide a range or average. Or do you think the more negative numbers in December of 2021 could be because of differences in airflow vs October 2019? I’m not sure what to make of the description of the other study’s data. Can you better connect this study to your findings? If it’s that the values are about the same, then provide the values from the Araujo et al 2008 paper.
L355: Avoid 1 sentence paragraphs. This is a nice topical sentence for the statements that follow.
L361: The Carvalhais et al approach also integrates over longer time periods, no? Isn’t that based on annual productivity and respiration values? You should start here by stating that their 14 year turnover time is consistent with your range from 2021, though you observed a shorter mean residence time range in 2019. Then describe how they derive their value, that it’s integrative over time, doesn’t have quite the same ability to identify sources, and so on.
L372: Start again by comparing your results to theirs. “In contrast, in studies close to Manaus reported mean residence times similar to our range in 2019…” or something along those lines. Then describe how they derived their numbers and what the differences in methods, site, timing, and so on might mean. What about the Sierra et al results for Colombia? They sort of fall right in the middle of what you observed in 2019 and 2021.
L388: I think it’s better not to say “not necessarily mean a limitation of the method” but to go ahead and emphasize that your method provides resolution that the others do not in being able to look at spatial and temporal variability.
L395: Specify for ecosystem respiration – this has been done for soils previously. Also take care, because there may be some mean transit times for ecosystem respiration from non-forested sites using NEE chambers (I’m thinking specifically of CIPHER and other experiments of Ted Schuur’s group in Alaska on permafrost thaw). You have very cool and very novel results, but don’t overstate their uniqueness by being to general in your statements.
L400-403: I don’t follow this statement at all. Better to not introduce normal vs non-normal distributions of transit times in the Conclusion section! Cut or reframe – or move to the discussion where it won’t confuse your takeaway message.
L410-415: This is good, but the paragraph before this is not very interesting. Better to move this last paragraph closer to the beginning of the Conclusions and follow with you demonstrate a method that can help address these open questions about variability and change in mean residence time of C in tropical forests.
- AC1: 'Reply on RC1', Ingrid Chanca, 11 Jul 2024
- AC2: 'Reply on RC1', Ingrid Chanca, 11 Jul 2024
-
RC2: 'Comment on egusphere-2024-883', Anonymous Referee #2, 13 Jun 2024
The work by Chanca et al. provides an empirical determination of the C turnover in the Amazon rainforest, the most important ecosystem for the global carbon balance. The determination of this parameter is therefore of great importance, also because there is hardly any data of this kind. In this study, however, the authors were only able to provide a rough quantification, which underlines the need for further investigations.
The estimation of the mean transit time in this study is subject to large variations. This is due in part to methodological uncertainties and in part to biological variations such as seasonal changes. The authors are aware of the sources of these variations and the discussion of these points is well done.
Overall, the authors have submitted a solid manuscript, which I am happy to support. I have only one somewhat larger criticism related to Table 1 and a number of smaller comments, which I list below:
Regarding Table 1: The table is basically clear, but the text describing the data does not match the data in the table. The Keeling plot data are not described at all. The Mitter-Trans data are different in the table and in the text. Have I missed something here? In my opinion, the text for this table needs to be fundamentally revised. Also, just for clarification: Please also state the data origin (i.e. references) for the lower 3 values in the table legend.
L52: Do you mean “leaf sugars”?
L54: I think it is not about the age of the organic matter itself, but about the age of the respired C (from organic matter), right?
L76-79: As you rightly point out, the Keeling plot approach is based on two sources. But the “CO2 released from the ecosystem respiration” is not a stable source, but composed of many different individual sources. This is not ideal, and I would like to encourage the authors to clarify this point further (either in the M&M or in the discussion).
L140: Please provide some details on the nature of this quality control. This may be helpful to other researchers.
L233/4: We are talking about a difference of 1.2 per mil, right. In my opinion, this indicates different respiration sources or different isotopic composition of these sources. Please clarify.
L238/9: Higher daytime del13C: Any idea what is going on?
L300-3: The references from 2002 and 2003 are too old to support the interpretation of the del13C data. What was the precipitation in 2019 and 2021? Was 2019 drier than 2021?
Figures: The figures have slightly different designs, i.e. points, information on equations, r2,… Please homogenize.
Citation: https://doi.org/10.5194/egusphere-2024-883-RC2 - AC3: 'Reply on RC2', Ingrid Chanca, 11 Jul 2024
- AC2: 'Reply on RC1', Ingrid Chanca, 11 Jul 2024
Status: closed
-
RC1: 'Comment on egusphere-2024-883', Karis McFarlane, 10 Jun 2024
In this paper, the authors present 14C values for ecosystem respiration in a tropical rainforest in the Amazon, which they derive from a series of 14C measurements of CO2 from 2 field campaigns with samples collected at multiple heights above the ground surface. They estimate the 14C and 13C end-members of ecosystem respiration using both the Keeling Plot and Miller-Trans approach and compare the results from these two approaches to one another. They further compare their 14C values for ecosystem respiration to atmospheric values to derive mean transit times and compare these mean transit times to similar variables (mean turnover time) from the literature determined using different approaches. The approach used in this paper is sound and the data novel and interesting. The manuscript itself could use some additional editing prior to publication. Suggestions for a revised manuscript are outlined below with specific locations of these issues marked in the manuscript pdf with highlight or strikethrough.
In some cases, it seems like things are misplaced or not consistently called out in the appropriate places. It reads as if perhaps it was first written as a short-format paper with the Methods last so some things that should be in the Methods are instead in the Results and Discussion.
At the end of the introduction, you state that you discuss your results in comparison to model predictions. How interesting – why not include this in the Abstract? The section of the Results that describes this comparison needs some reworking. It is quite confusing and is not followed up with sufficient discussion of what the reader should take away from these comparisons.
The use of “14C/C” is confusing, since we do not discuss 14C/C data and 14C is such a small fraction of total C (1 out of 1 trillion C atoms!) You’re also reporting Delta values, which are an isotopic ratio not 14C/C ratios. Either use Delta 14C values or 14C isotopic signatures or something similar throughout.
I found the methods section a bit awkward in that the introduction to the Keeling and Miller-Trans methods came first, but then there are not very clear connections to how the data you collected are used in these approaches (which would justify explaining how the approaches work before discussing how you collected samples and generated the data). Consider describing the study site and field approaches first, then analytical methods, then introduce the approaches for deriving the end-member 14C and 13C for ecosystem respiration. You might add more in-text citations to the Miller-Trans 2003 paper (not just the Phillips et al paper). It’s a little awkward that you don’t cite the older papers when discussing the approach in the methods.
It is not clear what exact approach was used to solve for the 14C of ER. There are different ways to fit linear regressions, which one did you use in R? e.g., Did you use lmodel2 or lm() or something else? See Pataki et al 2003 methods section for discussion and recommendation on dealing with minimizing both x and y errors (model II regression). It’s cited elsewhere in the paper. Whatever you used, this should be in the methods. How are the reported errors, 95% confidence intervals, and ranges derived? This should also be in the methods. It was not clear until one looks at Figure 3 that the Keeling plot method was applied across the 4 tower heights to achieve the range in CO2 concentration. This should be clear from the methods as the Keeling plot approach can also be applied over time at night when CO2 from respiration accumulates. The height-based approach is indeed more sensitive to varying backgrounds across the 4 heights – this would be easier to follow if it were clear that this how the Keeling plot approach is applied here. Also, it is not clear from the methods that the two approaches are used for 13C also until you see the results. This should be clarified in the methods.
How were Delta 14C-ER values converted to mean transit time? The abstract and methods (L206) and results (L273) are vague, please be more specific.
Figure 2 is not very helpful. Can you replot in a way that distinguishes between day and night? The text states that day and night values differed and that they differed between sampling dates, but it is not very easy to compare in this plot. If it can’t be improved to highlight your results more effectively, move it to the supplement.
Section 3.3 is difficult to follow. The header should be something more like “Estimates of mean transit time and comparison to other values from the literature”. I am not convinced this should be in the results, especially considering there is no explanation of how the 14C data were used with the atmospheric data to determine the mean ages or meant transit times. Was this just comparing the end-member values to the atmospheric data and matching them to years in the atmospheric data set or did you use a one pool model? I do not understand where the ranges reported here for the Miller-Trans approach come from. The reported ranges in the text and those in the table are not the same – but appear to be from this study. The paragraph does not describe the results from the Keeling plot approach, but they are in the table. These results are reported as single values in the section 3.2 then as very large ranges in 3.3. All of this needs to be better explained. I do not understand why there are decimal places for everything when the ranges and uncertainties are magnitudes higher. It makes this section more difficult to read and conveys a false level of precision and accuracy. This section needs to be rewritten. I think the comparison to the other methods is useful, but perhaps in the discussion is better. Table 1 also needs footnotes or other clarification of where these values came from (with the references in the table).
There are numerous sentences and phrases that are awkward and could benefit from editing. In some cases, these are word choice, in others they are long sentences that could be easier to read if split into multiple statements. There are several single sentence paragraphs – these need to be restructured into cohesive paragraphs. Some of these are highlighted in the PDF preprint. I have also marked unnecessary text with a strikethrough in the PDF preprint. Specifics are below:
L15: what do you mean by “combining”? Please be more specific. If you are up against the word count, you can reduce the text in the lines above reporting the Keeling plot and Miller-Trans methods – they’re close enough you can provide ranges. And no need to provide decimal places when the errors are greater than 1.
L16-19: The final sentence of the Abstract is problematic. The relatively young mean transit time of ecosystem respiration does not suggest anything about the size of the fraction that is assimilated for decades or longer. Arguably the fact that respired C is so young is a good sign that these systems are not losing older C. Revise this. I understand why the authors might not want to conclude that the presumed increase in mean transit time in 2019 from 6 years to 18 years in 2019 suggests this forest is losing older C but it seems that a stronger statement could be made about what the observed variation might mean and the possible implications should this be a real trend. At minimum it points to something we might want to watch as these forests become increasingly impacted by climate change.
L29: “in land photosynthesis” is awkward phrasing. Consider “among terrestrial ecosystems” or something else.
L30: “might compensate most” is missing a word but I suggest more specific, if not quantitative, language here.
L49-50: This is very awkward and confusing because you’ve not yet explained that the 14C in the atmosphere is oxidized to CO2 – you might specify this in the previous sentence.
L-53-54: Revise this it’s very awkwardly written. The first part of the sentence is about respiration but the end is about organic matter ages. Maybe the age of the organic matter is several years should come first.
L55: integrates rather than reflects seems like a better word choice.
L58: Do not use 14C/C – you are not deriving the 14C/C ratio you are deriving the isotopic ratio and they are not the same thing.
L68: use “and” instead of “/”
L116: I do not understand what “also counts with two more towers” means.
L135: flasks come up here but not earlier – when would flasks from the tall tower not be available? Perhaps it would be better to describe the tall tower data in the section above where the other towers are introduced? Otherwise rather than starting with the tall tower start with the lower altitude samples, then say you used the tall tower as the background (for Miller-Trans approach, I assume – you could specify that here too) and that when you needed to you filled in missing tall tower data with measurements from the 80m walk up tower.
L153: Clarify why then you bothered to remove the water or omit the statement before that you don’t need to remove the water – or both.
L185: just “graphite” the underscore and italics are unnecessary and make it harder to read.
L189: It is difficult to tell for sure if all of the 14C data were corrected for 13C measured on the AMS or if this only happened at MPI. It is also confusing that 13C measurements via IRMS are described between two paragraphs describing 14C measurements. Can you reorganize this section and make this more clear?
L195: Yes, we are lacking in monitoring atmospheric 14C data for the tropics but I think it’s better to state what you did and point out later (perhaps in the Discussion) if you think this lack of data is important for your interpretation of your results.
L233-235: Cut, this type of statement belongs in the discussion, but probably not worth discussing. Be sure to put the previous sentence with the preceding paragraph.
L245-247 and elsewhere: it is a little odd to provide both F14C and Delta. You might explain why you do this in the methods and/or use one throughout but provide the other in parenthesis or in a data table. It’s also not clear if you applied the Keeling plot to F14C and to Delta 14C or if you applied it to F14C and then converted the ecosystem respiration end members to Delta 14C using the year of sampling (2019 and 2021). This should be specified or just use one 14C notation throughout. Figure 3 shows F14C but the figure caption starts with Delta 14C. If you want to show the F14C figure, the caption should read: “Keeling plot of F14C….” Also here, no need to provide decimal places for the Delta values.
L252: It is odd to reference the small variations that have not been presented in the results. Also not clear what you mean by variation at 79 m – between dates? Cut this. I would rephrase that the delta 13C of ER are similar between the Keeling and Miller-Trans approaches, despite the explicit incorporation of background variation in the later method.
L255: “qualify as a violation” is a bit awkward. Perhaps you mean of the assumption of a stable background implicit in the Keeling plot method? I’d omit this statement here and save it for discussion, but you could rephrase that this suggests the small variations in CO2 and 13C at 79 m were small enough not to violate the assumption of a stable background made in the Keeling plot approach. Where are these results shown?
L266: This and the next paragraph should come first seem to be about how you quantified the background for the Miller-Trans approach. This should come first in this section before the first paragraph that reports the results. There is also quite a lot of detail here that arguably could be in the methods or supplement.
L264: “we selected” sounds like you arbitrarily picked one. Better to say “we used” or you should explain how you selected this specifically and what implications this has vs choosing something else.
L287-268: This is unnecessary if all you did was convert Delta to back to Fraction. You must have rearranged Eq 4 to do so and so this is confusing as well as unnecessary.
L270: Why do you provide the dates here (days of month) but not elsewhere? I don’t see why it is necessary.
L275: Where did this large range come from? In L270-271 you report 32 permil for October 2019 from the Miller-Trans method. Same comment for December in the lines that follow. You can’t even see the ribbon for the 95% confidence interval for October – how can it be so large? If these are the correct numbers they should be provided in the section above.
L290: “We were able to obtain” sounds like someone gave them to you. Try “We estimated the mean transit time…”
L299: Why isn’t 2 years included in the range in Table 1? I see 4-24 in Table 1, 2-28 in the text in Section 3.3 and 2-30 here. Why aren’t these values consistent across text, table, and sections?
L301: Delete “may” – “suggests” already hedges your statement. Elsewhere you make the point that this may be variation rather than a [linear trend] change from 2019 to 2021 – I think that it is worth pointing that out directly here when you discuss that variation is observed.
L305: Use difference rather than variation. With 2 data points it is difficult to call this “variation” you just know the 2 values differ. Do you mean variability in mean transit times or do you mean in 13C? This paragraph is mostly about 13C so it seems out of place. Perhaps discuss 13C first and then 14C together or in separate paragraphs.
L306-310: this is very hard to follow with some grammatical issues. Please break this down into multiple sentences. It seems to say that you need to know the background conditions because you use them for the Miller-Trans estimate and because you need them for estimating mean transit times. This is confusing because the first part is about concentrations and isotopes but the second is really just about the 14C. Conflating the requirements of the approaches for the end-members with the mean transit time estimation is confusing. Start with one paragraph to discuss point i and a second paragraph for point ii.
L313: Delete “nevertheless” as you have overcome this limitation – I don’t know that you need to spend so much space on this. You did a good job filling in data gaps appropriately and you can address this in the methods and not belabor the point in the results and discussion. You make the point nicely in the next paragraph that these atmospheric background and calibration data are generally lacking for the tropics.
L330: I think you mean “steady annual decline”. In the next sentence, be clear that you mean the end-member from the Keeling plot vs Miller-Trans approach. You still report a large range (e.g. in Table 1 and the results and beginning of the Discussion) and again I don’t follow where this 2-30 year range (or 15-45 permil range ) comes from when you provide single values from the Keeling and Miller-Trans approaches that are so close to one another.
L333: Are you suggesting that the analytical errors on the raw 14C values are larger than the intercept errors from your linear fits? This is where it is very important to know how you fit the linear regressions as some approaches take into account the uncertainties on the data points, in both x and y, and others do not.
L335: Yes this has already been demonstrated, so it is promising. It is a nice result that it worked here but I don’t understand why you say “especially for the tropical regions”. You might rephrase that you demonstrate that it works for tropical regions, not just temperate ones as demonstrated previously. In the next sentence, it would be more compelling to call for more work to better quantify spatial and temporal variation for the tropics as it is difficult to explain why you have different results for 2019 vs 2021 and whether these differences capture interannual variation or a trend in shifting mean transit times of respired CO2 in the tropics.
L345: This is the first I’ve seen the difference in seasonality pointed to as an explanation for why October 2019 and December 2021 might provide different mean transit times (or 13C). This seems like something that could be mentioned earlier – then you can include seasonal variation in your statement that the method is useful and data is interesting but more is needed to understand what is happening in the tropics (see comment about L335). Perhaps this could be included in the paragraph that starts on L298.
L350: remove the decimal places from this paragraph and if you aren’t going to describe the difference from 2019 to 2021 provide a range or average. Or do you think the more negative numbers in December of 2021 could be because of differences in airflow vs October 2019? I’m not sure what to make of the description of the other study’s data. Can you better connect this study to your findings? If it’s that the values are about the same, then provide the values from the Araujo et al 2008 paper.
L355: Avoid 1 sentence paragraphs. This is a nice topical sentence for the statements that follow.
L361: The Carvalhais et al approach also integrates over longer time periods, no? Isn’t that based on annual productivity and respiration values? You should start here by stating that their 14 year turnover time is consistent with your range from 2021, though you observed a shorter mean residence time range in 2019. Then describe how they derive their value, that it’s integrative over time, doesn’t have quite the same ability to identify sources, and so on.
L372: Start again by comparing your results to theirs. “In contrast, in studies close to Manaus reported mean residence times similar to our range in 2019…” or something along those lines. Then describe how they derived their numbers and what the differences in methods, site, timing, and so on might mean. What about the Sierra et al results for Colombia? They sort of fall right in the middle of what you observed in 2019 and 2021.
L388: I think it’s better not to say “not necessarily mean a limitation of the method” but to go ahead and emphasize that your method provides resolution that the others do not in being able to look at spatial and temporal variability.
L395: Specify for ecosystem respiration – this has been done for soils previously. Also take care, because there may be some mean transit times for ecosystem respiration from non-forested sites using NEE chambers (I’m thinking specifically of CIPHER and other experiments of Ted Schuur’s group in Alaska on permafrost thaw). You have very cool and very novel results, but don’t overstate their uniqueness by being to general in your statements.
L400-403: I don’t follow this statement at all. Better to not introduce normal vs non-normal distributions of transit times in the Conclusion section! Cut or reframe – or move to the discussion where it won’t confuse your takeaway message.
L410-415: This is good, but the paragraph before this is not very interesting. Better to move this last paragraph closer to the beginning of the Conclusions and follow with you demonstrate a method that can help address these open questions about variability and change in mean residence time of C in tropical forests.
- AC1: 'Reply on RC1', Ingrid Chanca, 11 Jul 2024
- AC2: 'Reply on RC1', Ingrid Chanca, 11 Jul 2024
-
RC2: 'Comment on egusphere-2024-883', Anonymous Referee #2, 13 Jun 2024
The work by Chanca et al. provides an empirical determination of the C turnover in the Amazon rainforest, the most important ecosystem for the global carbon balance. The determination of this parameter is therefore of great importance, also because there is hardly any data of this kind. In this study, however, the authors were only able to provide a rough quantification, which underlines the need for further investigations.
The estimation of the mean transit time in this study is subject to large variations. This is due in part to methodological uncertainties and in part to biological variations such as seasonal changes. The authors are aware of the sources of these variations and the discussion of these points is well done.
Overall, the authors have submitted a solid manuscript, which I am happy to support. I have only one somewhat larger criticism related to Table 1 and a number of smaller comments, which I list below:
Regarding Table 1: The table is basically clear, but the text describing the data does not match the data in the table. The Keeling plot data are not described at all. The Mitter-Trans data are different in the table and in the text. Have I missed something here? In my opinion, the text for this table needs to be fundamentally revised. Also, just for clarification: Please also state the data origin (i.e. references) for the lower 3 values in the table legend.
L52: Do you mean “leaf sugars”?
L54: I think it is not about the age of the organic matter itself, but about the age of the respired C (from organic matter), right?
L76-79: As you rightly point out, the Keeling plot approach is based on two sources. But the “CO2 released from the ecosystem respiration” is not a stable source, but composed of many different individual sources. This is not ideal, and I would like to encourage the authors to clarify this point further (either in the M&M or in the discussion).
L140: Please provide some details on the nature of this quality control. This may be helpful to other researchers.
L233/4: We are talking about a difference of 1.2 per mil, right. In my opinion, this indicates different respiration sources or different isotopic composition of these sources. Please clarify.
L238/9: Higher daytime del13C: Any idea what is going on?
L300-3: The references from 2002 and 2003 are too old to support the interpretation of the del13C data. What was the precipitation in 2019 and 2021? Was 2019 drier than 2021?
Figures: The figures have slightly different designs, i.e. points, information on equations, r2,… Please homogenize.
Citation: https://doi.org/10.5194/egusphere-2024-883-RC2 - AC3: 'Reply on RC2', Ingrid Chanca, 11 Jul 2024
- AC2: 'Reply on RC1', Ingrid Chanca, 11 Jul 2024
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 419 | 134 | 43 | 596 | 44 | 33 | 33 |
- HTML: 419
- PDF: 134
- XML: 43
- Total: 596
- Supplement: 44
- BibTeX: 33
- EndNote: 33
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1