the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 05 Feb 2025
Global CH4 Fluxes Derived from JAXA/GOSAT Lower Tropospheric Partial Column Data and the CTE-CH4 Atmospheric Inverse Model
Abstract. Satellite-driven inversions provide valuable information about methane (CH4) fluxes, but the assimilation of total column-averaged dry-air mole fractions of CH4 (XCH4) has been challenging. This study explores, for the first time, the potential of the new lower tropospheric partial column (pXCH4_LT) GOSAT data, retrieved by the Japan Aerospace Exploration Agency (JAXA), to constrain global and regional CH4 fluxes. Using the CarbonTracker Europe-CH4 atmospheric inverse model, we estimated CH4 fluxes between 2016–2019 by assimilating the JAXA/GOSAT pXCH4_LT and XCH4 data and surface CH4 observations, independently of each other. The Northern Hemisphere CH4 fluxes derived from the JAXA/GOSAT pXCH4_LT data were similar to the estimates derived from the surface observations, but was underestimated by about 35 Tg CH4 year-1 (∼6 % of the global total) using the JAXA/GOSAT XCH4 data. For the Southern Hemisphere, the estimates from the both GOSAT inversions were about 15–30 Tg CH4 year-1 higher than that derived from surface data. The evaluations against independent data from the Atmospheric Tomography Mission aircraft campaign showed good agreement in the lower tropospheric CH4 from the inversions using the JAXA/GOSAT pXCH4_LT and surface data. However, the modelled North-South gradients showed significant overestimation in the upper troposphere and stratosphere, possibly due to relatively uniform inter-hemispheric OH distributions that control CH4 sinks. Overall, we found that the use of the JAXA/GOSAT pXCH4_LT data shows considerable potential in constraining global and regional CH4 fluxes, advancing our understanding of the CH4 budget.
-
Notice on discussion status
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
-
Preprint
(12623 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(12623 KB) - Metadata XML
- BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2025-159', Anonymous Referee #1, 14 Mar 2025
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-159/egusphere-2025-159-RC1-supplement.pdf
-
RC2: 'Comment on egusphere-2025-159', Anonymous Referee #2, 18 Mar 2025
General comments
Since CH4 is one of the most important greenhouse gases in Earth's atmosphere, at lot of research effort is put into measuring its concentration either in situ or remotely. Those measurement can be used to derive methane fluxes at the surface using various inversion techniques. The authors use a novel methane retrieval from JAXA/GOSAT in an existing ensemble Kalman filter based inversion system. The main conclusion is that assimilating the lower tropospheric column from the novel retrieval is more consistent with the results from a retrieval using surface data than with a retrieval using total column data. This is an interesting conlusion in
view of other synergistic retrievals using the SWIR and TIR wavelength bands from other instruments. Therefore, the paper presents an important contribution to the field and is worthy of prompt publication, after careful consideration of the comments presented below.Specific comments
Section 2.1:
-) Some clarification of the horizontal resolutions would be appreciated. If the fluxes are optimized at 1x1 over Eurasia and 2x3 over other land areas, how does this work with a model resolution of 1x1 over Europe and 6x4 over the rest of the globe?
-) Why choose for 1x1 over Europe and 6x4 over the rest of the globe anyway? You focus on global results, so wouldn't it make more sense to use the same resolution everywhere? You did run the model on 1x1 and 3x2 outside Europe anyway for figure A2.
-) It is mentioned that the atmospheric chemistry is the same as in Houweling (2014), which uses the OH distribution from Spivakovski (2000). That is also mentioned in line 365. But then please clarify the use of the ECHAM/MESSy1 model in line 98.Section 2.3.1:
-) In lines 133-135, two major shortcomings in the current approach are mentioned: the lack of vertical interpolation and application of averaging kernels. That may be understandable for the AKs, since they are lacking the v2 version of the GOSAT product. However, more explanation on why vertical interpolation was not applied should be added to the text. And in the discussion, the possible effects of not including vertical interpolation or not applying the AK should be discussed.
-) In lines 139-141 (and again in line 203), the observational uncertainty is set to 30 or 50 ppb depending on the product and location, and the rejection threshold is set to two times those values. But if the uncertainty is set to a fixed value, then the observations will never be rejected. So some clarification on the use of the rejection threshold should be added to the text.Section 2.4:
-) Since the CTE-CH4 implements the ensemble Kalman filter, it is surprising that the number of enseble members is not mentioned in the text. According to the caption of figure 4, there are 500 ensemble members, but that may be a bit high. Please clarify.
-) Why are the four inversion experiments from this section selected? More specific, what is the reason for doing the InvGLT_land experiment?Section 3.1:
-) Lines 218-221: Is it possible that not applying vertical interpolation (see section 2.3.1) or the coarse model resolution outside Europe also contribute to the observed biases? For example, the surface elevation of the model over mountainous regions such as western South America (i.e. Andes mountains) will be different from the surface elevation of the satellite measurements. This difference in elevation will affect the derived methane column. And how will the lack of AKs affect these biases?
-) Line 230: Comparing figure A2 with with figure 1, it appears that the (SH) tropical bias deteriorates with higher resolution. If smaller biases over Europe are the result of a higher model resolution, then why is that not the case globally? I agree that the biases between Europe and Russia are smaller for 3x2 (A2, def), but this is not true for the inversion with the total column (A2, b) on global 1x1 resolution.
-) Line 232: ... smaller during NH winter than summer... However, if I look at Figure 2, I would say that the minimum is more in October than in NH winter.Section 3.2:
-) Figure 3: What are the "obs", surface values or total columns? I don't think that they are total columns, since the observations are "surface ground based and shipboard". But InvGTOT assimated the total columns, so are you showing the surface concentration from that experiment here? In other words, are you comparing the same quantities here? Total column XCH4 will be lower than surface observations due to the shape of the CH4 profile. In the caption, please rephrase the line "For (a), the data that were assimilated in InvSURF were used". But that inversion is shown in every plot, so that is meant by that line?
-) Line 255: "... InvGTOT below 4000 meter ...", but Fig 3b mentions 2000m. Please clarify.
-) Line 259/260: I'm not sure that InvGTOT better captures the latitudinal gradient. InvGTOT still underestimates the the observations in the 4000-8000 meter range. Please add some more explanation.Technical corrections
line 7: change was to were
line 55: change inversion to inversions
line 59: remove the first occurence of thermal
line 66: remove "in"
line 82 / eq. 1: Below this equation, you do not mention x^b, and in line 83/84 y and x are in italic, while in the equation they are in bold. And there's an error in the equation as well: the observation operator does not operate on x^b, but on x.
line 110: in general, I think the English in section 2.3.1 should be reviewed
line 111/112: Replace with "The JAXA/GOSAT retrieval algorithm is based on the Full Physics algorithm and is extended to use simultaneous both the 2-orthogonal SWIR and TIR signals."
line 113: replace "and use..." with "and uses..."
line 114/115: Please rephrase the line "The an ... retrieval process".
line 116/117: Replace "Two layers ... are derived" with something like "CO2 and CH4 partial column-averaged concentrations are derived for two layers in the troposphere and three layers in the stratosphere. The H2O concentration is derived on 11 vertical layers."
line 118: Add "CO2 and CH4" to the start of the line: "The five CO2 and CH4 layers..."
line 143: "examined" should be "justified"?
line 144: add a "d" to "compare"
line 228/229: Please rephrase the line "This resulted...the shift".
line 233: Remove "it" from "... biases it showed ..."
line 256: Replace "this" with "these" in "In this altitude..."
line 280: Add a "degree sign" to "(30 latitude band)"
line 387: Replace "is" with "occur" in "These differences is probably..."
Figure A1: In the caption, InvSURF refers to one of the inversion experiments described in section 2.4. But the first time this figure is referred to is in section 2.1 (line 87), when the inversion experiments are not mentioned yet. So suggest to update this caption with a reference to section 2.4, e.g.: "... ground-based surface observations (dots and x-marks) assimilated in the InvSURF inversion experiment (see section 2.4)..."
Figure A4/A5: replace TOM with ATom in the legend of both figures
Figure A7: the dates on the x-axis overlapCitation: https://doi.org/10.5194/egusphere-2025-159-RC2 -
AC1: 'Replies to reviewers on egusphere-2025-159', Aki Tsuruta, 02 May 2025
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-159/egusphere-2025-159-AC1-supplement.pdf
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2025-159', Anonymous Referee #1, 14 Mar 2025
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-159/egusphere-2025-159-RC1-supplement.pdf
-
RC2: 'Comment on egusphere-2025-159', Anonymous Referee #2, 18 Mar 2025
General comments
Since CH4 is one of the most important greenhouse gases in Earth's atmosphere, at lot of research effort is put into measuring its concentration either in situ or remotely. Those measurement can be used to derive methane fluxes at the surface using various inversion techniques. The authors use a novel methane retrieval from JAXA/GOSAT in an existing ensemble Kalman filter based inversion system. The main conclusion is that assimilating the lower tropospheric column from the novel retrieval is more consistent with the results from a retrieval using surface data than with a retrieval using total column data. This is an interesting conlusion in
view of other synergistic retrievals using the SWIR and TIR wavelength bands from other instruments. Therefore, the paper presents an important contribution to the field and is worthy of prompt publication, after careful consideration of the comments presented below.Specific comments
Section 2.1:
-) Some clarification of the horizontal resolutions would be appreciated. If the fluxes are optimized at 1x1 over Eurasia and 2x3 over other land areas, how does this work with a model resolution of 1x1 over Europe and 6x4 over the rest of the globe?
-) Why choose for 1x1 over Europe and 6x4 over the rest of the globe anyway? You focus on global results, so wouldn't it make more sense to use the same resolution everywhere? You did run the model on 1x1 and 3x2 outside Europe anyway for figure A2.
-) It is mentioned that the atmospheric chemistry is the same as in Houweling (2014), which uses the OH distribution from Spivakovski (2000). That is also mentioned in line 365. But then please clarify the use of the ECHAM/MESSy1 model in line 98.Section 2.3.1:
-) In lines 133-135, two major shortcomings in the current approach are mentioned: the lack of vertical interpolation and application of averaging kernels. That may be understandable for the AKs, since they are lacking the v2 version of the GOSAT product. However, more explanation on why vertical interpolation was not applied should be added to the text. And in the discussion, the possible effects of not including vertical interpolation or not applying the AK should be discussed.
-) In lines 139-141 (and again in line 203), the observational uncertainty is set to 30 or 50 ppb depending on the product and location, and the rejection threshold is set to two times those values. But if the uncertainty is set to a fixed value, then the observations will never be rejected. So some clarification on the use of the rejection threshold should be added to the text.Section 2.4:
-) Since the CTE-CH4 implements the ensemble Kalman filter, it is surprising that the number of enseble members is not mentioned in the text. According to the caption of figure 4, there are 500 ensemble members, but that may be a bit high. Please clarify.
-) Why are the four inversion experiments from this section selected? More specific, what is the reason for doing the InvGLT_land experiment?Section 3.1:
-) Lines 218-221: Is it possible that not applying vertical interpolation (see section 2.3.1) or the coarse model resolution outside Europe also contribute to the observed biases? For example, the surface elevation of the model over mountainous regions such as western South America (i.e. Andes mountains) will be different from the surface elevation of the satellite measurements. This difference in elevation will affect the derived methane column. And how will the lack of AKs affect these biases?
-) Line 230: Comparing figure A2 with with figure 1, it appears that the (SH) tropical bias deteriorates with higher resolution. If smaller biases over Europe are the result of a higher model resolution, then why is that not the case globally? I agree that the biases between Europe and Russia are smaller for 3x2 (A2, def), but this is not true for the inversion with the total column (A2, b) on global 1x1 resolution.
-) Line 232: ... smaller during NH winter than summer... However, if I look at Figure 2, I would say that the minimum is more in October than in NH winter.Section 3.2:
-) Figure 3: What are the "obs", surface values or total columns? I don't think that they are total columns, since the observations are "surface ground based and shipboard". But InvGTOT assimated the total columns, so are you showing the surface concentration from that experiment here? In other words, are you comparing the same quantities here? Total column XCH4 will be lower than surface observations due to the shape of the CH4 profile. In the caption, please rephrase the line "For (a), the data that were assimilated in InvSURF were used". But that inversion is shown in every plot, so that is meant by that line?
-) Line 255: "... InvGTOT below 4000 meter ...", but Fig 3b mentions 2000m. Please clarify.
-) Line 259/260: I'm not sure that InvGTOT better captures the latitudinal gradient. InvGTOT still underestimates the the observations in the 4000-8000 meter range. Please add some more explanation.Technical corrections
line 7: change was to were
line 55: change inversion to inversions
line 59: remove the first occurence of thermal
line 66: remove "in"
line 82 / eq. 1: Below this equation, you do not mention x^b, and in line 83/84 y and x are in italic, while in the equation they are in bold. And there's an error in the equation as well: the observation operator does not operate on x^b, but on x.
line 110: in general, I think the English in section 2.3.1 should be reviewed
line 111/112: Replace with "The JAXA/GOSAT retrieval algorithm is based on the Full Physics algorithm and is extended to use simultaneous both the 2-orthogonal SWIR and TIR signals."
line 113: replace "and use..." with "and uses..."
line 114/115: Please rephrase the line "The an ... retrieval process".
line 116/117: Replace "Two layers ... are derived" with something like "CO2 and CH4 partial column-averaged concentrations are derived for two layers in the troposphere and three layers in the stratosphere. The H2O concentration is derived on 11 vertical layers."
line 118: Add "CO2 and CH4" to the start of the line: "The five CO2 and CH4 layers..."
line 143: "examined" should be "justified"?
line 144: add a "d" to "compare"
line 228/229: Please rephrase the line "This resulted...the shift".
line 233: Remove "it" from "... biases it showed ..."
line 256: Replace "this" with "these" in "In this altitude..."
line 280: Add a "degree sign" to "(30 latitude band)"
line 387: Replace "is" with "occur" in "These differences is probably..."
Figure A1: In the caption, InvSURF refers to one of the inversion experiments described in section 2.4. But the first time this figure is referred to is in section 2.1 (line 87), when the inversion experiments are not mentioned yet. So suggest to update this caption with a reference to section 2.4, e.g.: "... ground-based surface observations (dots and x-marks) assimilated in the InvSURF inversion experiment (see section 2.4)..."
Figure A4/A5: replace TOM with ATom in the legend of both figures
Figure A7: the dates on the x-axis overlapCitation: https://doi.org/10.5194/egusphere-2025-159-RC2 -
AC1: 'Replies to reviewers on egusphere-2025-159', Aki Tsuruta, 02 May 2025
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-159/egusphere-2025-159-AC1-supplement.pdf
Peer review completion
Journal article(s) based on this preprint
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 329 | 97 | 15 | 441 | 16 | 34 |
- HTML: 329
- PDF: 97
- XML: 15
- Total: 441
- BibTeX: 16
- EndNote: 34
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Akihiko Kuze
Kei Shiomi
Fumie Kataoka
Nobuhiro Kikuchi
Tuula Aalto
Leif Backman
Ella Kivimäki
Maria K. Tenkanen
Kathryn McKain
Omaira E. García
Frank Hase
Rigel Kivi
Isamu Morino
Hirofumi Ohyama
David F. Pollard
Mahesh K. Sha
Kimberly Strong
Ralf Sussmann
Voltaire A. Velazco
Mihalis Vrekoussis
Thorsten Warneke
Minqiang Zhou
Hiroshi Suto
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(12623 KB) - Metadata XML