Computation and analysis of atmospheric carbon dioxide annual mean growth rates from satellite observations during 2003–2016

Buchwitz, Michael; Reuter, Maximilian; Schneising, Oliver; Noël, Stefan; Gier, Bettina; Bovensmann, Heinrich; Burrows, John P.; Boesch, Hartmut; Anand, Jasdeep; Parker, Robert J.; Somkuti, Peter; Detmers, Rob G.; Hasekamp, Otto P.; Aben, Ilse; Butz, André; Kuze, Akihiko; Suto, Hiroshi; Yoshida, Yukio; Crisp, David; O'Dell, Christopher

doi:https://doi.org/10.5194/acp-18-17355-2018

Articles | Volume 18, issue 23

https://doi.org/10.5194/acp-18-17355-2018

© Author(s) 2018. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/acp-18-17355-2018

© Author(s) 2018. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 18, issue 23

Research article

|

07 Dec 2018

Research article |

| 07 Dec 2018

Computation and analysis of atmospheric carbon dioxide annual mean growth rates from satellite observations during 2003–2016

Michael Buchwitz, Maximilian Reuter, Oliver Schneising, Stefan Noël, Bettina Gier, Heinrich Bovensmann, John P. Burrows, Hartmut Boesch, Jasdeep Anand, Robert J. Parker, Peter Somkuti, Rob G. Detmers, Otto P. Hasekamp, Ilse Aben, André Butz, Akihiko Kuze, Hiroshi Suto, Yukio Yoshida, David Crisp, and Christopher O'Dell

Download

Final revised paper (published on 07 Dec 2018)
Preprint (discussion started on 14 Mar 2018)

Interactive discussion

Status: closed

AC: Author comment | RC: Referee comment | SC: Short comment | EC: Editor comment

- Printer-friendly version

- Supplement

RC1: 'Review', Anonymous Referee #1, 07 Apr 2018
- AC1: 'Reply to Anonymous Referee #1', Michael Buchwitz, 05 Jun 2018
RC2: 'Review of Buchwitz et al (2018)', Anonymous Referee #2, 17 Apr 2018
- AC2: 'Reply to Anonymous Referee #2', Michael Buchwitz, 05 Jun 2018

Peer-review completion

AR: Author's response | RR: Referee report | ED: Editor decision

AR by Michael Buchwitz on behalf of the Authors (05 Jun 2018) Author's response Manuscript

ED: Referee Nomination & Report Request started (10 Jun 2018) by Christoph Gerbig

RR by Anonymous Referee #1 (14 Jun 2018)

Suggestions for revision or reasons for rejection

My two main criticisms of this paper were their use of: 1) regional atmospheric growth rates and 2) a simple statistical model to attribute changes in atmospheric CO2 to human emissions and ENSO. Neither comment is addressed well by the authors.
Their response to my first comment is limited to two additional statements neither of which I fully understand:
a) " Growth rate time series for several latitude bands are shown in Fig. 4. As can be seen from Fig. 4, the growth rates are similar in all latitude bands including the global results (for numerical values see Tab. 2). The reason for this is that atmospheric CO2 is long-lived and therefore well-mixed.”
b) “As a result of atmospheric transport and mixing, similar mean annual CO2 growth rates, within their measurements error, are expected for all values derived at the different latitude bands. This behaviour is shown in Fig. 4 and is interpreted as an indication of the good quality of the satellite XCO2 data product and the adequacy of the method used to compute the annual mean CO2 growth rates.”
Atmospheric CO2 is well mixed but the authors are trying to determine the small, annual changes that sit on the growing well-mixed background. Atmospheric CO2 has an interhemispheric gradient, which is determined by hemispheric differences in emissions but also by the ~1 year mean interhemispheric crossing time. It can take weeks-months for signals to be transported from the tropics to midlatitudes so a seasonal change in one latitude band in year one may straddle growth rates in that year and year two. I also don't follow their argument about using regional growth rates to comment on the quality of the CO2 data.
Their response to my second comment is a bit odd in my opinion. I agree that attempting to attribute observed changes in atmospheric CO2 to human emissions and ENSO is of great importance. However, my original comment said that the method they used to attribute human emissions and ENSO was a fool's errand. I made this comment for a number of reasons:
* the spatial and temporal distributions of human emissions and the manifold responses of the land biosphere to regional weather patterns are not necessarily distinct.
* is there a disprovable result reported by this correlation analysis? The approach/results are so amorphous and ill defined that I find it hard to understand what new knowledge I have gained from this analysis.
* if, as the authors claim, their analysis are not in conflict with Liu et al then they should show it.

Hide

RR by Anonymous Referee #2 (22 Jul 2018)

ED: Reconsider after major revisions (06 Aug 2018) by Christoph Gerbig

AR by Michael Buchwitz on behalf of the Authors (21 Aug 2018) Author's response Manuscript

ED: Referee Nomination & Report Request started (26 Aug 2018) by Christoph Gerbig

RR by Anonymous Referee #3 (14 Sep 2018)

Suggestions for revision or reasons for rejection

Review of “Computation and analysis of atmospheric carbon dioxide annual mean growth rates from satellite observations during 2003-2016” by Buchwitz et al.

The authors presented a new Level 3 XCO2 product, based on data from SCIAMACHY and GOSAT, and examined the atmospheric growth rate of CO2 captured by these data. They showed that the annual mean CO2 growth rate estimated from the XCO2 data is consistent with that estimated from the NOAA surface CO2 data. They also did an analysis to determine the relative contributions of ENSO and anthropogenic CO2 emissions to variations in the annual mean CO2 growth rate. The new Level 3 data will be a useful product for the community since working with Level 2 data can be challenging. The fact that the XCO2-based CO2 growth rate is consistent with that estimated from the surface data is reassuring. However, I cannot recommend the manuscript for publication in ACP. I do not believe that the manuscript contains sufficiently new scientific results to warrant publication in ACP.

In responding to the previous reviews, the authors described what new knowledge is contained in the manuscript. They stated that:
• "We present a new global total column CO2 (“XCO2”) data set (based on satellite data) covering 14 years
• We present a new method to compute annual mean XCO2 growth rates from this data set
• We present a new annual mean CO2 growth rate time series (covering the entire atmosphere, not only near-surface CO2) including a comparison with growth rates from NOAA based on surface CO2 observations; we find agreement within the reported uncertainty ranges and therefore consider our growth rates to be validated
• We present an answer to the question “Assuming that the variability of the CO2 growth rate is dominated by ENSO and by human emissions, which of the two considered causes dominates the growth rate variability given the satellite-derived growth rates and their uncertainty?” To answer this question we used a statistical analysis method, which we clearly explain. Our answer is given in the Conclusions section: “Our analysis also shows that the ENSO impact on CO2 growth rate variations dominates over that of human emissions throughout the period 2003-2016 (14 years) but in particular during the period 2010-2016 (second half of the investigated time period) due to strong La Niña and El Niño events. We estimate the probability that the impact of ENSO on the variability is larger than the impact of human emissions to be 63% for the time period 2003-2016. If the time period is restricted to 2010-2016 this probability increases to 94%.”

However, only the fourth bullet contains any new science results, and this is minimal. It is generally accepted that natural variability in the tropics is the main driver of the atmospheric growth rate, and ENSO is the dominant source of tropical variability. As noted in the IPCC AR5, “the causes of the year-to-year variability observed in the annual atmospheric CO2 accumulation … are estimated with a medium to high confidence to be mainly driven by terrestrial processes occurring in tropical latitudes as inferred from atmospheric CO2 inversions and supported by ocean data and models.” Of course, there is a need for attribution studies to better understand the processes driving interannual variability, but the simple analysis presented in this manuscript does not represent substantial new knowledge. It was suggested by Referee #2 that the authors consider publishing in Atmospheric Measurement Techniques (AMT), and I would agree with that suggestion. Indeed, the first three bullets describing the new knowledge in the manuscript suggest that the manuscript would be better suited for AMT. In its current form, I believe that the manuscript would be a good, short AMT paper. If the authors insist on publishing in ACP, they need to significantly expand the scope of the growth rate analysis, and perhaps include a model to help with the attribution analysis.

Technical comment

Page 4, line 23: This line mentions the “(ii) the spatial variability of the XCO2 within the selected region.” What region? Is this referring to the analysis of the different latitude bands that was removed?

Hide

RR by Anonymous Referee #4 (14 Oct 2018)

Suggestions for revision or reasons for rejection

Review of “Computation and analysis fo atmospheric carbon dioxide annual mean growth rate from satellite observation during 2003-2016” submitted for possible publication in ACP by M. Buchwitz et al.

This paper presents satellite observation data of the atmospheric Carbon Dioxide, derives an ad-hoc method to estimate a growth rate, and analyses the variability of the annual growth rate with respect to the anthropogenic (human emissions) and natural (ENSO) contributions.

This paper already went through two rounds of exchanges between the authors and two reviewers. I understand that the two reviewers have given-up and I have been asked to provide a “final” review on the revised version of the paper.

The two reviewers had rather consistent comments. One comment focused on the analysis of the growth rate per latitude band and its interpretation. This analysis has now been removed from the main part of the paper and moved into an Annex. One can therefore state that this concern is resolved. The other comment was on the poor significance of the paper. Although I can agree with the fact that the scientific content of the paper is limited, I do see a large fraction of the papers published with similar or even lower scientific content. The current trend in the scientific literature is unfortunate, but this particular paper should not be a first target in this context.
My opinion is then that the version of the paper that I have been asked to review can be published in ACP

I nevertheless take this opportunity to point a few things that apparently were not mentioned by the two original reviewers

• The last sentence of the abstract states what has been done in the paper concerning the variability of the growth rate but does not provide the conclusion that can be fairly easily stated. I recommend that the conclusion is given in the abstract
• Page 4, line 20-28, the authors describe how they compute an uncertainty on the growth rate. The method is far from rigorous and there is absolutely no argument why the uncertainty should be estimated as the average of three terms. It should be made clear that the value that is derived is only an indicator, but in no mean a proper uncertainty estimate.
• The annual growth rate that is computed is an annual average of monthly estimates that are themselves computed from the differences between the XCO2 averages over a year. Thus, the 2010 growth rate (for instance) involves measurements from July 2009 to July 2011. Conversely, the NOAA growth rate (based on surface measurements) that is used for an evaluation from they production uses measurements of January and December 2010. The time periods are different and this should be made very clear.
• When doing the comparison of the growth rate based on surface and satellite data (comment above) (page 5 line 9 and Figure 2), the author argues that the difference is linked to the difference vertical sampling (line 9). To substantiate that statement, they could use a the results from a global transport model and compare the annual growth rate at the surface from that of the full column. I am sure that several of the authors have such simulations available so that this simple check would be easy to achieve.
• Figure 3 very clearly shows that the anthropogenic emission variability is mostly linked to a trend whereas the El Nino signal is variable on much shorter time scales. Thus, the relative impacts of the anthropogenic and natural contributions will very much depend on the length of the time series. The shorter the time series, the smaller the anthropogenic variability is. It is then very much expected that the natural contribution to the variability of the growth rate gets larger for a shorter time series. This should be made clear in the manuscript and conclusions that are somewhat misleading in this respect

Hide

ED: Publish subject to minor revisions (review by editor) (05 Nov 2018) by Christoph Gerbig

AR by Michael Buchwitz on behalf of the Authors (15 Nov 2018) Author's response Manuscript

ED: Publish as is (29 Nov 2018) by Christoph Gerbig

AR by Michael Buchwitz on behalf of the Authors (30 Nov 2018)

Short summary

We present a new satellite data set of column-averaged mixing ratios of carbon dioxide (CO₂), which covers the time period 2003 to 2016. We used this data set to compute annual mean atmospheric CO₂ growth rates. We show that the growth rate is highest during 2015 and 2016 despite nearly constant CO₂ emissions from fossil fuel burning in recent years. The high growth rates are attributed to year 2015-2016 El Nino episodes. We present correlations with fossil fuel emissions and ENSO indices.