Exploring Gaps between Bottom-Up and Top-Down Emission Estimates Based on Uncertainties in Multiple Emission Inventories: A Case Study on CH4 Emissions in China
Round 1
Reviewer 1 Report
My biggest concern with this manuscript is that it overstates the goal of this work. Authors claim in the title, key words, and introduction that they are trying to use inverse modeling results to study the discrepancy between top-down and bottom-up methods. However, neither inverse modeling was performed by authors themselves, nor results from other inverse modeling studies were cited to make any analysis. Authors only compared three existing bottom-up methane emission inventories. I do not see how to ‘link’ bottom-up and top-down without doing any analysis about top-down.
It seems that authors ignored the uncertainties associated with the top-down method. By stating that the bottom-up inventories ‘overestimate’ (line 14, 16, 589) the methane emissions, the top-down results ‘indicates’ or ‘implies’ the uncertainties of bottom-up (line 13, 52,59), and the bottom-up should be ‘improved’ (line 60), authors assume that the top-down method is the benchmark and provides the correct methane emissions, which is for sure not the case. Top-down analysis often conducts inverse modeling using observation data from satellite, flight, and/or tall tower as well as a chemical transport model (CTM). The validation and bias correction of the satellite data could be highly uncertain due to the sparse coverage of ground measurements. CTM itself also has uncertainties. And another big source of uncertainty of top-down is the source attribution. It usually assumes that the source attribution of the posteriori emissions is the same as a priori, which is hard to validate and highly uncertain. Therefore, the discrepancy between top-down and bottom-up is potentially caused by uncertainties of both methods. And we can only say there is a difference between top-down and bottom-up but cannot make a judgement that which one is right and which one is wrong.
Authors show that three bottom-up inventories, PENG, EDGAR, and GAINS, yield different methane emissions. However, it is not clear that what we can learn from it. It is not convincing that the difference of these three inventories can be directly translated into uncertainties of the bottom-up method. Bottom-up estimations usually survey only a subpart of target sources for a limited amount of time and assume that all target sources always have the same emission factor as the surveyed sources. Therefore, a major source uncertainty of the bottom-up method lies in its insufficiency of capturing the spatial and temporal variability of target sources. For this study, are emission factors (EF) reported by these three bottom-up inventories able to represent all EF in China reported in the literature? Are they sufficient to represent the temporal and spatial variabilities of the methane EF in China? How did they come up with the EF when they were developing these bottom-up inventories? Is the difference of EF used in these three inventories more like a systematical difference caused by different measurement methods, or it is the real variabilities of methane EF? Did these three inventories report any uncertainties of their own EF and activity data?
The difference of three bottom-up emission inventory also provides few useful information on explaining the discrepancies between bottom-up and top-down. How does the methane emissions difference of these three bottom-up inventories compare with the difference suggested by the comparison of bottom-up and top-down? Even if they might be comparable, it could be a coincidence since the top-down method itself is highly uncertain as discussed above.
Therefore, if authors want to explain the difference between bottom-up and top-down results, they need to reconstruct the study to take into considerations of uncertainties associated with both top-down and bottom-up. If authors still plan to just compare tree existing bottom-up inventories, they need to carefully rethink about the purpose and value of the study and make a proper statement in the introduction.
Author Response
Dear editor and reviewer,
We are very grateful for the reviewer on the thoughtful comments provided by editor. Please kindly see the attached document for our response which providing in point by point as the suggestion.
Regards,
Penwadee
Author Response File: 
Author Response.doc
Reviewer 2 Report
This paper analyses the three major sources of CH4 emissions in China in the period 1990-2010, which were rice cultivation, coal mining and livestock. Three CH4 emission inventories were used, PENG, GAINS and EDGAR, where a detailed contribution of sources is presented. Moreover, an uncertainty analysis was also conducted. A key point of this study is the spatial analysis, since the calculation of emission factors for sectors of rice cultivation and coal mining is presented in the regions of China, where the contrasts are noticeable. However, some minor changes should be introduced previously to its publication. At the end of the Introduction, a short description of the CH4 inventories is introduced, l. 75-83. However, this text should be placed in Section 2.1, since it corresponds to Materials and methods. Moreover, the last sentence, l. 85-86 could be considered as a result and should not be placed in the Introduction.
Additional remarks
L. 98, from “This inventory indicates …” to l. 104. This is a result.
L. 115, from “This data set…” to l. 123. This is a result.
L. 136, from “According to the inventory data…” to l. 144. This is a result.
These results should be placed at the beginning of the Results.
Section 2.2, examples of variables xi and y should be presented. Moreover, some additional explanation about the uncertainty analysis would be acknowledged.
L. 162, is r the number of trajectories or the number of input variables?
L. 293, “cultivtion” or “cultivation”?
L. 298, “condtions” or “conditions”?
L. 315, zone AEX8 does not appear in Fig. 4.
L. 328, the uncertainty, about 85-106%, should be valued. May it be considered as normal? Could it be reduced?
L. 496, “manaure” or “manure”?
Are webs appearing in references 7, 8 and 9 currently available?
Author Response
Dear editor/reviewer
Please kindly see the attached document for the responses to the reviewer's comment.
Best regards,
Penwadee
Author Response File: Author Response.doc
Reviewer 3 Report
The manuscript submitted by the authors presents with high level of detail three CH4 emission inventories for China, i.e. PENG, GAINS and EDGAR, and provide very useful information on the activity data, emission factors and sources of information in each case. The authors present the uncertainties behind each inventory, which is very useful for users of those emission inventories. I would suggest modifying the title of the manuscript. The current title leads to the reader towards inverse modelling which is not evaluated in this study. I would suggest a title focusing on the work that they have carried out, emphasizing on the evaluation of these three emission inventories and the uncertainty analysis.
Specific suggestions.
Line 175: it says “CH4 emissions from China during the years 1990 to 2010 tended to show a continuous rising trend from a range of 30 Tg to 47 Tg …”. I suggest to indicate that this refers to total CH4 emissions, as in the figures are disaggregated in subsectors and this values are not visible.
Line 178: it says “EDGAR showed the lowest annual growth rate, the annual amount was significantly higher, i.e., it was 35% and 30% above that of PENG and GAINS, respectively”. I suggest referring to table 2 where the total emissions are reported, as this is not easily visible from the figure.
Figure 3, caption: I suggested including the meaning of AD, EF and All within the caption and no above. Similarly for Figure 4, Figure 5,
Line 464, Section 3.1, it says “Uncertainty of CH4 emissions….” I would not called this section “uncertainty” but “assessment of CH4 emissions …”
Line 359, it says “This temporal is unrepresented for ….”. I suggest saying “This temporal distribution…”
Line 496, it says “manaure”, I assume it refers to “manure”, correct in that case.
In equation 5 and in the text, it says “CH4utilzed”, I assume the authors intended to say “utilized”.
Author Response
Dear editor/reviewer
Thank you for the thoughtful comments and kindly suggestion. Please kindly see the uploaded document for the response to the reviewer's comment.
Best regards,
Penwadee.
Author Response File: Author Response.doc
Round 2
Reviewer 1 Report
I have no further comment.
