This manuscript describes an analysis of ground-based Fourier Transform Infrared Spectrometer (EM27/SUN) observations in India from 2015-2016. The authors use the EM27/SUN data to analyze emissions of methane and carbon dioxide in Southern India. The authors show the utility of the EM27 measurements to evaluate satellite observations and then compare the observations to model results and estimate emissions. The paper is a very good analysis providing satellite data validation and emission estimates for a region of the globe that has not traditionally had extensive ground-based observations of methane and carbon dioxide.

The analysis is an important contribution in the area of using ground based and satellite observations of greenhouse gases in South Asia. They do a nice comparison of satellite data to the surface observations and then use them to help evaluate emission modeling analysis. The scientific quality is also very good, they use the well-established methods for using their surface data with the satellite observations and model results. The paper is well written, concise, and effective at communicating the key results.

My comments below are mostly minor and a few wording issues or typos. I think the conclusions could be strengthened somewhat to more clearly state the key findings.

Line 35: continuously (not continuous)

Line 38: OCO-2’s gets global coverage every 16 days (not 15)

Table 1, bottom row, right column: forgot word spectrometer after grating

Line 61: Feels like a reference would be appropriate here, maybe Laughner et al., 2024 https://doi.org/10.5194/essd-16-2197-2024

Line 86: Type for spelling of However

Line 87: delete “a few”

Line 90-91: I was not familiar with the Pathakoti (2024) study. Do the authors feel that the larger bias seen in that analysis was due to an earlier version of the OCO-2 data? The way it is phrased is confusing. Also when the authors discuss the results in Table 4, much improved biases for OCO-2, maybe refer back to the Pathakoti study and highlight reasons for the improved comparisons?

Lines 104-105: maybe the sentence beginning with “KIT has developed” could be reworded to make the point more clear, how the sun tracker works.

Lines 114-115: Mention the “reference IFS 125HR” is the Karlsruhe TCCON site?

Line 121: Typo, space needed before pyplot

Line 122: Type, add “The” before PROFFAST

Line 126: Define the acronym NCEP

Line 167: (NASA, USA) instead of (NASA), USA

Line 173: This is the first time you define the acronym OCO-2, maybe that should also be done with the first mention in the Introduction?

Line 184: Typo, misspelling artifacts

Line 188: OCO-2 uses different metrics for its bias correction. TCCON data is one of the data sets used, but models, small area analysis are also used.

Line 191: Typo, “contain a quality flag”

Line 198: Define acronym NILU

Line 218: Define acronym GAINS

Line 228: Define WetCHARTS?

Line 242: SWAMPS, missing S at the end

Line 274: When discussing the time coincidence criteria, I found the wording confusing. Maybe reword the sentence to make clear that you use data within two hours of the observation, but that data can come from a three-day time period.

Line 280: Could the authors speak to the importance of step number 3? Does it make a big difference in the bias and standard deviation results?

Table 4: Because the authors are stressing the importance of the bias and standard deviation calculations meeting the CCI requirements, you could highlight somehow (red numbers? Filled in cell?) the cases that do not match the requirements.

Line 364-365: Maybe instead of “increase or decrease” just use “changes” or “variations”

Line 369: “observed” could be removed

Line 376: Define acronym “JAMSTEC”

Line 382-383: Since the FLEXPART, maybe remove “the”

Lines 394-403: Maybe expand a bit more about the importance of the box sizes and the trade off between number of comparisons and improvement in bias and standard deviation. The GOSAT studies used larger boxes to improve the number of comparisons, the authors could expand on their conclusions relative to this.

Lines 404-408: Similarly, the conclusions relative to the model results and comparisons to the observation could be spelled out a little more clearly. What is the key message the authors would like to get across from that part of the analysis?

see PDF containing review.