Comment on amt-2021-395 Improved calibration procedures for the EM27/SUN spectrometers of the COllaborative Carbon Column Observing Network (COCCON) by Carlos Alberti et al.

General comments

Manuscript continues series of papers devoted to the development of the COCCON network which is based on EM27/SUN FTSs (Fourier transform spectrometer) observations. This type of FTS designed by KIT in close collaboration with Bruker Optics has a number of unique characteristics including portability, robustness, and ease of use. The combination of EM27/SUN FTS together with the state-of-the-art open-source codes (PREPROCESS and FROFFAST) designed at KIT for processing of interferograms and spectra allows provision the highest accuracy/precision values of atmospheric XCO2, XCH4, XH2O and XCO. Since 2014, EM27/SUN spectrometers have been successfully tested in various kinds of environments including a number of intensive field campaigns. The COCCON community is growing rapidly, therefore one of most important tasks is to develop tools to ensure rigorous QA/QC throughout the network. Namely, the manuscript focuses on the improved procedure of EM27/SUN calibration including the analysis of ILS (Instrumental Line Shape) by means of newly designed cell filled with C2H2-air mixture in comparison with standard procedure based on open path measurements of H2O spectral signatures.

The manuscript is well written nevertheless the current version mostly resembles a technical report. In conclusion, it would be useful to give a brief recommendation on how the COCOON community could (or should) implement the findings presented in paper in practice.

Specific comments

Abstract, lines 19-20: It is worth mentioning that new calibration cell is filled with air-C2H2 mixture.

Line 119: The distance "about 4 m" is mentioned, while Figure 1 caption says about "distanse of 4.20 m". How critical is the precise knowledge of the distance between lamp and the first mirror of EM27/SUN solar tracker?

Figure 1: In addition to the existing panels it would be helpful providing an extra panel with the side view on the set-up for open-path measurements.

Lines 131-132: “The spectrometer is now oriented in such a way that the cell can be conveniently located in the infrared beam on top of the spectrometer housing (see Figure 1 and Figure 4 A).”: This is not clear from the photos in Fig.1 and Fig.4A where/how the cell is placed.

Line 138-139: The term "instrument entrance" is not mentioned/presented in the list in Figure 2, so it is unclear which path (inaccessible) is being measured.

Fig.2: It would be helpful to indicate the position of an fine optical target in Fig.2.

Lines 193-195: “The latter window resides in the spectral overlap region covered by both detectors, allowing a check for a degraded ILS of the CO channel with respect to the primary channel, because in this spectral window the retrieval of ILS parameters can be performed from both main channel and CO channel spectra.” Could authors explain why ILS of the CO channel is degraded?

Line 264: “2.4 Error budget of the cell measurement for measuring ILS parameters of the EM27/SUN spectrometer” More detailed discussion of error budget is expected in section 2.4.

Fig.7: The blue dotted line in right panel does not correspond to y=x.

Technical corrections

Line 128: An excess colon.

Line 129: “In” should be written with a capital letter.

Line 132: An excess parenthesis after “ Figure 4 A)”.

Line 186: Please clarify the "ca." abbreviation.

Line 505: “Figure 19: ... (derived from high-resolution IFS125-LR spectra using GGG2014, red)...”. Is “IFS125-LR” correct here?

The network COCCON focusses on column measurements of greenhouse gases (GHGs). Sub-percent variations are of interest for the GHG columns. Hence the measurements need to resolve these small variations and small biases between different instruments could lead to erroneous conclusions. For this reason the calibration of the instruments is of high importance. The manuscript describes very detailed how the calibrations are performed and compares the new and old calibration methods on the technical level (e.g. modulation efficiency amplitudes). This is certainly very important for documenting the COCCON calibration and also for readers working with FTIR spectrometry. However, currently the paper does not discuss the impact of the improved calibration on the GHG retrievals. In my opinion it would be good to address the following questions: Does the improved calibration improve the consistency within COCCON and if yes, how much for each gas. I understand that there might be no improvement, because of the instrument specific correction factors. In that case it should be discussed how much the improved calibration procedure influence the instrument specific correction factors. Also it would be good to discuss if these correction factors could be impacted by atmospheric conditions (e.g. variations in water content) of by the environment the instrument is operated in (temperature, …) and if this has been investigated.