The article was written in a fairly good systematic manner by presenting Simultaneous measurements of greenhouse gases (CH4, CO2 and N2O) at atmospheric levels using a gas chromatography system. The article also presents and compares several result analyses from different samples using the GC System such as compressed air mixtures, direct air ambient, and standard gas mixtures, and also compares the GC measurement with another method from CRDS measurement. 

Remarks/Additional comments were written in the pdf file. Please see the attachment for the reviewer’s comments on this article.

General comments

This paper reports a method for analyzing three greenhouse gases simultaneously using a relatively simple GC system. The strength of this paper is that the developed method enables us to separate and quantify CH4, CO2, and N2O in air samples on a single column with a single detector. This technique would reduce sample size, time, and resources for the gas analyses. A shortcoming of this paper is that the precision or repeatability of the method is insufficient for atmospheric monitoring at background concentration level. In this regard, I think the title is misleading and should be revised to mean that the method is most suitable for source gases such as soil emission. Another concern is that the results of experiments for optimizing the GC setting are mainly discussed in the context of CV without further consideration of sensitivity of the detector. I believe combination of split ratio and sample size affect the amount/concentration of the target species delivered to the detector. For example, split ratio of 1 with 1-mL sample loop should give peak area that is equal to the area obtained at split ratio of 3 with 2-mL loop. In Table 1, I see results of CO2 and N2 are consistent with this idea, but it is not the case for other gases.

In summary, I recommend the publication of this paper after the authors address the issues above and specific points below.

Specific comments

L40–41. I think “the most modern techniques and devices” should be adopted only if they provide results with precision and accuracy that are sufficient for the purpose.

L46. I think a chapter from an e-book is referred here. Correct the citation information in the References section.

L53–55. I think mass spectrometer is one of the detectors used in gas chromatography. I cannot understand why the mass spectrometry is specifically noted.

L56–63. This paragraph is difficult to read because there is a lot of duplication. It seems that the authors try to list several types of “systems”, but they just mention detectors and GC columns.

L64. What does “natural wear” mean? Because the half-life of the radiation source Ni-63 is 100 years, I wonder other factors are meant.

L78. It is not clear what “and/or” means. In a certain case, both the two-column system and the single- column system are required?

L81–83. This sentence is difficult to understand. For example, what is compared using “as well as”?

L95. I recommend to add a schematic figure showing the developed system.

L101–102. “warmed” at what temperature?

L102–103. This means gas sample with high moisture (e.g., soil gas) might cause problems. Is the system designed for already dried samples or does it withstand moist samples?

L107–110. I guess the length and inner diameter of the columns are described here, but the splitting ratio of 1:5 cannot be achieved with the dimensions shown here. Add other parameters such as thickness of the inner coating of the columns. Also, combination of column and detector should be clearly described or shown using a figure. Is the TCD connected to the molecular sieve column?

L111. Quantitative information should be given instead of “extremely low baseline noise”.

L115–117. It is not clear at which position the discharge gas is added to the flow system.

L117. I cannot understand what this sentence means.

L121–122. Do the authors mean the final temperature of 200C is kept for 1 min? Revise the sentence.

L123–124. As described in the previous section, the flow after sample injection was divided into two columns. Are these flow parameters common to them?

L140. Decrease from what temperature?

Figures 1 and 2. Labels on the x and y axis are difficult to read.

L158. Does “vapour” mean water vapor? Please specify. Also, do the authors mean that the retention time of H2O peak shown in Figure 2 changes depending on the amount of water in the sample?

L165. It is not clear to me for what purpose the authors made experiments with different combinations of split ratio and sample size. I think the amount of sample (and water vapor) injected to the column is determined by the two parameters. For example, if 1 mL sample is processed with split ratio of 1, the amount of sample injected to the column is 1×1/(1+1) = 0.5 mL. If 2 mL sample is processed with split ratio of 3, the amount would be 2×1/(1+3) = 0.5 mL, which is the same as the first case.

L177 and elsewhere. Since TCD, FID, and ECD are acronyms of “xxx detector”, notation like “TCD detector” is awkward.

Table 1. It seems “SD” does not show the standard deviation of peak area, because dividing this value with “area” gives much smaller CV value. This is also the case for Tables 2 and 3. Please correct.

L185. Specify “the main atmospheric gases”.

L202. What does “reportable results” mean?

L212–214. If the authors use the system for soil gases, should the sample be dried before analysis?

Table 3. Is the area for N2O at split ratio of 3 correct? It is extremely higher than those obtained at split ratio of 4 and 5.

L234 split ratio of 4

L235–237. I cannot understand these sentences. Do the authors mean the “GC” results of CO2 obtained after 25/11/23 18:00 (Figure 3) have a systematic error due to a shift in sensitivity of the detector?

L243–245. Although it is not clear what “typical N2O measurements” means, 5% error is NOT satisfactory if one would like to know diurnal/seasonal cycle or secular trend of atmospheric N2O at background level.

L250–251. Since no TCD chromatogram is shown, this statement cannot be verified.

L261. Consider other quantitative or scientific expression for “very good”.

L268. water vapour?

I am afraid I have to reject this paper in its present form, simply because to my opinion the instrument in its present state lacks being fit-for-purpose in any application.

While the simultaneous detection of the three anthropogenic greenhouse gases using just one detector and only He is an improvement compared to previous GC systems, the precision and accuracy is much too poor for any application, most certainly for atmospheric measurements. Compare for instance to van der Laan et al (2009): A single gas chromatograph for accurate atmospheric mixing ratio measurements of CO2, CH4, N2O, SF6 and CO. Atmospheric Measurement Techniques 2, 549–559 (2009).  (not cited in the paper). They use two detectors, FID and EC, but they reach precisions : ±0.04 ppm for CO2, ±0.8 ppb for CH4, ±0.8 ppb for CO, ±0.3 ppb for N2O, and ±0.1 ppt for SF6, which  are within the reach of the WMO recommendations for atmospheric monitoring. The present instrument gets uncertainties more than two orders of magnitude worse!

If the authors have other applications in mind, soil air or agricultural greenhouse air for example (as several of the papers they cite, and which is hinted at by the values for their reference cylinder with strongly enhanced concentrations compared to ambient), then they should make that very clear in the paper. Furthermore, I think that even for these applications the present quality is not good enough.

The authors should compare their instrument, with a proper, regular calibration scheme in place (periodically, with at least two reference cylinders, preferably more) for a longer time, with a Picarro, or with  cylinder air input, and then show the uncertainty (in ppm / ppb), stability and linearity (!) of the system.