The authors present a non-commercial automated chamber system for measurement of soil greenhouse gas fluxes and compare its performance to manual chamber measurements. This is a nice set-up and experimental design presented and such data regarding the performance of automated chamber systems are relevant since automated systems start to become more widespread in greenhouse gas flux studies. However, there are some details and information in the manuscript about the automated chamber system that are either missing or not explained in such a way that I understood them. Also, despite the authors having generated a very good dataset, the discussion could benefit from more depth. I recommend major revision to strengthen this paper for publication.

General comments

The objective of the paper is to present an “innovative non-commercial soil GHG measurement system”. However, I did not understand what is exactly new about this system. The ‘Queensland design’ described has already been in use since 2000 and has been referenced e.g. in the “Nitrous oxide chamber methodology guidelines” by de Klein & Harvey (2015). Please highlight more the innovative updates to the system.

In the abstract and the last paragraph, the authors refer to Mediterranean conditions. I am not sure how important this is with regard to the chamber methodology itself. What are environmental conditions that the automated chambers still have to properly function in which are different from automated chamber studies from other regions? What is challenging about measuring soil GHG fluxes in Mediterranean conditions?

Lines 268 – 277: This discussion is a bit too vague for me. The authors have started some really good discussion points, but they didn’t really explore them further in depth with their data. Instead of saying that the chamber dimensions could explain flux differences, why not use the measurement data to further explore this subject, e.g. by calculating the Minimum Detectable Flux according to Christiansen et al. (2015) and Nickersen (2016). This would be especially interesting for CH₄ with fluxes fluctuating around zero. The MDF specifically includes chamber design and chamber closure time. It also includes the analytical precision of the gas analysis system. This is a point I am completely missing here. The authors compare a photoacoustic multi-gas analyzer and a gas chromatograph which differ significantly in their analytical precision. Does this significantly impact the results? A question regarding the effect of air-mixing: Could the fans have flushed out a bit of air from the soil pore system, thus contributing to the higher flux estimates?

Specific comments

Lines 29f.: I disagree with this sentence. It is possible to resolve short-term emission events with manual chambers. The common low sampling frequency is simply a result of the high labor-intensity of this method. You explain this point very well in the introduction.

Lines 39ff.: Maybe include some numbers in the abstract, e.g. how much higher were the measured fluxes.

Line 47: Use the latest IPCC assessment report from 2023. There the entire AFOLU sector is listed with 22 % contribution.

Lines 83 – 86: I would include two more aspects here: 1) As a result of the constraints lower spatial coverage compared to manual chambers, and 2) also more and more companies start selling automated chambers systems; it is becoming a market

Lines 176ff.: Despite the reference, it would be good to include information if a collar was used with the manual chambers and the insertion depth.

Lines 190ff.: Include the source for the climate data.

Lines 209ff.: There are some words mixed up/writing mistakes which make this paragraph a bit hard to read.

Line 345: Was there a specific reason for sampling at 6:00 GMT?

Lines 346f.: Wu et al. (2021) state 10:00 am. Do you consider 6:00 really close to that?

Lines 357f.: You don’t say anything about the costs in your manuscript. Compared to what is your system more affordable?

Section 2.5: I presume you used a linear fit in equation 1? The molar weights in the brackets are not correctly displayed. You have to write C-CO₂, C-CH₄, and N-N₂O. The R version used is the same as mentioned previously in the manuscript? Did you use any special R packages for the flux calculation or just the base packages?

Figure 3: In this figure it looks like the system is not a closed-loop system but the air drawn from the chambers is simply discarded to the atmosphere. Could this be one contributing factor to the higher fluxes measured with the automated chambers, e.g. pressure fluctuations despite the installed chamber vent? How long were the sampling lines and how long were your purge times?

Line 370: I don’t see how you can easily modify the number of chambers (also compared to other automated system). It is not only about the chamber number itself. What about power consumption, adjustment of the sampling protocol, length of tubing in the field, quick movability in case of field operations?

Figure 4: Is the sign for the p-value accidently flipped around or is it really >?

Figure 5: What are the cumulative soil gas emissions in this figure (sorry, maybe I just don’t get it for whatever reason; or are these average fluxes or do you mean scaling up from hourly to daily fluxes)? The color description in the caption is the wrong way around (also in Fig. 6). I am missing error bars. How certain are your flux estimates? I know error bars can make a plot unreadable, but at least include some information about the uncertainty range in the caption.

Figure 6: Why is CO₂ in Mg and the other gases in kg? There is one “soil” too many in the first sentence of the caption. What statistical test did you use?

Here the authors have compared measurements of soil gas fluxes using chambers and an automated chamber design based on the 'Queensland design''. Please see the uploaded pdf for grammatical/text suggestions. The manuscript is reasonably cler to read but would benefir from editing. The abstract contains repetition of ideas. As I began to read the manuscript I wondered what the new knowledge to be delivered to the reader would be. I'm not sure after finishing the text I have learnt anything new. Please emphasise what is novel with the study. Others have compared automated and manual chamber methods as noted in the text. Others have used automated results to help decide when to take manual chamber samples to reduce diel variation/bias in manual sampling (e.g. van der Weerden et al. 2013) - such an approach could be used here too. As a reader I want to know what method was more acccurate. Obviously we know automated methods provide greater frequency. How significant was the diel variation? What error might be incurred by taking manual samplease at any given time? What were the respective detection limits for the two methods? What other biases/benefits did the methods give? - e.g. automated chambers have clear walls allowing photosynthesis and plant function but chambers were opaque shutting down photosynthesis so implications for the CO2 flux and actually what the flux represents. Overall I think some more detail around the methodologies is clearly needed and the discussion should be better developed to clearly state what is novel about the manuscript.

Grace et al. 2020 Global Research Alliance N2O chamber methodology

guidelines: Considerations for automated flux measurement. J Environmental Quality

van der Weerden, T. J., Clough, T. J., & Styles, T. M. (2013). Using nearcontinuous measurements of N2O emission from urine-affected

soil to guide manual gas sampling regimes. New Zealand Journal of

1140 GRACE et al.

Agricultural