General comments

Peatlands are globally important carbon (C) sinks and storage through their significant uptake of   carbon dioxide (CO₂) from the atmosphere and accumulation of undecomposed plant material as peat. However, peatlands also emit CO₂ as well as another potent greenhouse gas methane (CH₄.). These gas flux dynamics in peatlands are regulated by many environmental variables, such as temperature and water level, that are impacted by the ongoing climate change, and thus there is an urgent need for better understanding of the C cycling of the peatlands and their climate feedback under the warming climate. More studies are especially still needed about spatio-temporal variation of the fluxes in different peatland ecosystems for more accurate climate modeling. This technical note introduces a new version of the previously presented novel measurement method called “skirt-chamber” that can be used for greenhouse gas flux measurements, even in remote locations. Compared to the commonly used dynamic chamber method, the skirt-chamber is more cost-effective and non-invasive as it does not require collars for air-sealed chamber closure. The new “modulated-light skirt-chamber” is specifically designed to determine photosynthesis-irradiance (PI) curves by measuring CO₂ flux rates under different light levels, to which it seems to fit well. The authors have polished the chamber design and carefully thought through the measurement set-up as well as the flux calculation method. The resulted PI-curves are showed to be comparable with previous models and eddy covariance measurements. It is also noted that the method of temperature measurements used in the current set-up still requires improvement. Thus, I find that the authors have done comprehensive work with improving their new chamber method for suiting net ecosystem exchange (NEE) measurements and are aware of the remaining limitations of their method.

The manuscript is well written, and its overall quality is good. The abstract is concise and details successfully the central background, methods and results. The introduction flows nicely and highlights the advantages and limitations of different GHG flux measurement methods. The aims of the study are clearly stated. Materials, especially the new version of the skirt-chamber design, as well as measurement methods and mathematical formulae for flux calculation are explained in detail. The authors have also tested several different shading steps to optimize their method regarding the use of different light intensity levels. Moreover, the PI curve generation is validated by comparing two different existing models. All the figures and tables are informative, clear, and include comprehensive captions. Furthermore, the supplementary material is well made giving further information about the chamber design, mass balance calculations, and result validation. I have only a few comments and questions regarding the measurement protocol, chamber design, and sampling. I recommend this manuscript to be accepted after minor revision. Please, see my more detailed comments below.

Our response:

We sincerely thank Reviewer 1 for the careful review and positive evaluation of our work. The comments provided are constructive and insightful, and we believe they will help us improve the quality of the manuscript in the revised version. We are particularly grateful that the reviewer evaluated both the current study and our previous publication (Thalasso et al., 2023; https://doi.org/10.5194/bg-20-3737-2023), thus providing an informed and coherent assessment of the skirt-chamber method and its development. Below, we provide detailed responses to the six specific comments raised. We are confident that the clarifications and forthcoming revisions will further enhance the clarity and robustness of the manuscript.

_________________________

Comment 1: Study site, campaign, and flux measurements, lines 140-148: In the description of the study site and flux measurements the studied bog is said to have different microforms from hummocks to bare peat surfaces without living Sphagnum cover, and that the measurements covered random locations across different vegetation covers and topographies. Did you also measure the bare peat surfaces or other relatively wet surfaces?

Even though the targeted bog is described not to be submerged, with the water table typically between 10 – 60 cm below surface, it seems that it could still have some wet spots. As I mentioned in my previous review of the author’s first technical note regarding the skirt-chamber, bare peat surfaces in bogs can especially be very wet or event water-saturated in my experience, which can make them tricky to measure with chambers. Even if not directly applicable to the studied bog here, I am still curious about how well the authors think the enhanced skirt-chamber would work for measurements in more wet peatland conditions?

As one of the main advantages of the skirt-chamber compared to traditional dynamic chamber is that it is does not need any collars, it could still be difficult to conduct the measurements without causing disturbance at the measured site without boardwalks and by placing down the chamber, especially when the water table is high.

Our response 1 – Measurement coverage across topography:

We confirm that the chamber was deployed across a representative range of topographic conditions, including hummocks, hollows, transitional zones, and with varying vegetation covers including bare peat surfaces. In response to this comment, we will include in the revised manuscript a new Table (as supplementary material) listing the 27 chamber deployment sites, along with a short description of the dominant vegetation or surface condition. A brief discussion will also be added to the main text to contextualize the diversity of sites sampled and the corresponding results observed.

Our response 2 – Applicability to wetter conditions and submerged sites:

The studied peatland section presented in this manuscript did not include any submerged area. In the revised manuscript, we will add a paragraph explicitly noting this limitation and discussing the expected performance of the chamber in submerged areas. We anticipate that in submerged or near-saturated zones, the chamber would form a seal with the peat surface, functioning similarly to a static closed chamber, with gas accumulation occurring—potentially without reaching a steady state. We agree with the reviewer that testing the chamber under such conditions would be an important future step, and we will emphasize this in the forthcoming revision of our manuscript.

Our response 3 – Site disturbance:

We agree that a potential limitation of the skirt-chamber is that it requires the presence of an operator, which may lead to pressure disturbances during deployment — especially in wet or water-saturated areas where pressure is transmitted more effectively through the peat matrix. In a separate and more recent study (not part of the current manuscript), we observed that operator proximity influenced the occurrence of ebullition: gas release was triggered when the operator stepped close to the chamber but was avoided when the operator maintained a maximum distance (40–50 cm) allowing chamber operation. In the revised manuscript, we will explicitly mention this drawback and recommend mitigation strategies such as the use of snowshoes (as we did) or pressure-distributing boards. This consideration will be included in the discussion of the method’s limitations.

_________________________

Comment 2: Related to the possible disturbances and difficulties when measuring greenhouse gas fluxes in wet and intermediate peatland microforms, did you detect any ebullition during the measurements this time?

Our response – Ebullition:

We did not detect any sudden increases in CH₄ or CO₂ concentrations that would indicate ebullition events during the present study. We attribute this observation to the fact that, as previously mentioned, all measurements reported in this manuscript were conducted in non-submerged areas, where the water table was below the peat surface. In such conditions, any gas bubbles formed in deeper layers are likely to be gradually released and diluted as they pass through the unsaturated peat and vegetation, reducing the likelihood of detectable ebullition at the surface.

_________________________

Comment 3: 2.3 Measurement protocol, lines 127-138: It is said that all PI curve determinations followed the same four-stepped protocol, with the third one being the chamber closure, during which two to four light conditions were tested. Do I understand correctly that the chamber was not ventilated between each light level, but the light conditions were altered during one chamber closure? If so, how often the chamber was ventilated? What are the benefits of not ventilating the chamber between each light level? What about the possible disadvantages or issues?

Was there any condensation in the chamber during a closure?

Our response 1 – Chamber closure and light modulation without ventilation:

Yes, indeed, the reviewer understood correctly: the chamber remained closed, i.e., not ventilated, for three to four minutes, during which two to four light conditions were tested. The main reason for not ventilating between each light level was to ensure accurate determination of the chamber gas residence time (θ_C; Section S2) after each chamber opening/closing, achieved by injecting a methane pulse and monitoring the chamber methane concentration over the longest possible time for improved precision. Since the methane injection did not interfere with the CO₂ concentration inside the chamber, θ_C could be determined precisely while applying different light conditions within the same closure phase. In our view, this approach increased the accuracy of θ_C determination and reduced the data processing effort by minimizing the number of θ_C determinations required. We will clarify this point explicitly in the forthcoming version of the manuscript.

Our response 2 – Condensation during measurements:

The reviewer raised a valid point. We did observe occasional condensation on the chamber window, but only under the highest light condition—i.e., when no fabric was used to reduce light intensity and under direct sun exposure. This condition was not always observed and never maintained for more than two minutes, in such manner that only slight condensation was observed during this short period. Although limited, the condensation may have slightly affected the incident light by (i) scattering direct irradiance into a more diffuse pattern and (ii) causing a minor reduction in overall transmission. However, because light intensity was measured inside the chamber, any potential effect of condensation on incident light was directly accounted for. Given the short duration and limited extent of the condensation, we consider its impact on the PI curve measurements to be negligible. We will make sure this is clearly addressed in the revised version of the manuscript.

_________________________

Comment 4: Did you also monitor the light intensity in real time in addition to the two light/temperature data loggers (so that you could see what PAR was at any given moment) or did you only filter through it later during the data processing? On the lines 189-190 it is said that 26% of the measurements failed, e.g., due to fluctuating or limited solar irradiance. This percentage is pretty high (1/4 of the measurements), and at least for the light it can easily be improved by monitoring PAR also in-situ so that the measurement can be stopped and started from the beginning if the light intensity changes too much during the chamber closure.

Our response – Real-time light monitoring:

We greatly appreciate the reviewer’s suggestion, which would indeed represent a valuable improvement to the current implementation of the skirt-chamber method for PI curve determination. As mentioned in the manuscript, light intensity was monitored using sensors placed inside the chamber, and the data were processed afterward during analysis. We fully agree that real-time PAR monitoring would provide the operator with immediate feedback on irradiance conditions, enabling the repetition of measurements in case of sudden fluctuations. We will include this point as a recommendation for future applications of the method in the revised version of the manuscript.

_________________________

Comment 5: The presented modulated-light skirt-chamber seems to work well for estimating PI-curves based on CO₂ measurements under different light levels. Despite the different base for flux calculation, modulated-light skirt-chamber method requires a gas analyzer for measuring CO₂ and a selected tracer gas concentrations. In the example study, CH₄ was selected as the tracer gas since it was also detected by the used gas analyzer. Nowadays there are quite many portable gas analyzers that commonly measure CO₂ and CH₄ at the same time. However, in the peatland gas exchange studies it is often interesting to be able to measure both of these gas fluxes at the same time, which saves time and effort from the “old-traditioned” measurements, when CO₂ and CH₄ needed to be measured in separate campaigns in the absence of the modern gas analyzers. Are there some other gases that the authors would recommend as potential tracer gas for the measurements for the scientists who wish to also use the CH₄ data as it is and not to interfere with it by injecting CH₄ into the chamber?

Our response – Other tracer gas:

The reviewer raised an important point regarding the possibility of using an alternative tracer gas, particularly when simultaneous measurement of CH₄ and CO₂ emissions is of interest. We acknowledge that this aspect was not clearly explained in the current manuscript. Importantly, the use of CH₄ as a tracer in our method does not prevent the determination of CH₄ emissions. In our previous study (Thalasso et al., 2023; https://doi.org/10.5194/bg-20-3737-2023), we showed that modifying light conditions for PI curve construction had no effect on CH₄ fluxes. Therefore, any segment of the CH₄ concentration data recorded prior to the CH₄ pulse injection can be used to calculate CH₄ emissions. In the present study, following each chamber closure, CH₄ concentrations were monitored for approximately one minute before the CH₄ pulse injection. This time window is sufficient to estimate CH₄ fluxes, although extending it by an additional 30 seconds in future applications would improve the accuracy of CH₄ flux estimation when CH₄ emissions are a core objective. It is also worth noting that each chamber closure provides an independent opportunity to estimate CH₄ fluxes, allowing for multiple emission values to be derived from a single chamber deployment. From our perspective, this approach is more practical than switching to an alternative tracer gas, which would inevitably increase cost and equipment weight, potentially compromising the portability of the method. We will clarify this point in the revised manuscript and include a recommendation for adjusting the protocol when simultaneous CH₄ and CO₂ flux measurement is desired.

_________________________

Comment 6: Have you been considering the possibility to add a cooling system to the skirt-chamber? In my experience, temperature in the chamber can increase significantly already in a couple of minutes especially in sunny weather with high light intensity, which alters the conditions for the gas fluxes that are targeted with the measurements. However, many chamber systems do not have a cooling system to regulate the possible warming effect of the chamber. Did the temperature in the skirt-chamber change during the chamber closures?

Our response – Impact on temperature, cooling system

We fully agree with the reviewer’s concern regarding potential temperature increases inside the chamber during closure. In the current manuscript, we acknowledge (around L285) that one limitation of our study is the use of a suboptimal temperature sensor inside the chamber. As a result, we are unable to provide a reliable assessment of temperature dynamics during chamber closure. In response to this comment, we will expand the discussion in the revised manuscript to include the importance of using a more appropriate temperature sensor for future applications. We will also discuss possible strategies to limit temperature increases, including the use of active cooling systems such as Peltier elements, as previously suggested by Jentzsch et al. (2024; https://doi.org/10.5194/bg-21-3761-2024). These improvements would enhance the ability of the skirt-chamber to maintain near-ambient conditions during measurements. We will make sure this is clearly addressed in the revised version of the manuscript.

General comments

Chamber-based measurements have long been used to determine greenhouse gas (GHG) fluxes in peatlands. As globally important carbon stores that are sensitive to environmental conditions such as air/soil temperature, they are susceptible to being turned from carbon sinks into carbon sources when environmental conditions change (e.g. as a result of climate change or anthropogenic land-use conversion). In near-natural peatlands carbon dioxide (CO2) and methane (CH4) are the dominant GHG that decide over the GHG balance of the ecosystem. Because of constraints of traditional chamber designs, such as the (semi-)permanent installation of collars into the soil as a measurement base, GHG measurements with chambers in remote areas remain limited. The installation of collars into the soil also constitutes a spatial constraint for measurements and can potentially create pseudo-replication and, thus, misleading statistical results. The proposed novel “modulated-light skirt chamber” builds on the previously introduced “skirt-chamber”. Both chamber-types allow to minimise invasion into the ecosystem and can relatively easily be deployed in remote locations as they do not require the installation of collars into the soil ahead of GHG flux measurement. The new chamber and study design outlined in this technical note improves on the initial skirt-chamber design in that it tries to minimize shading inside the chamber and allows to build photosynthesis-irradiance curves. The design concepts for the chamber and flux calculation methods used are well-described in the manuscript. The authors also compare the results produced with their new chamber-design to values calculated via the well-established eddy covariance method for the same peatland, as well as values from other peatland studies for validation. They outline remaining limitations within their study, such as an unsuitable temperature sensor chosen in their measurement set-up.

Overall, the manuscript is well-written and easy to follow. It successfully outlines the thought-process behind the new chamber design and which calculation procedures were used, with additional details provided in the supplementary materials. The abstract summarizes the study well. The materials and methods section provides ample information about the chamber design, measurement protocol (including different levels of shading) and mathematical background for flux calculations. It also describes the PI curve models used in this study in sufficient detail. I have some remarks about the introduction and the strengths and weaknesses of the method within the results and discussion section in particular. While these parts of the manuscript are equally well-written as the rest of the manuscript, I do wonder if there is too much of a focus on the new chamber design filling a gap between the ecosystem-level eddy covariance and leaf scale measurements. Chamber-based studies at the intermediate level with the more commonly used dynamic chamber method are fairly frequent in peatland ecosystems (e.g. in Canada and Europe). There might not be so much of a scale-gap that needs to be bridged, rather than existing chamber methods that have limitations and can be improved upon. I am aware that the authors focused a bit more on this aspect in their previous manuscript when they introduced the skirt-chamber, so it might be a good idea to at least make a mention of this facet and reference the previous manuscript here. My second concern is that the introduction section focuses on peatlands in general, but I would argue that the current chamber design is better suited to being used in sphagnum-dominated bog ecosystems rather than fen ecosystems which can have taller vegetation that might limit the use of the newly developped chamber and make sealing the chamber-base to the ground more difficult. Even though I would like to see these two points addressed, I do recommend the manuscript for publication after minor revision. Please see below for my specific comments.

Our response:

We thank Reviewer 2 for the kind evaluation and thoughtful feedback on our manuscript. Below, we address each of the two general comments and ten specific comments, outlining how we intend to revise the manuscript in response. We are confident that these planned improvements will further strengthen the manuscript.

Our response to general comment 1: Scale-gap and chamber method:

We fully agree with this comment regarding the actual benefits of the skirt-chamber for PI curve determination. After a fresh reading of our manuscript, we also recognized that our initial presentation may have overstated the existence of a scale gap, rather than focusing on the specific methodological advances of our chamber design. In the revised manuscript, we will present the three main approaches currently used (Eddy Covariance, leaf-level measurements, and chamber methods), highlighting their respective advantages. We will then clarify that the skirt-chamber is a refinement of our previous design—offering specific improvements over standard collar chambers—rather than a solution to a scale gap.

Our response to general comment 2: Chamber suitability for bog versus fen ecosystems:

We thank the reviewer for highlighting this important distinction. We will clarify in the revised introduction that the chamber is better suited for Sphagnum-dominated bogs than for fens and we will address this also in the subsection “Strengths and weaknesses of the method.”

_______________

1. 2.1 Modulated-light skirt chamber concept, line 74: as mentioned above, chambers are already not only used to measure soil but also ecosystem fluxes, so the latter should be mentioned here as well

Our response – Chambers for ecosystem studies: We will update the text at line 74 to accurately reflect the current use of chambers in ecosystem studies.

_______________

1. 2.1 Modulated-light skirt chamber concept: While the new chamber design allows for very flexible spatial deployment anywhere in the peatland, the lack of platforms around the measurement site might lead to unwanted effluxes of soil gases while putting the chambers on the ground but also if someone walks around the chamber during measurements. Is wearing snowshoes (as described in their previous technical note) enough of a preventative measure to avoid this?

Our response – Ebullition and site disturbance: We acknowledge this important point, which was also raised by Reviewer 1 (See AC1; first comment). In the revised manuscript, we will explicitly mention the use of snowshoes and other mitigation measures. This limitation and practical guidance will be included in the discussion of the method.

_______________

1. 2.2 Modulated-light skirt chamber design: I wonder if the authors can elaborate a bit more on which advantages they think their chamber design has over more traditional cylindrical or cubical chambers that are made fully of transparent PVC. It seems to me that their chamber has more of a risk of creating unwanted shading within the chamber if not deployed properly. With their new design it seems to me that any type of chamber could be used on top of their skirt-base, so why not use a more traditional design that has been tried and tested?

Our response – Chamber design: Reviewer 2 raises a legitimate question here. In the revised manuscript, we will discuss the main advantages of our chamber design, such as the circular section (which allows the chamber to be rotated without disturbing the base) and the truncated cone shape (which permits selection between shaded and direct light). We will also mention that, based on our experience, any chamber design—including traditional transparent models—can probably be effectively used, as long as it is equipped with a skirt-base that allows easy closing, opening, and aeration.

_______________

1. 2.2 Modulated-light skirt chamber design: What is the author’s reasoning to not use a sensor that measures PAR directly and rather choose a sensor that measures lux and needs to be calibrated against a PAR sensor? Would it be possible to include the results of this calibration in the supplementary materials in order to better be able to assess the quality of the light intensity measurements?

Our response – Lux vs PAR sensors: This is an important point. Our choice of a lux-based sensor (HOBO MX2202) was primarily guided by availability, but also by its compact size, durability, self-logging capability, and ease of deployment within the chamber. In the revised manuscript, we will provide the HOBO MX2202 calibration data collected over 60 hours of continuous operation, including the calibration equation and relevant statistics, in the supplementary materials.

_______________

1. 2.4 Study site, campaign, and flux measurements, line 141: should either read “-54.940° N, -67.644° E” or “54.940° S, 67.644° W”

Our response – Errors in coordinates: We thank the reviewer for pointing out this error. In the revised manuscript, we will correct to the standard notation: “54.940° S, 67.644° W.”

_______________

1. 2.6 Data treatment and statistical analysis: is there any particular reason Michaelis-Menten-type rectangular hyperbolic function proposed by Falge et al (2001) was not tested for creating the PI curves? Originally proposed as a gap filling strategy for eddy covariance data it has also been used to model the relationship between GPP and PAR in many chamber-based European peatland studies.

Our response – Monod or Michaelis-Menten type function: There is no substantive difference between the Monod and Michaelis-Menten hyperbolic functions, which are mathematically equivalent. The choice of terminology (Monod vs. Michaelis-Menten) mainly reflects convention rather than differences in data treatment or model fitting. We agree that it would be important to reference the widely cited Falge et al. (2001) study, and we will include this citation in the revised manuscript to acknowledge its relevance.

_______________

1. 3.1 PI curves: Labelling of the x-scale seems to be missing for figure 2, having labels would be beneficial to better be able to visualize temporal information regarding chamber closure times/ventilation periods described in the text.

Our response – Error in Figure 2 label: We thank Reviewer 2 for this helpful observation. Indeed, the x-axis label was inadvertently omitted in Figure 2, and we will correct this in the revised manuscript.

_______________

1. 3.3 PI curves and model parameters, lines 245-247: I expect there should be more than two studies out there reporting these values for peatland ecosystems, so it would be nice to include the results of a few more studies to solidify the comparison.

Our response – Literature comparison: The reviewer’s suggestion to include a broader range of studies is appreciated. Expanding the comparison will enhance the contextualization of our results. In the revised manuscript, we will incorporate additional reported values of the initial slope from peatland ecosystems across diverse geographic regions and peatland types.

_______________

1. 3.3 PI curves and model parameters, lines 261ff: Respiration rates should get their own subsection within the results and discussion section as they fall neither under the categories of PI curves or model parameters in my opinion; alternatively, the section header should be changed to include respiration rates.

Our response – Respiration: We fully agree that ecosystem respiration is distinct from PI curve parameters and warrants separate discussion. Accordingly, we will include a new subsection titled “Ecosystem respiration and net ecosystem emissions” in the revised manuscript.

_______________

1. 3.4 Strengths and weaknesses of the method: Even when relying on natural light, it should be possible to optimize PI curves by planning campaigns well (e.g. try to measure at different times of the day in cloudless conditions). I do understand that having limited field time in a remote location might not allow to always conduct measurements in ideal, cloudless weather. It would therefore be beneficial to monitor irradiance levels while taking measurements so that measurements can immediately be repeated in case irradiance was too changeable during a measurement.

Our response – Planning and real time monitoring: We appreciate this important suggestion, which aligns with a similar comment from Reviewer 1 (See AC1; comment 4). As noted in the manuscript, light intensity was monitored inside the chamber and processed post-measurement. We agree that real-time PAR monitoring would be beneficial by providing immediate feedback to operators and allowing measurement repetition during unstable irradiance. We will highlight this as a recommendation for future applications of the skirt-chamber method in the revised manuscript.