the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Technical note: Resolving Vertical CO₂ Concentration Gradients at the Air-Water Interface Using a Novel Membrane Equilibration Technique
Abstract. Quantifying diffusive greenhouse gas emissions from inland waters often relies on the concentration gradient between dissolved gases in the water and their equilibrium concentration in the atmosphere. However, while gas exchange occurs directly at the surface, dissolved CO2 is mostly measured further below the surface, potentially introducing bias. Recent studies have shown the presence of vertical CO₂ gradients during calm conditions, which can lead to systematic errors in flux estimations when measurements are not taken at the correct depth. To address this issue, we developed a novel Floating Membrane Equilibrator (FME)—a thin (0.6 cm), flexible device equipped with gas-permeable silicone tubing arranged in a flat plane with a theoretical vertical resolution of 1 cm. By directing airflow through the tubes, for instance using an infrared gas analyzer, CO2 in the water equilibrates with the gas phase inside the tubing. The resulting CO2 concentration reflects the local dissolved CO2 in the adjacent water layer. We tested the FME in a controlled pool experiment using CO2-supersaturated water. Two FMEs were deployed at 1 cm and 25 cm depth. Results show that the FME provides reliable equilibration with the surrounding water and delivers accurate CO2 measurements. Compared to conventional CO2 probes, the FME shows faster response times and higher temporal resolution, enabling detection of short-term fluctuations that are typically missed by standard sensors. Moreover, our measurements revealed a distinct vertical CO2 gradient in the pool, with higher concentrations at 25 cm depth compared to 1 cm, which is consistent with surface outgassing. This highlights the risk of misestimating fluxes when relying on deeper CO2 measurements. The FME is a valuable tool for resolving near-surface CO2 profiles with fine vertical resolution, thereby improving our understanding of the dynamics and drivers of lake-atmosphere gas exchange. Ultimately, the FME helps reduce uncertainty in CO2 flux estimates and supports the development of more accurate models for greenhouse gas emissions from aquatic systems.
- Preprint
(894 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2026-124', Anonymous Referee #1, 21 Jan 2026
- AC1: 'Reply on RC1', Patrick Aurich, 14 Apr 2026
-
RC2: 'Comment on egusphere-2026-124', Anonymous Referee #2, 11 Mar 2026
The authors Aurich et al. present a new design to measure vertically CO2 concentration (or pCO2) in lakes. The authors present a smart idea which addresses a critical issue in understanding CO2 gradients in near-surface layers of water bodies. However, I have some concerns with the design and practical use. For example, the authors claim in the text (line 69), that manual sample disturb the layer being sampled. However, the FME floating would also induce artificial disturbance in the upper 3 cm layer being sampled. I think that is not something which can be totally avoided, but needs to be addressed. The fact that FME can only be deployed at very calm conditions – as silicone tubes may get in contact with atmospheric CO2 – raise the question on how practical the technique is? What the good depth without running into the risk to “contaminate” data with atmospheric CO2? A more sophisticated design with adjustable depth deployment, pressure sensor and multiple FME (rather than two depths) is actually needed. The design presented here is rather simplified.
The authors did the experiments using CO2-supersaturated water (950 and 3000 ppm, and I am wondering how the FME performs under more natural saturation levels (~ 380-480 ppm)?
Line 44: Some of the literature refers to estimating k over the ocean. I suggest to explicitly say so, and make clear which ones refer to lake studies
Line 61: remove comma after reference
Line 65: Please re-phrase “safe depth”
Line 84: replace by “calm surface conditions in lakes”
Line 108: Provide more detaisl on the NIR analyzer in terms of performance (accuracy, calibration setup….)
Line 139: How was the AMT standard probe was calibrated, and what is a typical accuracy?
Line 145: How did the authors ensure that the aquarium pumps efficiently avoided the buildup of a gradient for the calculation of the error between both FME?
Line 149: Authors refer to the “same depths”, which strictly speaking was not the case as indicated in the following sentence by writing “proximity”.
Figure 3A: I do not understand why a concentration of 950 ppm was measured as in the method section a concentration of 3000 ppm was referred to. The authors should also explain how the access the “true” CO2 concentrations as reference values. I assume that pool of water was outgassed “naturally”, but a reference value would be beneficial to assess the “sampling efficiency”.
Figure 4b: Is the steep increase at 2cm between 6:00 and 9:00 real or artifact from pressure buildup? There seems to be also small increase by switching from 29 to 2cm, at the each beginning of the black line sections.
Figure 5: It would help readers to indicate in the plot when the aquarium pumps were turned on.
Line 180: Authors refer to a standard CO2 probe shown in Figure 2B, but that is not visible or clearly indicated.
Line 182: equilibrated faster to the in-situ concentration than what?
Figure 6. The authors should show data from FME at 25cm depth to compared with the standard CO2 probe.
Line 194: Use the acronym FME
Line 194; Measuring at two depths cannot be referred to as “high vertical resolution”. The paper represents two points with depth only.
Line 196: Replace “saw” with “observe”
Line 205: Optical observation seems to be critical, but not described in the method section. What means “wavy conditions”? Presence of capillary waves?
Line 210: I believe that the chamber technique quantifies the water-air flux, not the surface concentration. Please verify and correct if necessary
Line 221: The authors compare FME at 2cm with CO2 probe at 25 cm in terms of signals from the rain events. That is inappropriate because the impact on rain is highest at the surface – for obvious reason – and decrease with depth. I would not expect rain effects at 25 cm, but for a full assessment data on wind speed would be helpful.
Line 228: See previous comment. Comparison between FME and standard CO2 probe should be done only at the same depth.
Line 231: How fast? This should be given with a quantity.
Line 251; Have water traps being used in this study? If not, why? I also feel that deployments only at very calm conditions is a strong limitations, as CO2 gradients will change with increased wind-driven turbulence.
Citation: https://doi.org/10.5194/egusphere-2026-124-RC2 - AC2: 'Reply on RC2', Patrick Aurich, 14 Apr 2026
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 227 | 96 | 13 | 336 | 19 | 23 |
- HTML: 227
- PDF: 96
- XML: 13
- Total: 336
- BibTeX: 19
- EndNote: 23
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
My comments are in the attached .pdf