Response to Reviewer 1

We thank the reviewer for their contributions to the manuscript. Our answer to each comment is addressed separately below.

Specific comments:

Lines 55-56: “The soil H2 sink is caused by microbial activity, both under aerobic and anaerobic condition” à clarify that the soil sink for atmospheric H2 (which is typically what term “soil sink” refers to) occurs under aerobic conditions. 

Text reworded:

The soil sink of atmospheric H₂ is caused by microbial activity, both under aerobic and anaerobic conditions (Piché-Choquette & Constant, 2019).

Lines 60-61: The referenced paper does not discuss “the high H2 demand of microbes”, rather I believe the implication of their findings is that the large capacity of soils for H2 uptake is primarily constrained by diffusive limitation.

The original text was poorly worded. We have removed the reference for clarity.

Lines 61-62: Clarify that such measurements exist, but are sparse/few in number (e.g., Conrad and Seiler, 1985; Yonemura et al., 1999, 2000a, 2000b; Smith-Downey et al., 2006, Smith-Downey 2008, Lallo et al., 2009, Khdhiri et al., 2015, etc.)

We have amended the text and included some of the higher relevancy studies the reviewer has provided.

Text changed to:

Studies investigating specific H₂ uptake rates for different soil types and conditions have been carried out but are sparse and limited to a small number of countries (primarily North America, Europe and Japan, e.g., Yonemura et al., 1999; Yonemura et al., 2000; Smith-Downey 2008, Lallo et al., 2009, Hammer, S., Levin, I., 2009; Khdhiri et al., 2015).

References added:

Yonemura, S., Kawashima, S., Tsuruta, H., 1999. Continuous measurements of CO and H2; deposition velocities onto an andisol: uptake control by soil moisture. Tellus B: Chemical and Physical Meteorology. https://doi.org/10.3402/tellusb.v51i3.16465

Yonemura, S., Kawashima, S., Tsuruta, H., 2000. Carbon monoxide, hydrogen, and methane uptake by soils in a temperate arable field and a forest. J. Geophys. Res. https://doi.org/10.1029/1999jd901156

Khdhiri, M., Hesse, L., Popa, M.E., Quiza, L., Lalonde, I., Meredith, L.K., Röckmann, T., Constant, P., 2015. Soil carbon content and relative abundance of high affinity H2-oxidizing bacteria predict atmospheric H2 soil uptake activity better than soil microbial community composition. Soil Biology and Biochemistry. https://doi.org/10.1016/j.soilbio.2015.02.030

Lines 68-71: The second part of this sentence does not seem tied in with the idea of temperature controlling enzyme activity.

The second part of the sentence is not tied to the first. To clarify we have reworded the text.

Temperature is another important factor as it determines the rate of microbial enzyme reactions, in addition a carbon source is also required for heterotrophic microbial activity…

Lines 70-72: This appears to be an inaccurate/incomplete discussion of the interaction between the soil sinks of CH4 and H2. The authors seem to be referencing H2 dynamics in methanogenic environments, as suggested by the cited reference. If such environments, H2 is not competing with CH4 as the energy source for microbes but is instead being consumed by methanogens to produce CH4 (i.e., CH4 is the product of metabolism, not a direct competitor for H2). This is likely not directly relevant to the atmospheric H2 sink, except insofar as H2 emissions from such environments could contribute to atmospheric H2 levels. In terms of H2 vs. CH2 specifically as energy sources with implications for the soil sinks, a more relevant discussion would be the interaction between high-affinity H2 oxidizers and methanotrophs in aerated soils.

To avoid confusion, we have removed the text and reference to Conrad et al., 1999.

Line 73: Specifically, spatial variation in microbial diversity.

Text changed to:

The biological sink of atmospheric H₂ has been suggested to be more sensitive to spatial variations of drivers (specifically microbial diversity) compared to the fluxes of other gases with high variability such as nitrous oxide (N₂O); however, H₂ measurement data are limited (Baril et al., 2022). 

Line 255-260: Looks like total carbon varies between 2.2-4% at the woodland site (Fig. S4)—does this really indicate high spatial variability? Which figure or table compares the variability in C and N at the chamber-scale vs at the plot scale?

The 95% C.I. of the mean soil C & N for each chamber location is shown in Figure S4 (samples taken in a 0.5 m radius around a chamber). The magnitude of the uncertainty in the mean for each individual chamber location is similar to the spread of all samples across the field site. Thus, spatial variability (or analytical noise) in the samples is too large to do meaningful comparisons of the impact of soil C and N on H₂ flux. While the spatial variability is relatively small in real terms, it is approximately the same magnitude (or greater) as any treatment effect so has a relatively large impact.

Text reworded to clarify:

Variability in C and N in the replicated cores per chamber was similar to the magnitude of spatial variability at the plot scale, suggesting that localised soil samples were not adequately representative of the soil within a chamber (high spatial variability of C and N observed in the soil at the <1 m² scale is similar to variability observed across the entire experimental area).

Line 295-296: Suggestion to clarify that the “increase” is referring to the difference between measured and predicted uptake rates. Also reference Table 4 (this should be called Table 3?).

Reworded to clarify:

Using regression to model (gap-fill) flux data, we estimate annual H₂ uptake of 3.1 ± 0.1 and 12.0 ± 0.4kg H₂ ha^-1 yr^-1 for the grassland and woodland sites, respectively, which increases the modelled mean uptake at the grassland site to 4.3 ± 0.2 nmol m^-2 s^-1 (in comparison to a measured mean flux of -2.6 ± 0.4 nmol m^-2 s^-1) while the expected mean uptake at the woodland site remains near 18 nmol m^-2 s^-1 (Table 3).

Line 303: Table 3 or 4?

Typo, corrected to Table 3.

Lines 347-350: Perhaps, it is worth emphasizing that while both soils have similar porosity values (based on Table 1), the difference in texture implies variations in pore size distribution and connectivity, leading to different sensitivities to moisture changes.

Text reworded:

While both soils have similar particle density, the difference in clay content implies variations in pore size distribution and connectivity which will likely lead to different sensitivities to moisture changes. We hypothesise from this assessment that the high fraction of clay soil at the grassland site results in the soil becoming highly anaerobic when moisture levels increase, as can be seen in the switching from CH₄ uptake to CH₄ emission when WFPS exceeded 40%.

Lines 350-351: Minor suggestion, but it may be helpful to rephrase this sentence to clarify that it’s the “well-drained” property of the soil that’s providing “ideal conditions” for H2 uptake, and not the litter lying on top. (The litter layer being thin is possibly contributing to this effect).

Our intention is to highlight that the forest litter is a potential source of microbial nutrients which may increase H2 uptake. It has been suggested that methanogens are active on plant surfaces in forests,  https://doi.org/10.1038/s41586-024-07592-w, but we will not go as far as to claim we have evidence of this for H₂ flux.

Text changed to clarify:

At the woodland site, a thin layer of organic materials (forest litter that could provide a source of labile carbon) lies on top of a sandy, well-drained soil, which may provide ideal conditions for H₂ uptake.

Figure S1: This figure is supposed to show the deposition velocity, but it looks identical to Fig. 2. The y-axis label also says “H2 flux” not “deposition velocity”

Yes, the wrong figure has been uploaded in the supplementary file which will be rectified.

Reviewer 2

We thank the reviewer for their contributions to the manuscript. Our answer to each comment is addressed separately below.

The authors analyze temperature and moisture effects on H₂ uptake as independent variables, as evidenced by Figure 3, which separately plots uptake against temperature (panel a) and moisture (panel b). This approach obscures the joint effects of these variables, as it is unclear what range of soil moisture corresponds to each temperature data point in panel b (likely contributing to the large scatter). Decades of experimental and modeling studies (since Conrad’s work in the 1980s) demonstrate that temperature and moisture act jointly on H₂ uptake. At least a contour plot showing uptake as a function of both variables (e.g., Smith-Downey et al., GRL, Figure 4) would clarify the interplay between them. This issue also extends to the polynomial fit presented in Figure 4, where the results imply additive effects of temperature and moisture (uptake = f(moisture) + g(temperature)). However, these variables typically interact multiplicatively (uptake = f(moisture) × g(temperature)). Incorporating this distinction is critical. While I acknowledge the authors’ decision not to use uptake models for their analysis, I recommend reviewing relevant studies (e.g., Smith-Downey 2006, Ehhalt et al. 2011, 2013, Tellus B; Bertagni et al. 2021, GBC) to inform the interpretation of their results.

A heatmap figure of temperature and soil moisture will be added to supplementary materials.

The initial rate used in the deposition velocity could be indicated as (dC/dt)_t=0.

(dC/dt)_t=0: it is defined in the text that dC/dt is the initial rate of change in concentration at t = 0

Text added to improve clarity by repeating the above in the description of the equation:

The references to Hutchinson and Mosier (1981) and HM/HMR models (acronyms not defined) seem unnecessary; this is a simple first-order closure for the rate.

Acronyms: References to HM method have been changed to HMR.

While the equations used by the Hutchinson and Mosier (1981) study are not unique, we feel providing some context to the R package that we use is relevant and useful for those who wish to follow our steps. It is perhaps a bit more detail than is required, but may help people who wish to carry out chamber work in future to understand the resources we have used.

 C_max is an unfortunate definition if it is the (minimum) value at equilibrium (proposing C_eq?).

Agreed.

The authors claim to report results as water-filled pore space (WFPS), defined as the volume of water per volume of pore space. However, the reported values (always below 0.5) suggest they may have used volumetric soil water content (volume of water per volume of soil). I support the Authors’ choice of using the WFPS, but I think those values should be divided by the porosity.

Additionally, it is unclear the range of soil moisture values encountered by the authors on the different days of measurements. How much of the spread of H2 uptake measurements in a single day of measurements is due to soil/soil moisture heterogeneity? I believe this critical aspect should be addressed, maybe with an additional graphical element.

Some graphs have used volumetric water instead of WFPS by accident. This will be rectified throughout the manuscript.

I would suggest defining the classes of the soils: in USDA, the two soils are clay (grassland) and sandy clay loam (forest). Soil texture is fundamental in the biogeophysical processes of H2 uptake, including moisture dynamics and soil H2 diffusion (e.g., Bertagni et al., 2021).

Further soil classifications will be added to the text.

In the keywords, I would suggest having hydrogen rather than methane and carbon

It is recommended that keywords are words that do not already appear in the title of the manuscript which aid search engines find relevant work.

Line 11. Although H2 might be present geologically, it is usually referred to as an energy carrier rather than a source.

Text changed to

“e.g. the use of H2 gas as a fuel”

Line 14. Within soils?

Corrected

Line 18. Nonlinear fit i.e., exponential?

Changed to exponential

Line 44. Add stratospheric to water vapor.

Stratospheric added

Line 54. Although both H2 sinks are uncertain, the soil sink is much more uncertain.

Text Changed to:

Whilst both removal mechanisms are highly uncertain (especially the soil sink), the fate and impact of increased H₂ emissions depends largely on the soil sink strength (Ehhalt & Rohrer, 2009).

Line 55. As I understood, most atmospheric H2 uptake would happen in aerobic conditions.

Yes, but can also occur in anaerobic conditions when H₂ is available.

Line 62-68. The authors are right that soil porosity and structure (together) greatly influence H2 uptake, especially regarding H2 diffusivity in the soil (Bertagni et al., 2021).

Reference included.

Lines 70-72. I am no microbiologist, so take this as a discussion comment, but I am unsure that H2 and CH4 are competing energy sources for soil microbes.

Paragraph lines 76. I would add that the only extensive H2 uptake dataset with coupled measurements of soil moisture and temperature is Meredith et al., 2016. Although at a lower temporal resolution, your measurements are a critical addition to what we have in terms of complete data (very little!).

Section 2.1 It would be very helpful to have a map, or at least indications, about the spatial extension of the campaigns. How far were the point measurements taken? This is somewhat related to the comment on moisture variability.

We will included a figure in the supplementary to show soil moisture at the scale of the measurement site.

Lines 208-210. Clarification is needed: is the deposition velocity negative for H2 emissions?

Individual measurements of positive flux were observed, resulting in what would be a negative deposition velocity. However, as this is leading to confusion, we will remove it for both sites.

Lines 208–210: Does the deposition velocity become negative for H₂ emissions?

See above. Further discussion of observed measurements in discussion section of manuscript.

Figure 2: Since these data are not equally distributed in time, you may consider using an x-axis that reflects the actual temporal spacing of measurements.

The data are too spaced out over 1 year, thus the figure is mostly empty if this approach is taken. We prefer to keep things as are to better observe the data presented.

Figure 3: This is, in my view, the most important figure in the paper, but it could benefit from significant improvement. Refer to my earlier comments on representing temperature and moisture interactions more effectively.

Suggested figures will be included in supplementary.

Lines 310–311: The phrase “While dealing with the exponential non-linearity of the rate of change” is wrong. While the concentration follows an exponential (and nonlinear) decay, the rate of change itself is linear, as discussed earlier.

Text will be re-worded to clarify the exponential nature of the H2 concentration changes throughout the text.

Lines 320–321: Consider adding “unknown microbial and geological processes” to account for the possibility of abiotic H₂ production pathways.

Text changed

Lines 366-367. Is this due to H2 leakage? This a very interesting aspect that you may want to highlight, given the interest in H2 containment/leakages.

Test added:

“due to leakage rates from the rubber septum materials used to cap vials”

Line 408: Besides soil aeration, I suggest discussing H₂ diffusion, as it plays a crucial role in the uptake process.

Diffusivity added to text.