Article title: Determination of respiration and photosynthesis fractionation coefficients for atmospheric dioxygen inferred from a vegetation-soil-atmosphere analog of the terrestrial biosphere in closed chambers

The marine and terrestrial biosphere regulate the concentration of O₂ in the atmosphere through the processes of photosynthesis and respiration (both photorespiratory and mitochondrial). These processes also modify the oxygen isotope composition of O₂ (d¹⁷O and d¹⁸O) differently providing a promising tool to investigate variations in biosphere function in response to changes in climate. However, to use these tracers to resolve the global mass balance and interpret ice core data, it is necessary to understand how the various biosphere processes fractionate isotopically and how variable these isotope effects are for different components of the biosphere (e.g. marine vs terrestrial organisms). This need has been hampered somewhat, as measuring d¹⁷O and d¹⁸O variations represents a technical challenge and only a few labs in the world measure routinely the ¹⁷D of O₂.

Up until quite recently the scientific community assumed that during photosynthesis, there was no isotopic fractionation of the O₂ evolved based on some of the first studies by Stevens et al., (1975) and Guy et al., (1993) on enzymes, intact cells and spinach thylakoids. These experiments found that the d¹⁸O of O₂ was identical in isotope composition to the substrate water. Since then, further studies have been made using far more precise IRMS techniques on phytoplankton cultures/communities (Helman et al., 2005; Eisenstadt et al., 2010). Collectively, these studies and data from other ocean studies now indicate that during photosynthesis, marine organisms produce O₂ significantly enriched in ¹⁸O with respect to the seawater bathing the phytoplankton. Furthermore, the extent of ¹⁸O enrichment may vary depending on whether the photosynthetic organisms are eukaryotic or prokaryotic, with the former producing O₂ relatively more ¹⁸O-enriched than the latter. Although a number of datasets now exist for marine phototrophs, there are very few data for terrestrial plants. The first measurements on plants by Dole and Jenks, (1944) concluded that plants and algae caused an ¹⁸O enrichment of O₂, however, these results were challenged repeatedly by several later studies (Vinogradov et al., 1959; Guy et al., 1993; Helman et al., 2005; Luz & Barkan, 2005). The present study of Paul et al., under review suggests that grass plants (Festuca arundinacea) may actually produce ¹⁸O-enriched O₂ during photosynthesis. Given that the global flux of O₂ produced by land plants is significant, a photosynthetic fractionation of 3.7+/- 1.3 per mil would impact the overall value of the terrestrial Dole Effect and our estimates of terrestrial biosphere productivity during past glacial and interglacial periods using ice cores.

A further novelty of the present study is the independent measurements of ¹⁷D of O₂ consumed by soils in the dark and the ¹⁷D of O₂ produced and consumed by plants and soils exposed to day/night light cycles. These experiments demonstrated that soils incubated without plants did not exhibit any overall trend in ¹⁷D values over extended incubation periods of 1 to 2 months. In contrast, replicated incubations with both plants and soils exhibited an overall increase in ¹⁷D of ~80 ppm over the same timeframe demonstrating that the biosphere and especially the metabolism of plants has a strong impact on the ¹⁷D of the atmosphere in contrast to other abiotic processes that occur in the stratosphere that tend to decrease ¹⁷D signals. Collectively these experimental data may be useful to help parameterise models that are used to interpret variations in biosphere productivity from ice core records during glacial and inter-glacial periods.

Overall the data are of great interest to the scientific community. However, not all information is communicated for certain components of the study rendering it difficult for the reader to confirm unequivocally some of the important advances particularly those linked to the revised photosynthetic fractionation values for plants of 3.7 per mil. For example, one of the key variables to calculate 18epsilon photosynthesis is the oxygen isotope composition of leaf water (see eq 14). The authors explain that leaf samples were collected and IRMS measurements made to establish the d18O and d17O values but I could not find any reference to the values obtained or used in eq 14 or how this varied during the experiments and how stable the closed water irrigation values were during the experimental runs. It would be important that this information is provided where available in either Table 1 and/or Table 2. In addition, there seems to be some inconsistencies in the development of the 18epsilon calculations. Specifically, as written it is not clear how Eq 14 is simplified to Eq 18. Currently equation 18 has some issues with signs and a number of R’s are missing. Thus, it is not possible for the reader to calculate and check the conclusions related to 18epsilon photosynthesis as valuable data and definitions are not provided. I am sure everything is fine but for the moment it is just not transparent and requires communication.

There are also certain parts of the introduction and discussion that assume a certain level of reader prior knowledge and if this paper is to appeal to a wider audience a little more work on briefly explaining the key processes involved (Mehler reactions, COX versus AOX, photorespiration) and some biological explanations could be appreciated as well as how they may vary in importance between environmental conditions for example dark respiration in the dark vs in the light and also how dark respiration rates and isotope ratios may vary in soils with and without roots. The discussion could also benefit from summarizing the different phototrophs that have been measured in the past and how these vary and rather than stating that the new value is 3.7 perhaps the reality is that this value is somewhat variable across plant functional types and thus this parameter may require further investigation as hinted in the conclusion.

Specific comments

Please do a search in the manuscript to correct “leave” to “leaves” when referring to plant foliage.

Line 34 change to “one of the most important biogeochemical cycles on Earth as oxygen…”

Line 39 change to “variations of the low latitude”

Line 45 change to “knowledge of the various fractionation..”

Line 46 change to “interpreting the relationship between d18Oatm variations in ice core air with the low latitude water cycle”

Line 48 change to “multiple processes involved”

Ln 48 change to “had shown that d18Oatm”

Ln 51 change to “several fractionations caused by biosphere processes”

Ln 52 First measurements, there were some measurements before Guy albeit less precise but still very provocative and it would be good to summarise which organisms were measured by Guy et al and others.

Perhaps refer to the review of Tcherkez and Farquhar 2007 for a discussion on the theoretical aspects of the oxygen evolving complex.

Ln 54 perhaps mention the process either as water photolysis, water-splitting or photosynthetic water oxidation and refer to its location in photosystem II of the chloroplast

Line 56 change to “in the terrestrial biospherethe d18O of water split during photosynthesis…”

Ln 58 for higher latitudes/elevations please cite Dansgaard, 1964 and during leaf evaporation please cite Dongmann et al., 1974

Ln 59 change to “The mean d18O enrichment of leaf water isotopic composition has been estimated between…”

Ln 60 change to “composition of mean global ocean water”

Ln68  it is not clear to the reader the logic that connects the +6 per mil enrichment to the low latitude water cycle. In fact this latter part of the paragraph discussing past hydrology and d18O signals is not clearly presented and could benefit from being a separate paragraph after a clear explanation of the hydrological connections perhaps with the aid of a diagram explaining the budget fluxes, current understanding in the size and drivers and uncertainties.

Ln79 I would invert these processes and start with the MIF in the atmosphere the describe the MDF that is then followed logically by the definition for the MDF.

Ln 88 change to “increases”

Ln 103 change to “to the biological”

Ln 105 is the variability between COX and AOX the only possibility for soil fractionation? What about non-enzymatic weathering? Or decomposition of different substrates varying in oxidation level? Other enzymes linked to other biogeochemical cycles? Soil community composition? What about roots?

Ln 109 change to “a simplified vegetation-soil-atmosphere analog

Ln 110 change to “in a closed chamber of 120 dm3”

Fig 1 No light sensor in the drawing. What is the impact on the d18O2 if it equilibrates with water vapour in the glass flask? Would it not be prudent to have a drier on the fLask inlet? How did the irrigation water isotope composition vary between each experiment and during the experimental runs with and without plants?

Ln 123 dm3 is the preferred SI unit

Ln 125 change to “Microcosm”

Ln 135 change enlightenment to the explicit number hours in the dark and light expressed as a ratio.

Ln 138 change to approximately

Ln 139 invert to read mature fescue

Ln 143 change “one “ to “smaller cchamber”

Ln 151 how was the Oxy1-SMA O2 concentration calibrated?

Ln 155 invert to “closed water”

Ln 156 change to “was condensed”

Ln 162 provide model ref for the foil bags

Ln 164 change to “inflated or deflated”

Ln 164 change to “variations caused either by”

Ln 166 place a space before Almeno

Ln 178 please provide info on the flow rate

Ln 186 change to “immersed”

Ln187 change to “pressure was reached”

Ln 189 change to “vial for the water was then immersed”

Ln 189 remove the brackets and the text therein and replace “and” with “whilst”

Ln 190 change to “transferred to a water bath”

Ln 196 “an IRMS”

Ln 198 define D17O

Ln 201 “the isotopic composition”

Ln 217 please define dO2/Ar

Ln 220 “the isotopic”

Ln 228 “leaf dark respiration”

Ln 233 I would rearrange this sentence so that 2 weeks is before 23 days

Ln 236 maybe “characterise the” experimental repeatability

Ln 246 why no light dark cycle?

Ln 254 change during to for

Ln 255 controlled to ~ 400umol mol-1 by a…

Ln256 change to composition as this

Ln 257 change “enlightenment” to “light cycle” and provide day/night cycle in hrs here.

Ln 275 change subscripts to alphas not epsilon to be consistent with the equation that follows

Ln 283 “breathed”?  overall the notation throughout is difficult to follow and not intuitive

Ln 287 remove the phrase “evolution of the” if you really want to define n(O2) as evolution implies something that changes i.e. would require the definition of a flux

Eq 8 definition sign not intuitive

Eq 10 R’s should be deltas

Eq 12 perhaps worth pointing out which leaf water pool is likely most important but an assumption is made that it can represented by bulk leaf water signal.

Eq 322 maybe also important to note how the differences in dark respiration in the light and dark may differ.

Eq 18 this equation needs to be revised it is incorrect in its current form and is not consistent with the previous eq 14

Table 1 Strongly suggest a third column that provides information about all the values used or if they are variable and what the units are.

R*O in the table 1 of O2 in air?

Fig 2 x axis would be easier to follow if the Day # was provided instead of Date

Fig 2 would also be usefu to indicate the variation of the soil water d18O over time.

Ln 403 provide mean value plus SD

Ln 411 respiration not significantly different? Test

Ln 412 you cannot explain only speculate you did not measure this. Furthermore this should be in the discussion

Fig 4 legend not consistent with the axis purple is O2 not CO2

Ln 437-440 Again this is a bit of discussion not really results unless you actually compare with the leaf water data from the experiment that is not presented in the paper. Pease provide the leaf water information from the experiment.

Ln 444 Is this caused by a technical problem?

Ln 453 assuming that respiration rates or fractionation during the dark and light do not vary.

Eq 28 same as Eq18 and has problems with missing R’s

Ln 558 change to “photorespire”

Paul et al. describe a novel environmental chamber apparatus, as well as its first results, focused on obtaining isotopic fractionation factors associated with respiration and photosynthesis in terrestrial analogue systems. In this case, the authors report results from a study of a commercial potting soil and a grass (tall fescue). It is a difficult system to control and to study, and the authors have done perhaps the best job of controlling the environment compared to all the terrarium studies done over the past two decades in this vein (i.e., those led by Luz, Angert, and Yeung), namely by maintaining carbon dioxide concentrations and a closed water cycle with constant relative humidity below saturation. In this sense the study is quite welcome and I look forward to seeing more studies come of this apparatus. However, I have technical concerns about a couple elements of the manuscript, which are listed below:

1. When describing the mass balance equations for the experiment, the sign of dn sometimes does not make sense relative to the direction of the oxygen flux. For example, in 2.4.3, equation 13 describes dn_{photosynthesis} as “the number of molecules of O₂ produced by photosynthesis,” yet to have the correct sign I believe it needs to have a negative value in the equation (i.e., it is of opposite sign to dn_{total_respi}, which is a consumption term and has a positive value like in equation 8). I am not sure whether this confusion is an error in words only or if it propagates into the mass balance equations, but the authors should check.
2. Accurate isotopic scaling between VSMOW and air is taken as a matter of fact when there is a known discrepancy of order 0.1 per mil in both δ¹⁸O and δ¹⁷O differences between labs that measure the O₂ analyte together with Ar (e.g., Hebrew U., Princeton, the present study) and those who measure it as pure O₂ (U. New Mexico, Gottingen, Open University, UCLA, Rice U.). It may seem like a minor point, but Yeung et al. RCMS (2018) showed that inconsistent/assumed scaling can lead to spurious disagreements in discrimination factors and triple-isotope slopes in the range of 0.1 per mil and 0.005, respectively. It poses a problem for the soil-respiration γ value because this type of uncertainty is systematic and thus would not be included implicitly in the random errors; the reported uncertainty range is too small. Indeed, Stolper et al. GCA (2018) and Ash et al. ACS Earth Space Sci. (2020)  report evidence -- from two independent labs -- that dark respiration might not be characterized by the “canonical” 0.516 value. Many of the other reported uncertainty ranges are significantly larger than the level of these disagreements, but the photosynthetic endmember does depend strongly on the assumed value of VSMOW, which Wostbrock and others have shown are far from in agreement. I suggest the authors (1) acknowledge that this disagreement in the field exists, citing the relevant literature, and (2) make note of the possibility that the fractionation factors may need to be revised in the future once everyone gets on the same reference frame. I don’t necessarily believe that the reported values need revision per se, but the field would do well to acknowledge outstanding issues in papers rather than continue to ignore them.

Minor comments

1. There is some nonstandard notation: the use of R¹⁸O instead of the more common ¹⁸R when describing ¹⁸O/¹⁶O ratios; the use of γ without mention of its equivalence to the symbol θ used elsewhere in the triple-isotope literature; the use of “fractionation coefficient” rather than the more common term “fractionation factor” for α.
2. “Since” refers to a time in the past (e.g., since 1980) and “because” refers to a cause (“because Ar is an inert gas”). In most instances of “since” in the manuscript I think the authors should be using “because” instead.
3. In the abstract, Table 1, L531, and 546: “respiration of leave” --> “leaf respiration”
4. L451, 453, 458, and 460: “leave” --> “leaf”