Physiological control on carbon isotope fractionation in marine phytoplankton

Brandenburg, Karen M.; Rost, Björn; Van de Waal, Dedmer B.; Hoins, Mirja; Sluijs, Appy

doi:https://doi.org/10.5194/bg-2022-68

Preprints

https://doi.org/10.5194/bg-2022-68

Preprints

17 Mar 2022

Status: a revised version of this preprint was accepted for the journal BG and is expected to appear here in due course.

Physiological control on carbon isotope fractionation in marine phytoplankton

Karen M. Brandenburg¹, Björn Rost^2,3, Dedmer B. Van de Waal⁴, Mirja Hoins^1,2, and Appy Sluijs¹ Karen M. Brandenburg et al.

Show author details

¹Department of Earth Sciences, Faculty of Geosciences, Utrecht University, Princetonlaan 8a, 3584 CB Utrecht, the Netherlands
²Department of Marine Biogeoscience, Alfred Wegener Institute (AWI), Helmholtz Centre for Polar and Marine Research, Am Handelshafen 12, 27570 Bremerhaven, Germany
³Faculty of Biology/Chemistry, University of Bremen, Leobener Strasse, 28359 Bremen, Germany
⁴Department of Aquatic Ecology, Netherlands Institute of Ecology (NIOO-KNAW), Droevendaalsesteeg 10, 6708 PB Wageningen, the Netherlands

Received: 11 Mar 2022 – Discussion started: 17 Mar 2022

Abstract. One of the great challenges in biogeochemical research over the past half a century has been to quantify and understand the mechanisms underlying stable carbon isotope fractionation (ε_p) in phytoplankton in response to changing pCO₂. Partly, this interest is grounded in the use of fossil photosynthetic organism remains as a proxy for past atmospheric CO₂ concentrations. Phytoplankton organic carbon is depleted in ¹³C compared to its source because of kinetic fractionation by the enzyme RubisCO during photosynthetic carbon fixation, as well as through physiological pathways upstream of RubisCO. Moreover, other factors such as nutrient limitation, variations in light regime as well as phytoplankton culturing systems and inorganic carbon manipulation approaches may confound the influence of CO₂ on ε_p. Here, based on experimental data compiled from the literature, we assess which underlying physiological processes cause the observed differences in ε_p for various phytoplankton groups in response to C-demand/C-supply and test potential confounding factors. Culturing approaches and methods of carbonate chemistry manipulation were found to best explain the differences in ε_p between studies, although daylength was an important predictor for ε_p in haptophytes. Extrapolating results from culturing experiments to natural environments and for proxy applications therefore requires caution, and it should be carefully considered whether culture methods and experimental conditions are representative of natural environments.

Download & links

Preprint (PDF, 1064 KB)

Supplement (554 KB)

Karen M. Brandenburg et al.

Status: closed

RC1:
'Comment on bg-2022-68', Anonymous Referee #1, 06 Apr 2022

Brandenburg et al. synthesize the available data on stable carbon isotope fractionation in phytoplankton to test how this parameter (Ep) is controlled by CO2 and other environmental parameters. This is a very well written and interesting paper and the data collection/analysis/interpretation seem very sound. I have hardly any comments, although I must admit that my knowledge on isotope fractionation is a bit rusted as I haven’t followed the literature for a couple of years.

The key message of the paper is almost frustrating, nevertheless important. From my stand-point this paper requires only very minor revisions (but I hope the other Reviewer is more up to date on the topic than I am).

Minor comments:

Figure 2: I found the unit of the C-demand/C-supply a bit strange. Wouldn’t it be easier to keep the unit as for the individual components and put them in brackets i.e. (C-demand unit)/ (C-supply unit)? Just a suggestion to facilitate understanding what this parameter means.

Line 165: “…prevents diffusion of CO2 but is permeable for HCO3-…” This surprised me. Are is CO2 or HCO3 mixed up, perhaps? Just double-checking.

The supplementary material could be moved to the main text. I don’t see a reason to bury it there.

The figures are very well designed and informative.

Citation: https://doi.org/10.5194/bg-2022-68-RC1
- AC1: 'Reply on RC1', Karen Brandenburg, 21 Apr 2022
  
  Brandenburg et al. synthesize the available data on stable carbon isotope fractionation in phytoplankton to test how this parameter (Ep) is controlled by CO2 and other environmental parameters. This is a very well written and interesting paper and the data collection/analysis/interpretation seem very sound. I have hardly any comments, although I must admit that my knowledge on isotope fractionation is a bit rusted as I haven’t followed the literature for a couple of years.
  The key message of the paper is almost frustrating, nevertheless important. From my stand-point this paper requires only very minor revisions (but I hope the other Reviewer is more up to date on the topic than I am).
  We thank the reviewer for his/her kind words. In terms of proxy applications, the key message of our analysis is indeed discouraging. Below you will find our answers to the reviewer’s comments and suggestions.
  Minor comments:
  Figure 2: I found the unit of the C-demand/C-supply a bit strange. Wouldn’t it be easier to keep the unit as for the individual components and put them in brackets i.e. (C-demand unit)/ (C-supply unit)? Just a suggestion to facilitate understanding what this parameter means.
  We agree with the reviewer that the units may be clearer, both in the figures and also in the text. Whenever we use POC production/CO2 concentration, we now consistently use POC production/[CO2]. We also made figure legends clearer, by typing the unit between brackets (e.g. ([pg C cell^-1 d^-1] / [µmol kg^-1]) like the reviewer suggested.
  
  Line 165: “…prevents diffusion of CO2 but is permeable for HCO3-…” This surprised me. Are is CO2 or HCO3 mixed up, perhaps? Just double-checking.
  No they are not mixed up. Cyanobacterial photosynthesis takes place in the carboxysome, and to allow sufficient build-up of CO₂ around RubisCO ensuring effective carboxylation, the membrane of this compartment is not permeable to CO₂. It is for HCO₃^-, which upon diffusion into the carboxysome is converted to CO₂. See also the references (i.e. Espie and Kimber, 2011).
  
  The supplementary material could be moved to the main text. I don’t see a reason to bury it there.
  As we would like to keep to main text focused and condensed, we would prefer to keep the supplement as is.
  
  The figures are very well designed and informative.
  Thanks!
  
  Citation: https://doi.org/10.5194/bg-2022-68-AC1
RC2:
'Comment on bg-2022-68', Anonymous Referee #2, 12 Apr 2022

The manuscript by Brandenburg et al. presents a compilation of stable carbon isotope fractionation data in phytoplankton experiments grown under various culture conditions (day length, nutrient availability, temperatures..) and teases out their contribution to the theoretically straightforward and expected demand/supply relationship on ep. The manuscript is an important contribution to the field, very well written and the data is presented nicely. I have therefore only a few comments and suggestions.

General comments:

1) The authors should clarify the statistical approach. If I understood correctly, their linear models predicting ep had three factors, i.e. POC production/CO2, one influential condition (light, irradiance,.....), and species. While the influential condition and species factors have categorical or discrete factor levels, POC production/CO2 has not. Is that something the lmer function in R can handle? I was under the impression that all levels would need to be categorical or distinct (not a continuum without groups), as it is basically an ANOVA. Please clarify.

2) The authors have chosen to test POC production/CO2 as the main driving factor for ep (please see also comment 1). From a pale-reconstruction perspective, that would require estimating two physiological parameters, i.e. POC per cell and instantaneous growth rate, to infer ep. What about the more simple growth rate/CO2 approach? The authors could test if they come to the same conclusions. I reckon they would but better to check.

3) Again, from a reconstruction perspective, the authors could calculate how much explanatory power a multiple linear regression approach would generate. Of course, some of the factors would not work as being categorical (unless a generalised linear model would be used instead), but some could be retained (e.g. light, temperature) or changed over (nutrient concentration, e.g. nitrate as being a proxy for the degree of limitation). That could be done group-specific, and looking at the simple linear regression presented in Figure 2, I could imagine that it would be quite a success.

Specific comments:

1) L 241: either 'these systems' or 'this system'.

Citation: https://doi.org/10.5194/bg-2022-68-RC2
- AC2: 'Reply on RC2', Karen Brandenburg, 21 Apr 2022
  
  The manuscript by Brandenburg et al. presents a compilation of stable carbon isotope fractionation data in phytoplankton experiments grown under various culture conditions (day length, nutrient availability, temperatures..) and teases out their contribution to the theoretically straightforward and expected demand/supply relationship on ep. The manuscript is an important contribution to the field, very well written and the data is presented nicely. I have therefore only a few comments and suggestions.
  We thank the reviewer for his/her kind words and will address the comments and suggestions below.
  General comments:
  1) The authors should clarify the statistical approach. If I understood correctly, their linear models predicting ep had three factors, i.e. POC production/CO2, one influential condition (light, irradiance,.....), and species. While the influential condition and species factors have categorical or discrete factor levels, POC production/CO2 has not. Is that something the lmer function in R can handle? I was under the impression that all levels would need to be categorical or distinct (not a continuum without groups), as it is basically an ANOVA. Please clarify.
  The reviewer is correct that we used three predictor factors in our models (namely POC production/CO₂, one influential condition, and species). Lmer has no problem with fitting both continuous and discrete data as fixed predictor variables, as illustrated in the examples from Bates et al. (2015; https://cran.r-project.org/web/packages/lme4/vignettes/lmer.pdf). What the reviewer maybe refers to is that it does not make sense to use a continuous variable for generating a random intercept in R. However, we used only the discrete variable “species” to provide a random intercept, and no continuous variables, so our model structure is correct. Here is also the R syntax for clarification: lmer(Ep ~ POCproduction/CO2 + [influential factor] + POCproduction/CO2: [influential factor] + (1|Species), data). To make this more clear, we now specify that we used a random intercept also in the text (L112).
  2) The authors have chosen to test POC production/CO2 as the main driving factor for ep (please see also comment 1). From a pale-reconstruction perspective, that would require estimating two physiological parameters, i.e. POC per cell and instantaneous growth rate, to infer ep. What about the more simple growth rate/CO2 approach? The authors could test if they come to the same conclusions. I reckon they would but better to check.
  The reviewer is right that this more simple approach would be easier to apply from a paleo-reconstruction perspective. This is why we now include this analysis on instantaneous growth rate/CO2 in the supplementary (ε_p versus µ_i /[CO₂], Fig. S2 and S3) and also mention it in the main text (L126-130; 159-161). While these data confirm our conclusion, these figures clearly stress the need to also make estimations for cellular POC contents in the paleo-domain, as especially for haptophytes this makes a big difference with regard to the explanatory power.
  3) Again, from a reconstruction perspective, the authors could calculate how much explanatory power a multiple linear regression approach would generate. Of course, some of the factors would not work as being categorical (unless a generalised linear model would be used instead), but some could be retained (e.g. light, temperature) or changed over (nutrient concentration, e.g. nitrate as being a proxy for the degree of limitation). That could be done group-specific, and looking at the simple linear regression presented in Figure 2, I could imagine that it would be quite a success.
  We thank the reviewer for his/her suggestion, and tested how much explanatory power we could generate with a multiple linear regression approach using different environmental variables (L114-116). As explained above, this approach can use both continuous and discrete data. We found that the inclusion of the light regime and whether there was nutrient limitation yielded highest explanatory power in all groups, and mention these findings now in the results (L131-133) and in the discussion (L156-157; 280-281).
  
  Specific comments:
  
  1) L 241: either 'these systems' or 'this system'.
  Thanks for noting this mistake. We changed it accordingly.
  
  Citation: https://doi.org/10.5194/bg-2022-68-AC2

Status: closed

RC1:
'Comment on bg-2022-68', Anonymous Referee #1, 06 Apr 2022

Brandenburg et al. synthesize the available data on stable carbon isotope fractionation in phytoplankton to test how this parameter (Ep) is controlled by CO2 and other environmental parameters. This is a very well written and interesting paper and the data collection/analysis/interpretation seem very sound. I have hardly any comments, although I must admit that my knowledge on isotope fractionation is a bit rusted as I haven’t followed the literature for a couple of years.

The key message of the paper is almost frustrating, nevertheless important. From my stand-point this paper requires only very minor revisions (but I hope the other Reviewer is more up to date on the topic than I am).

Minor comments:

Figure 2: I found the unit of the C-demand/C-supply a bit strange. Wouldn’t it be easier to keep the unit as for the individual components and put them in brackets i.e. (C-demand unit)/ (C-supply unit)? Just a suggestion to facilitate understanding what this parameter means.

Line 165: “…prevents diffusion of CO2 but is permeable for HCO3-…” This surprised me. Are is CO2 or HCO3 mixed up, perhaps? Just double-checking.

The supplementary material could be moved to the main text. I don’t see a reason to bury it there.

The figures are very well designed and informative.

Citation: https://doi.org/10.5194/bg-2022-68-RC1
- AC1: 'Reply on RC1', Karen Brandenburg, 21 Apr 2022
  
  Brandenburg et al. synthesize the available data on stable carbon isotope fractionation in phytoplankton to test how this parameter (Ep) is controlled by CO2 and other environmental parameters. This is a very well written and interesting paper and the data collection/analysis/interpretation seem very sound. I have hardly any comments, although I must admit that my knowledge on isotope fractionation is a bit rusted as I haven’t followed the literature for a couple of years.
  The key message of the paper is almost frustrating, nevertheless important. From my stand-point this paper requires only very minor revisions (but I hope the other Reviewer is more up to date on the topic than I am).
  We thank the reviewer for his/her kind words. In terms of proxy applications, the key message of our analysis is indeed discouraging. Below you will find our answers to the reviewer’s comments and suggestions.
  Minor comments:
  Figure 2: I found the unit of the C-demand/C-supply a bit strange. Wouldn’t it be easier to keep the unit as for the individual components and put them in brackets i.e. (C-demand unit)/ (C-supply unit)? Just a suggestion to facilitate understanding what this parameter means.
  We agree with the reviewer that the units may be clearer, both in the figures and also in the text. Whenever we use POC production/CO2 concentration, we now consistently use POC production/[CO2]. We also made figure legends clearer, by typing the unit between brackets (e.g. ([pg C cell^-1 d^-1] / [µmol kg^-1]) like the reviewer suggested.
  
  Line 165: “…prevents diffusion of CO2 but is permeable for HCO3-…” This surprised me. Are is CO2 or HCO3 mixed up, perhaps? Just double-checking.
  No they are not mixed up. Cyanobacterial photosynthesis takes place in the carboxysome, and to allow sufficient build-up of CO₂ around RubisCO ensuring effective carboxylation, the membrane of this compartment is not permeable to CO₂. It is for HCO₃^-, which upon diffusion into the carboxysome is converted to CO₂. See also the references (i.e. Espie and Kimber, 2011).
  
  The supplementary material could be moved to the main text. I don’t see a reason to bury it there.
  As we would like to keep to main text focused and condensed, we would prefer to keep the supplement as is.
  
  The figures are very well designed and informative.
  Thanks!
  
  Citation: https://doi.org/10.5194/bg-2022-68-AC1
RC2:
'Comment on bg-2022-68', Anonymous Referee #2, 12 Apr 2022

The manuscript by Brandenburg et al. presents a compilation of stable carbon isotope fractionation data in phytoplankton experiments grown under various culture conditions (day length, nutrient availability, temperatures..) and teases out their contribution to the theoretically straightforward and expected demand/supply relationship on ep. The manuscript is an important contribution to the field, very well written and the data is presented nicely. I have therefore only a few comments and suggestions.

General comments:

1) The authors should clarify the statistical approach. If I understood correctly, their linear models predicting ep had three factors, i.e. POC production/CO2, one influential condition (light, irradiance,.....), and species. While the influential condition and species factors have categorical or discrete factor levels, POC production/CO2 has not. Is that something the lmer function in R can handle? I was under the impression that all levels would need to be categorical or distinct (not a continuum without groups), as it is basically an ANOVA. Please clarify.

2) The authors have chosen to test POC production/CO2 as the main driving factor for ep (please see also comment 1). From a pale-reconstruction perspective, that would require estimating two physiological parameters, i.e. POC per cell and instantaneous growth rate, to infer ep. What about the more simple growth rate/CO2 approach? The authors could test if they come to the same conclusions. I reckon they would but better to check.

3) Again, from a reconstruction perspective, the authors could calculate how much explanatory power a multiple linear regression approach would generate. Of course, some of the factors would not work as being categorical (unless a generalised linear model would be used instead), but some could be retained (e.g. light, temperature) or changed over (nutrient concentration, e.g. nitrate as being a proxy for the degree of limitation). That could be done group-specific, and looking at the simple linear regression presented in Figure 2, I could imagine that it would be quite a success.

Specific comments:

1) L 241: either 'these systems' or 'this system'.

Citation: https://doi.org/10.5194/bg-2022-68-RC2
- AC2: 'Reply on RC2', Karen Brandenburg, 21 Apr 2022
  
  The manuscript by Brandenburg et al. presents a compilation of stable carbon isotope fractionation data in phytoplankton experiments grown under various culture conditions (day length, nutrient availability, temperatures..) and teases out their contribution to the theoretically straightforward and expected demand/supply relationship on ep. The manuscript is an important contribution to the field, very well written and the data is presented nicely. I have therefore only a few comments and suggestions.
  We thank the reviewer for his/her kind words and will address the comments and suggestions below.
  General comments:
  1) The authors should clarify the statistical approach. If I understood correctly, their linear models predicting ep had three factors, i.e. POC production/CO2, one influential condition (light, irradiance,.....), and species. While the influential condition and species factors have categorical or discrete factor levels, POC production/CO2 has not. Is that something the lmer function in R can handle? I was under the impression that all levels would need to be categorical or distinct (not a continuum without groups), as it is basically an ANOVA. Please clarify.
  The reviewer is correct that we used three predictor factors in our models (namely POC production/CO₂, one influential condition, and species). Lmer has no problem with fitting both continuous and discrete data as fixed predictor variables, as illustrated in the examples from Bates et al. (2015; https://cran.r-project.org/web/packages/lme4/vignettes/lmer.pdf). What the reviewer maybe refers to is that it does not make sense to use a continuous variable for generating a random intercept in R. However, we used only the discrete variable “species” to provide a random intercept, and no continuous variables, so our model structure is correct. Here is also the R syntax for clarification: lmer(Ep ~ POCproduction/CO2 + [influential factor] + POCproduction/CO2: [influential factor] + (1|Species), data). To make this more clear, we now specify that we used a random intercept also in the text (L112).
  2) The authors have chosen to test POC production/CO2 as the main driving factor for ep (please see also comment 1). From a pale-reconstruction perspective, that would require estimating two physiological parameters, i.e. POC per cell and instantaneous growth rate, to infer ep. What about the more simple growth rate/CO2 approach? The authors could test if they come to the same conclusions. I reckon they would but better to check.
  The reviewer is right that this more simple approach would be easier to apply from a paleo-reconstruction perspective. This is why we now include this analysis on instantaneous growth rate/CO2 in the supplementary (ε_p versus µ_i /[CO₂], Fig. S2 and S3) and also mention it in the main text (L126-130; 159-161). While these data confirm our conclusion, these figures clearly stress the need to also make estimations for cellular POC contents in the paleo-domain, as especially for haptophytes this makes a big difference with regard to the explanatory power.
  3) Again, from a reconstruction perspective, the authors could calculate how much explanatory power a multiple linear regression approach would generate. Of course, some of the factors would not work as being categorical (unless a generalised linear model would be used instead), but some could be retained (e.g. light, temperature) or changed over (nutrient concentration, e.g. nitrate as being a proxy for the degree of limitation). That could be done group-specific, and looking at the simple linear regression presented in Figure 2, I could imagine that it would be quite a success.
  We thank the reviewer for his/her suggestion, and tested how much explanatory power we could generate with a multiple linear regression approach using different environmental variables (L114-116). As explained above, this approach can use both continuous and discrete data. We found that the inclusion of the light regime and whether there was nutrient limitation yielded highest explanatory power in all groups, and mention these findings now in the results (L131-133) and in the discussion (L156-157; 280-281).
  
  Specific comments:
  
  1) L 241: either 'these systems' or 'this system'.
  Thanks for noting this mistake. We changed it accordingly.
  
  Citation: https://doi.org/10.5194/bg-2022-68-AC2

Karen M. Brandenburg et al.

Supplement

https://doi.org/10.5194/bg-2022-68-supplement

Karen M. Brandenburg et al.

Viewed

Total article views: 484 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	Supplement	BibTeX	EndNote
362	105	17	484	20	8	10

HTML: 362
PDF: 105
XML: 17
Total: 484
Supplement: 20
BibTeX: 8
EndNote: 10

Views and downloads (calculated since 17 Mar 2022)

Month	HTML	PDF	XML	Total
Mar 2022	181	52	6	239
Apr 2022	98	30	9	137
May 2022	63	19	1	83
Jun 2022	19	4	1	24
Jul 2022	1	0	1

Cumulative views and downloads (calculated since 17 Mar 2022)

Month	HTML	PDF	XML	Total
Mar 2022	181	52	6	239
Apr 2022	98	30	9	137
May 2022	63	19	1	83
Jun 2022	19	4	1	24
Jul 2022	1	0	1

Viewed (geographical distribution)

Total article views: 453 (including HTML, PDF, and XML) Thereof 453 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 02 Jul 2022

Short summary

Reconstructions of past CO₂ concentrations rely on proxy estimates, with one line of proxies relying on the CO₂-dependence of stable carbon isotope fractionation in marine phytoplankton. Culturing experiments provide insight into which processes may impact this. We found, however, that the methods with which these culturing experiments are performed also influence ¹³C fractionation. Caution should therefore be taken when extrapolating results from these experiments to proxy applications.


Total:	0
HTML:	0
PDF:	0
XML:	0