Including filter-feeding gelatinous macrozooplankton in a global marine biogeochemical model: model-data comparison and impact on the ocean carbon cycle

Clerc, Corentin; Bopp, Laurent; Benedetti, Fabio; Vogt, Meike; Aumont, Olivier

doi:https://doi.org/10.5194/egusphere-2022-1282

Preprints

https://doi.org/10.5194/egusphere-2022-1282

Preprints

30 Nov 2022

| 30 Nov 2022

Including filter-feeding gelatinous macrozooplankton in a global marine biogeochemical model: model-data comparison and impact on the ocean carbon cycle

Corentin Clerc, Laurent Bopp, Fabio Benedetti, Meike Vogt, and Olivier Aumont

Abstract. Filter-feeding gelatinous macrozooplankton (FFGM), namely salps, pyrosomes and doliolids, are increasingly recognized as an essential component of the marine ecosystem. Unlike crustacean zooplankton (e.g., copepods) that feed on preys that are an order of magnitude smaller, filter-feeding allows FFGM to have access to a wider range of organisms, with predator over prey ratios as high as 10⁵: 1. In addition, most FFGM produce carcasses and/or fecal pellets that sink 10 times faster than those of copepods. This implies a rapid and efficient export of organic matter to depth. Even if these organisms represent <5 % of the overall planktonic biomass, the induced organic matter flux could be substantial. Here we present a first estimate of the influence of FFGM organisms on the export of particulate organic matter to the deep ocean based on the marine biogeochemical model NEMO-PISCES. In this new version of PISCES, two processes characterize FFGM: the preference for small organisms due to filter feeding, and the rapid sinking of carcasses and fecal pellets. To evaluate our modeled FFGM distribution, we compiled FFGM abundance observations into a monthly biomass climatology using a taxon-specific conversion. A model-observation comparison supports the model ability to quantify the global and large-scale patterns of FFGM biomass distribution, but reveals an urgent need to better understand the factors triggering the boom-and-bust FFGM dynamics before we can reproduce the observed spatio-temporal variability of FFGM. FFGM contribute strongly to carbon export at depth (0.4 Pg C yr^-1 at 1000 m), particularly in low-productivity region (up to 40 % of organic carbon export at 1000 m) where they dominate macrozooplankton by a factor of 2. The FFGM-induced export increases in importance with depth, with a simulated transfer efficiency close to one.

Received: 16 Nov 2022 – Discussion started: 30 Nov 2022

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this preprint. The responsibility to include appropriate place names lies with the authors.

Download & links

Preprint (PDF, 6787 KB)

Notice on discussion status
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
Preprint (6787 KB)

Download & links

The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.

Journal article(s) based on this preprint

28 Feb 2023

Including filter-feeding gelatinous macrozooplankton in a global marine biogeochemical model: model–data comparison and impact on the ocean carbon cycle

Corentin Clerc, Laurent Bopp, Fabio Benedetti, Meike Vogt, and Olivier Aumont

Biogeosciences, 20, 869–895, https://doi.org/10.5194/bg-20-869-2023,https://doi.org/10.5194/bg-20-869-2023, 2023

Short summary

Corentin Clerc, Laurent Bopp, Fabio Benedetti, Meike Vogt, and Olivier Aumont

Interactive discussion

Status: closed

RC1:
'Comment on egusphere-2022-1282', Rebecca Wright, 06 Jan 2023

I recommend the manuscript be accepted subject to minor revisions, which are detailed in the attached document.

Citation: https://doi.org/10.5194/egusphere-2022-1282-RC1
- AC1: 'Reply on RC1', Corentin Clerc, 17 Jan 2023
  
  The authors would like to thank reviewer #1 for her valuable and useful comments. The authors considered all suggestions and addressed the raised issues, which we believed increased the clarity of the revised manuscript. Below are given point-by-point answers to the comments.
  Reviewer comments are in bold, responses in normal font. Proposed changes to the manuscript are in italics.
  I think the authors have offered an interesting, thorough, and generally well-presented manuscript. The manuscript adds a valuable contribution to the growing body of research around modelling gelatinous zooplankton, further supporting the importance of gelatinous zooplankton in global carbon export while also highlighting key areas for future research.
  We would like to thank Reviewer #1 for her very supportive comments on our manuscript.
  SPECIFIC COMMENTS
  Consider using ‘Crustacean Macrozooplankton (CM)’, instead of ‘Generic Macrozooplankton (GM)’. As the generic macrozooplankton group is based entirely on crustacean macrozooplankton I think this would aid in the clarity and precision of the manuscript. Generic implies that the group represents organisms across the macrozooplankton i.e. more than just crustaceans.
  We agree with the reviewer that the definition of the generic macrozooplankton compartment (L136) was not precise enough, which may have lead to some confusion. Actually, our ‘generic macrozooplankton group’ intends to represent macrozooplankton that are not part of the Tunicata (“non-tunicate macrozooplankton”), which includes a lot of crustaceans but not only (e.g., pteropods). Since this compartment was parameterized using an allometric scaling relationship rather than group-specific parameterizations, we think that the group is generic enough to represent also non-crustacean types of macrozooplankton, such as pteropods. For this reason, the generic macrozooplankton notation is more appropriate, and the corresponding explanatory text in the Methods will be corrected as follows:
  L136: “GM, namely generic macrozooplankton, is intended to represent non-tunicate macrozooplankton, such as euphausids, pteropods or large copepods.» instead of “GM, namely generic macrozooplankton, is intended to represent crustacean macrozooplankton, such as euphausids or large copepods.»
  Also, L137, ”The parametrization is similar to that of mesozooplankton (Eq. 28 to 31 in Aumont et al. 2015)” will be replaced by “The parametrization is similar to that of mesozooplankton (Eq. 28 to 31 in Aumont et al. 2015) and parameter values have been derived using allometric scaling relationships (see section 2.2.1).”
  Line 151:
  
  Include an explanation on why the decision was made to have ‘feeding with identical preferences on both phytoplankton groups... as well as on microzooplankton’.
  We assumed non-selective predation when modelling filtration in FFGM, meaning that the organisms are able to ingest all filtered prey. This lack of food selectivity is a known characteristic of filter-feeders (Pauli. et al., 2021b). Recent studies suggest that the efficiency of prey capture by filter-feeding gelatinous organisms could depend on the prey size but also on their taxonomy (Sutherland et al., 2022). Yet, while “future work should bring more quantitative studies of grazing rates and selectivity under in situ conditions, expand the number of pelagic tunicate species that have been studied” (Sutherland et al., 2022), quantitative estimates remain insufficient to properly assess this phenomenon, therefore we believe that non-selective predation remains the best option in the current state of knowledge.
  To address the present comment, we propose to insert the following sentence at line 149 : “Although there is some recent evidence for selective feeding behavior in pelagic tunicates (Sutherland et al. 2022), the lack of quantitative study led to the simpler representation of FFGM as non-selective feeders (Pakhomov et al. 2002, Vargas et al. 2004, von Harbou et al. 2011). Therefore, we assume […]”
  New references :
  von Harbou, L., Dubischar, C. D., Pakhomov, E. A., Hunt, B. P., Hagen, W., and Bathmann, U. V.: Salps in the Lazarev Sea, Southern Ocean: I. Feeding dynamics, Marine Biology, 158, 2009–2026, 2011.
  Vargas, C. A. and Madin, L. P.: Zooplankton feeding ecology: clearance and ingestion rates of the salps Thalia democratica, Cyclosalpa affinis and Salpa cylindrica on naturally occurring particles in the Mid-Atlantic Bight, Journal of plankton research, 26, 827–833, 2004.
  
  Section 2.3: Sensitivity experiments
  
  Adding an introductory sentence to this section would help the flow and to guide the reader, i.e. ‘Five sensitivity experiments were carried out to...’.
  In accordance with the reviewer’s comment, the following introductory sentence will be added at line 190 : ‘Five sensitivity experiments were carried out to evaluate the sensitivity of the model to the chosen parameterization.”
  Rainbow colour palette for figures (fig. 3a, fig. 4, fig. 8, fig. A2e,f, fig. A3, fig. A4, fig. A5)
  
  I strongly recommend changing from a rainbow colour palette for figures, to a “perceptually uniform” palette which provides the same perceived colour change over the same change in value.
  In accordance with the suggestion by the reviewer, colour palette will be changed to a “perceptually uniform” palette in fig. 3a, fig. 4, fig. 8, fig. A2 e,f, fig. A3, fig. A4 and fig. A5. We will use the “cividis” colour palette from the matplotlib python package.
  Line 286:
  
  Include a brief mention of why 0-300m was chosen for analysis.
  The 0-300 m depth range was chosen for our analyses to ensure that we were capturing most of the organisms present in the epipelagic zone, which can be as deep as 250-300 m in the tropical gyres. This choice was applied for both macro- and mesozooplankton but there was an error in the manuscript indicating a 0-100 m depth range for macrozooplankton (L280). This typo was corrected.
  We justify this choice here with the AtlantECO dataset (see section 2.4.1) for FFGM species.
  First, when looking at the distribution of the maximal depth of sampling (see figure A2 in the attached supplement page 3), there is a clear cut off in the distribution at around 300 m, samples being sporadically distributed for deeper maximal depths. Spatial resolution of data below 300m is hence very coarse. Also, the depth distribution displayed below shows that using 100 m as a maximal depth would exclude a significant portion of the dataset.
  Second, the available data suggest that most of the biomass is present in the 300 m first meters. Indeed, when analyzing the raw AtlantECO dataset (i-e non gridded), the mean biomass for FFGM is 4.93 mg C m^-3 and the median 0.31 mg C m^-3 for the data with a maximum sampling depth shallower than 300 m. These metrics drop to 0.04 mg C m^-3 for the mean and 0.0 mg C m^-3 for the median when focusing at the data with a minimum sampling depth deeper than 300 m.
  Last, when focusing on the median by maximum depth bins (red points in the attached supplement page 3), we show that the median is null from 300 m to about 1000 m. Deeper values (>1000m) are very uncertain due to the few data points available as shown on the depth histogram.
  Thus, we propose to add the following at L286 and L280: “[vertically integrated between 0 and 300 m] to ensure that most of the organisms present in the epipelagic zone are included”
  We also propose to replace L264 “Then, we only retained observations from the upper 300 m to exclude deep water samples and focus on zooplankton communities that inhabit the euphotic layer” by “Then, we only retained observations from the upper 300 m to exclude deep water samples and focus on zooplankton communities that inhabit the euphotic layer because measured biomasses and sample numbers are low below 300 m (see. Fig. A2)”
  And to add figure A2 in the appendix (which would now contain 10 figures), with the following caption :
  Figure A2. Depth-Biomass scatter plot and histogram of FFGM observed biomass and maximal depth of the samples AtlantECO dataset before excluding deep samples (Section 2.4.1). Blue points are samples. The red dots represent the median biomass per depth bin.
  Line 324-325:
  
  It is interesting that doubling the complexity of the zooplankton (from 2 to 4 compartments) has minimal effect on chlorophyll. It would be interesting to have a short discussion on why this is and how it compares to other biogeochemical models.
  Chlorophyll is relatively little impacted by the addition of higher trophic levels (in PISCES-FFGM compared to PISCES-v2, the version published in Aumont et al. 2015). Indeed, macrozooplankton represent only 10% of the biomass of all organisms, and they consume less than 5% of the total primary production. Therefore, adding more macrozooplankton groups has a small effect on primary production in our model because grazing rates on phytoplankton are very low (< 10%), partly because GM eats little phytoplankton and then because metabolic rates are low. These low differences in chlorophyll fields also suggest that the mesozooplankton quadratic mortality used in PISCES-v2 is quite relevant to implicitly account for the effect of predation pressure by upper trophic levels on primary productivity.
  Note that our study focuses on the role of FFGM, and thus our analysis is mostly based on comparing PISCES-GM with PISCES-FFGM. The modest role of adding FFGM on lower trophic levels (phytoplankton and microzooplankton) is by the way mentioned and explained in Section 4.1.2
  Section 3.1.2:
  
  What is the reasoning behind using 0.5 mg C m-3 as the cut off for high/low biomass (Table 4 etc).
  The cut-off value corresponds to the rounded median value of macrozooplankton observations from MAREDAT (0.52 mg C m^-3), see Table 4). Regions where biomass (simulated or observed) is higher (resp. lower) than the cut-off are defined as high (resp. low) biomass regions.
  The following sentence will be added at the end of the legend of Table 4 :
  “The cut-off value of 0.5 mg C m-3, used for defining high and low biomass regions, corresponds to the rounded median value of the macrozooplankton observations from MAREDAT (see section 2.4.2).”
  Figure 5:
  
  Red/green is not a good choice for distinguishing between the two model runs, as red/green is the most common type of colour-blindness. This should be changed to a colour-blind friendly palette.
  We will switch our previous color palette to a colour blind friendly one in figure 5.
  Table 6:
  
  Include a description of the difference between the two Luo et al. (2020) columns.
  The following sentence will be included in the legend of Table 6 : “There are two columns for (Luo et al. 2020) as the authors tested two parameterizations of carcass and fecal pellet sinking speeds : 1000 m/d for carcasses and 650 m/d for fecal pellets (third column) or 800 m/d for carcasses and 100 m/d for fecal pellets (fourth column). »
  TECHNICAL CORRECTIONS
  All technical corrections will be applied to the revised version of the manuscript.
  
  Citation: https://doi.org/10.5194/egusphere-2022-1282-AC1
RC2:
'Comment on egusphere-2022-1282', Anonymous Referee #2, 07 Jan 2023

This is an excellent paper tackling to evaluate the impact of large filter-feeding gelatinous zooplankton (i.e. pelagic tunicates, except for appendicularians) on the marine geochemical cycle on a global scale. Although there will be some uncertainties about the calculated figures due to the paucity of data on the distribution, I believe it is worthwhile to publish them as a basis for raising the issue and for discussion. The manuscript is well-prepared and I have no critical comments for the publication except for minor points as shown below.

1. Some large salp species commonly show a large extent of diel vertical migration. So authors need to mention how they compile the data sets taken at different timing of the day. Also, the effect of the diel vertical migration on carbon transportation through the mesopelagic migrant pump (Boyd et al. 2019) could be added to the discussion.

2. Generally the size of doliolds residing in the epi-pelagic layer is small except for one species. Therefore I consider that role of doliolids in their estimation was negligible. Is it possible to estimate the impact depending on the taxonomic group?

3. Please add references to the CNP ratio of pelagic tunicates in line 103.

Citation: https://doi.org/10.5194/egusphere-2022-1282-RC2
- AC2:
  'Reply on RC2', Corentin Clerc, 17 Jan 2023
  The authors would like to thank reviewer #2 for his/her valuable and useful comments. The authors considered all suggestions and addressed the raised issues, which we believed increased the clarity of the revised manuscript. Below are given point-by-point answers to the comments.
  Reviewer comments are in bold, responses in normal font. Proposed changes to the manuscript are in italics.
  
  This is an excellent paper tackling to evaluate the impact of large filter-feeding gelatinous zooplankton (i.e. pelagic tunicates, except for appendicularians) on the marine geochemical cycle on a global scale. Although there will be some uncertainties about the calculated figures due to the paucity of data on the distribution, I believe it is worthwhile to publish them as a basis for raising the issue and for discussion. The manuscript is well prepared and I have no critical comments for the publication except for minor points as shown below.
  We would like to thank Reviewer #2 for her/his very supportive comments on our manuscript.
  Some large salp species commonly show a large extent of diel vertical migration. So authors need to mention how they compile the data sets taken at different timing of the day. Also, the effect of the diel vertical migration on carbon transportation through the mesopelagic migrant pump (Boyd et al. 2019) could be added to the discussion.
  
  The data retained for our analyses lie within the first 300 meters of the ocean’s water column. This depth range was indeed chosen to capture the majority of DVM behavior of the target PFT, and to reflect the observational biomass data, most of the observed Thaliacaea biomass is indeed located in the upper 300 m of the ocean. Indeed, when analyzing the raw AtlantECO dataset (i-e non gridded), the mean biomass for FFGM is 4.93 mg C m^-3 and the median 0.31 mg C m^-3 for the data with a maximum sampling depth shallower than 300 m. These metrics drop to 0.04 mg C m^-3 for the mean and 0.0 mg C m^-3 for the median when focusing at the data with a minimum sampling depth deeper than 300 m.
  We agree that tunicate biomass estimates are likely to be underestimated since some salp species can migrate several hundred meters per day (e.g. 600 m for Salpa fusiformis (Pascual et al. 2017), 300 m for Salpa thompsonii (Henschke et al. 2021)). Accounting for this bias as a function of sampling time would require 1) having a clear quantification of migration depth intervals for each tunicate species in the observation dataset, and 2) having an idea of the spatial heterogeneity of the migrations for a given species. However, these data are only available for a small number of species and are based on a few local studies that do not show the spatial variability of migration depths, and that are thus not representative of processes at global monthly scales. For all these reasons, we opted for the simplest approach, namely to average the data according to the month of collection independently of time of the day. We will add to the manuscript a mention of this potential bias in the observational data by adding the following sentence (L273): “Although some pelagic tunicate species show a large extent of diel vertical migration (Pascual et al. 2017, Henschke et al. 2021), the present observational data were averaged per months regardless of sampling time, due to the lack of precise quantitative information on the taxon-specific magnitude and spatial heterogeneity of these diel vertical migrations. A low […].”
  
  We agree with Reviewer#2 that diel vertical migration (DVM) is widespread in the zooplankton and can have significant impacts on carbon fluxes. However, the effect of active transport on the biological carbon pump is still poorly quantified at the scale of all migrating communities (Boyd et al., 2019). Quantitative data are even less available for a FFGM-specific parameterization of the DVM process and its associated fluxes of organic carbon.
  
  The following paragraph will be added at the end of the discussion, just before the conclusion (L616):
  “Diel vertical migration (DVM) is a key process that is currently not included in the model and that could deepen the production of carcasses and fecal pellets. Recent modeling studies that accounted for DVM at the community level demonstrated significant impact of migrants on carbon export (Aumont et al. 2018, Gorgues et al. 2019, Boyd et al. 2019). As some FFGM species undergo DVM (Pascual et al. 2017, Henschke et al. 2021b), this process is likely to strengthen their impact on carbon export by increasing the average depth at which carcasses and fecal pellets would be released into the water column, inducing a shorter path to the seafloor associated with lower total remineralization of these particles.»
  
  The corresponding section name will be changed from “4.3.2. Carcasses and fecal pellets transfer efficiency” to “4.3.2. Carcasses and fecal pellets”.
  
  New references :
  Pascual, M., Acuña, J., Sabatés, A., Raya, V., and Fuentes, V.: Contrasting diel vertical migration patterns in Salpa fusiformis populations, Journal of Plankton Research, 39, 836–842, 2017.
  Boyd, P. W., Claustre, H., Levy, M., Siegel, D. A., and Weber, T.: Multi-faceted particle pumps drive carbon sequestration in the ocean, Nature, 568, 327–335, 2019.
  Gorgues, T., Aumont, O., and Memery, L.: Simulated changes in the particulate carbon export efficiency due to diel vertical migration of zooplankton in the North Atlantic, Geophysical Research Letters, 46, 5387–5395, 2019.
  
  Generally the size of doliolids residing in the epi-pelagic layer is small except for one species. Therefore I consider that role of doliolids in their estimation was negligible. Is it possible to estimate the impact depending on the taxonomic group?
  
  The data do show a much lower median concentration for doliolids (0.01 mg C m^-3) than for salps (0.55 mg C m^-3) and pyrosomes (0.59 mg C m^-3) (taking the median of non-null biomass samples for each Order). This suggests that the role of doliolids on the carbon cycle would be negligible within large pelagic tunicates. For instance, assuming that carcasses and fecal pellets are identical within pelagic tunicates, and that the median of an Order divided by the sum of the medians of the Orders (Salpida + Pyrosomatida + Doliolida) is a good proxy of the contribution of the order to the biomass of large pelagic tunicates, doliolids would contribute to less than 1% of the production of carcasses and fecal pellets from FFGMs and thus to the associated carbon fluxes. These are very uncertain values based on very rough assumptions, but it would be difficult to assess the impact by taxonomic group using the PISCES-FFGM model, given the lack of constraints on the parameterization of the different groups, which led to the choice to represent only one group of "Large pelagic tunicates".
  
  Please add references to the CNP ratio of pelagic tunicates in line 103.
  
  Following the C:N:P ratio implemented for the other zooplankton groups modelled in PISCES-v2 (Aumont et al. 2015), and to avoid variable stoichiometry (which requires a substantial increase in computational resources), the C:N:P ratio of the two present macrozooplankton groups was set to the Redfield ratio (122:16:1). This will be clarified in the manuscript as follows:
  
  L103: “As with micro- and mesozooplankton in the standard version of PISCES, the C:N:P stoichiometric composition of the two macrozooplankton groups is assumed to be constant.”
  Will be replaced by: “As with micro- and mesozooplankton in the standard version of PISCES, the C:N:P stoichiometric composition of the two macrozooplankton groups is assumed to be constant and equal to the Redfield ratio (Aumont et al. 2015).”
  
  Citation: https://doi.org/10.5194/egusphere-2022-1282-AC2

Interactive discussion

Status: closed

RC1:
'Comment on egusphere-2022-1282', Rebecca Wright, 06 Jan 2023

I recommend the manuscript be accepted subject to minor revisions, which are detailed in the attached document.

Citation: https://doi.org/10.5194/egusphere-2022-1282-RC1
- AC1: 'Reply on RC1', Corentin Clerc, 17 Jan 2023
  
  The authors would like to thank reviewer #1 for her valuable and useful comments. The authors considered all suggestions and addressed the raised issues, which we believed increased the clarity of the revised manuscript. Below are given point-by-point answers to the comments.
  Reviewer comments are in bold, responses in normal font. Proposed changes to the manuscript are in italics.
  I think the authors have offered an interesting, thorough, and generally well-presented manuscript. The manuscript adds a valuable contribution to the growing body of research around modelling gelatinous zooplankton, further supporting the importance of gelatinous zooplankton in global carbon export while also highlighting key areas for future research.
  We would like to thank Reviewer #1 for her very supportive comments on our manuscript.
  SPECIFIC COMMENTS
  Consider using ‘Crustacean Macrozooplankton (CM)’, instead of ‘Generic Macrozooplankton (GM)’. As the generic macrozooplankton group is based entirely on crustacean macrozooplankton I think this would aid in the clarity and precision of the manuscript. Generic implies that the group represents organisms across the macrozooplankton i.e. more than just crustaceans.
  We agree with the reviewer that the definition of the generic macrozooplankton compartment (L136) was not precise enough, which may have lead to some confusion. Actually, our ‘generic macrozooplankton group’ intends to represent macrozooplankton that are not part of the Tunicata (“non-tunicate macrozooplankton”), which includes a lot of crustaceans but not only (e.g., pteropods). Since this compartment was parameterized using an allometric scaling relationship rather than group-specific parameterizations, we think that the group is generic enough to represent also non-crustacean types of macrozooplankton, such as pteropods. For this reason, the generic macrozooplankton notation is more appropriate, and the corresponding explanatory text in the Methods will be corrected as follows:
  L136: “GM, namely generic macrozooplankton, is intended to represent non-tunicate macrozooplankton, such as euphausids, pteropods or large copepods.» instead of “GM, namely generic macrozooplankton, is intended to represent crustacean macrozooplankton, such as euphausids or large copepods.»
  Also, L137, ”The parametrization is similar to that of mesozooplankton (Eq. 28 to 31 in Aumont et al. 2015)” will be replaced by “The parametrization is similar to that of mesozooplankton (Eq. 28 to 31 in Aumont et al. 2015) and parameter values have been derived using allometric scaling relationships (see section 2.2.1).”
  Line 151:
  
  Include an explanation on why the decision was made to have ‘feeding with identical preferences on both phytoplankton groups... as well as on microzooplankton’.
  We assumed non-selective predation when modelling filtration in FFGM, meaning that the organisms are able to ingest all filtered prey. This lack of food selectivity is a known characteristic of filter-feeders (Pauli. et al., 2021b). Recent studies suggest that the efficiency of prey capture by filter-feeding gelatinous organisms could depend on the prey size but also on their taxonomy (Sutherland et al., 2022). Yet, while “future work should bring more quantitative studies of grazing rates and selectivity under in situ conditions, expand the number of pelagic tunicate species that have been studied” (Sutherland et al., 2022), quantitative estimates remain insufficient to properly assess this phenomenon, therefore we believe that non-selective predation remains the best option in the current state of knowledge.
  To address the present comment, we propose to insert the following sentence at line 149 : “Although there is some recent evidence for selective feeding behavior in pelagic tunicates (Sutherland et al. 2022), the lack of quantitative study led to the simpler representation of FFGM as non-selective feeders (Pakhomov et al. 2002, Vargas et al. 2004, von Harbou et al. 2011). Therefore, we assume […]”
  New references :
  von Harbou, L., Dubischar, C. D., Pakhomov, E. A., Hunt, B. P., Hagen, W., and Bathmann, U. V.: Salps in the Lazarev Sea, Southern Ocean: I. Feeding dynamics, Marine Biology, 158, 2009–2026, 2011.
  Vargas, C. A. and Madin, L. P.: Zooplankton feeding ecology: clearance and ingestion rates of the salps Thalia democratica, Cyclosalpa affinis and Salpa cylindrica on naturally occurring particles in the Mid-Atlantic Bight, Journal of plankton research, 26, 827–833, 2004.
  
  Section 2.3: Sensitivity experiments
  
  Adding an introductory sentence to this section would help the flow and to guide the reader, i.e. ‘Five sensitivity experiments were carried out to...’.
  In accordance with the reviewer’s comment, the following introductory sentence will be added at line 190 : ‘Five sensitivity experiments were carried out to evaluate the sensitivity of the model to the chosen parameterization.”
  Rainbow colour palette for figures (fig. 3a, fig. 4, fig. 8, fig. A2e,f, fig. A3, fig. A4, fig. A5)
  
  I strongly recommend changing from a rainbow colour palette for figures, to a “perceptually uniform” palette which provides the same perceived colour change over the same change in value.
  In accordance with the suggestion by the reviewer, colour palette will be changed to a “perceptually uniform” palette in fig. 3a, fig. 4, fig. 8, fig. A2 e,f, fig. A3, fig. A4 and fig. A5. We will use the “cividis” colour palette from the matplotlib python package.
  Line 286:
  
  Include a brief mention of why 0-300m was chosen for analysis.
  The 0-300 m depth range was chosen for our analyses to ensure that we were capturing most of the organisms present in the epipelagic zone, which can be as deep as 250-300 m in the tropical gyres. This choice was applied for both macro- and mesozooplankton but there was an error in the manuscript indicating a 0-100 m depth range for macrozooplankton (L280). This typo was corrected.
  We justify this choice here with the AtlantECO dataset (see section 2.4.1) for FFGM species.
  First, when looking at the distribution of the maximal depth of sampling (see figure A2 in the attached supplement page 3), there is a clear cut off in the distribution at around 300 m, samples being sporadically distributed for deeper maximal depths. Spatial resolution of data below 300m is hence very coarse. Also, the depth distribution displayed below shows that using 100 m as a maximal depth would exclude a significant portion of the dataset.
  Second, the available data suggest that most of the biomass is present in the 300 m first meters. Indeed, when analyzing the raw AtlantECO dataset (i-e non gridded), the mean biomass for FFGM is 4.93 mg C m^-3 and the median 0.31 mg C m^-3 for the data with a maximum sampling depth shallower than 300 m. These metrics drop to 0.04 mg C m^-3 for the mean and 0.0 mg C m^-3 for the median when focusing at the data with a minimum sampling depth deeper than 300 m.
  Last, when focusing on the median by maximum depth bins (red points in the attached supplement page 3), we show that the median is null from 300 m to about 1000 m. Deeper values (>1000m) are very uncertain due to the few data points available as shown on the depth histogram.
  Thus, we propose to add the following at L286 and L280: “[vertically integrated between 0 and 300 m] to ensure that most of the organisms present in the epipelagic zone are included”
  We also propose to replace L264 “Then, we only retained observations from the upper 300 m to exclude deep water samples and focus on zooplankton communities that inhabit the euphotic layer” by “Then, we only retained observations from the upper 300 m to exclude deep water samples and focus on zooplankton communities that inhabit the euphotic layer because measured biomasses and sample numbers are low below 300 m (see. Fig. A2)”
  And to add figure A2 in the appendix (which would now contain 10 figures), with the following caption :
  Figure A2. Depth-Biomass scatter plot and histogram of FFGM observed biomass and maximal depth of the samples AtlantECO dataset before excluding deep samples (Section 2.4.1). Blue points are samples. The red dots represent the median biomass per depth bin.
  Line 324-325:
  
  It is interesting that doubling the complexity of the zooplankton (from 2 to 4 compartments) has minimal effect on chlorophyll. It would be interesting to have a short discussion on why this is and how it compares to other biogeochemical models.
  Chlorophyll is relatively little impacted by the addition of higher trophic levels (in PISCES-FFGM compared to PISCES-v2, the version published in Aumont et al. 2015). Indeed, macrozooplankton represent only 10% of the biomass of all organisms, and they consume less than 5% of the total primary production. Therefore, adding more macrozooplankton groups has a small effect on primary production in our model because grazing rates on phytoplankton are very low (< 10%), partly because GM eats little phytoplankton and then because metabolic rates are low. These low differences in chlorophyll fields also suggest that the mesozooplankton quadratic mortality used in PISCES-v2 is quite relevant to implicitly account for the effect of predation pressure by upper trophic levels on primary productivity.
  Note that our study focuses on the role of FFGM, and thus our analysis is mostly based on comparing PISCES-GM with PISCES-FFGM. The modest role of adding FFGM on lower trophic levels (phytoplankton and microzooplankton) is by the way mentioned and explained in Section 4.1.2
  Section 3.1.2:
  
  What is the reasoning behind using 0.5 mg C m-3 as the cut off for high/low biomass (Table 4 etc).
  The cut-off value corresponds to the rounded median value of macrozooplankton observations from MAREDAT (0.52 mg C m^-3), see Table 4). Regions where biomass (simulated or observed) is higher (resp. lower) than the cut-off are defined as high (resp. low) biomass regions.
  The following sentence will be added at the end of the legend of Table 4 :
  “The cut-off value of 0.5 mg C m-3, used for defining high and low biomass regions, corresponds to the rounded median value of the macrozooplankton observations from MAREDAT (see section 2.4.2).”
  Figure 5:
  
  Red/green is not a good choice for distinguishing between the two model runs, as red/green is the most common type of colour-blindness. This should be changed to a colour-blind friendly palette.
  We will switch our previous color palette to a colour blind friendly one in figure 5.
  Table 6:
  
  Include a description of the difference between the two Luo et al. (2020) columns.
  The following sentence will be included in the legend of Table 6 : “There are two columns for (Luo et al. 2020) as the authors tested two parameterizations of carcass and fecal pellet sinking speeds : 1000 m/d for carcasses and 650 m/d for fecal pellets (third column) or 800 m/d for carcasses and 100 m/d for fecal pellets (fourth column). »
  TECHNICAL CORRECTIONS
  All technical corrections will be applied to the revised version of the manuscript.
  
  Citation: https://doi.org/10.5194/egusphere-2022-1282-AC1
RC2:
'Comment on egusphere-2022-1282', Anonymous Referee #2, 07 Jan 2023

This is an excellent paper tackling to evaluate the impact of large filter-feeding gelatinous zooplankton (i.e. pelagic tunicates, except for appendicularians) on the marine geochemical cycle on a global scale. Although there will be some uncertainties about the calculated figures due to the paucity of data on the distribution, I believe it is worthwhile to publish them as a basis for raising the issue and for discussion. The manuscript is well-prepared and I have no critical comments for the publication except for minor points as shown below.

1. Some large salp species commonly show a large extent of diel vertical migration. So authors need to mention how they compile the data sets taken at different timing of the day. Also, the effect of the diel vertical migration on carbon transportation through the mesopelagic migrant pump (Boyd et al. 2019) could be added to the discussion.

2. Generally the size of doliolds residing in the epi-pelagic layer is small except for one species. Therefore I consider that role of doliolids in their estimation was negligible. Is it possible to estimate the impact depending on the taxonomic group?

3. Please add references to the CNP ratio of pelagic tunicates in line 103.

Citation: https://doi.org/10.5194/egusphere-2022-1282-RC2
- AC2:
  'Reply on RC2', Corentin Clerc, 17 Jan 2023
  The authors would like to thank reviewer #2 for his/her valuable and useful comments. The authors considered all suggestions and addressed the raised issues, which we believed increased the clarity of the revised manuscript. Below are given point-by-point answers to the comments.
  Reviewer comments are in bold, responses in normal font. Proposed changes to the manuscript are in italics.
  
  This is an excellent paper tackling to evaluate the impact of large filter-feeding gelatinous zooplankton (i.e. pelagic tunicates, except for appendicularians) on the marine geochemical cycle on a global scale. Although there will be some uncertainties about the calculated figures due to the paucity of data on the distribution, I believe it is worthwhile to publish them as a basis for raising the issue and for discussion. The manuscript is well prepared and I have no critical comments for the publication except for minor points as shown below.
  We would like to thank Reviewer #2 for her/his very supportive comments on our manuscript.
  Some large salp species commonly show a large extent of diel vertical migration. So authors need to mention how they compile the data sets taken at different timing of the day. Also, the effect of the diel vertical migration on carbon transportation through the mesopelagic migrant pump (Boyd et al. 2019) could be added to the discussion.
  
  The data retained for our analyses lie within the first 300 meters of the ocean’s water column. This depth range was indeed chosen to capture the majority of DVM behavior of the target PFT, and to reflect the observational biomass data, most of the observed Thaliacaea biomass is indeed located in the upper 300 m of the ocean. Indeed, when analyzing the raw AtlantECO dataset (i-e non gridded), the mean biomass for FFGM is 4.93 mg C m^-3 and the median 0.31 mg C m^-3 for the data with a maximum sampling depth shallower than 300 m. These metrics drop to 0.04 mg C m^-3 for the mean and 0.0 mg C m^-3 for the median when focusing at the data with a minimum sampling depth deeper than 300 m.
  We agree that tunicate biomass estimates are likely to be underestimated since some salp species can migrate several hundred meters per day (e.g. 600 m for Salpa fusiformis (Pascual et al. 2017), 300 m for Salpa thompsonii (Henschke et al. 2021)). Accounting for this bias as a function of sampling time would require 1) having a clear quantification of migration depth intervals for each tunicate species in the observation dataset, and 2) having an idea of the spatial heterogeneity of the migrations for a given species. However, these data are only available for a small number of species and are based on a few local studies that do not show the spatial variability of migration depths, and that are thus not representative of processes at global monthly scales. For all these reasons, we opted for the simplest approach, namely to average the data according to the month of collection independently of time of the day. We will add to the manuscript a mention of this potential bias in the observational data by adding the following sentence (L273): “Although some pelagic tunicate species show a large extent of diel vertical migration (Pascual et al. 2017, Henschke et al. 2021), the present observational data were averaged per months regardless of sampling time, due to the lack of precise quantitative information on the taxon-specific magnitude and spatial heterogeneity of these diel vertical migrations. A low […].”
  
  We agree with Reviewer#2 that diel vertical migration (DVM) is widespread in the zooplankton and can have significant impacts on carbon fluxes. However, the effect of active transport on the biological carbon pump is still poorly quantified at the scale of all migrating communities (Boyd et al., 2019). Quantitative data are even less available for a FFGM-specific parameterization of the DVM process and its associated fluxes of organic carbon.
  
  The following paragraph will be added at the end of the discussion, just before the conclusion (L616):
  “Diel vertical migration (DVM) is a key process that is currently not included in the model and that could deepen the production of carcasses and fecal pellets. Recent modeling studies that accounted for DVM at the community level demonstrated significant impact of migrants on carbon export (Aumont et al. 2018, Gorgues et al. 2019, Boyd et al. 2019). As some FFGM species undergo DVM (Pascual et al. 2017, Henschke et al. 2021b), this process is likely to strengthen their impact on carbon export by increasing the average depth at which carcasses and fecal pellets would be released into the water column, inducing a shorter path to the seafloor associated with lower total remineralization of these particles.»
  
  The corresponding section name will be changed from “4.3.2. Carcasses and fecal pellets transfer efficiency” to “4.3.2. Carcasses and fecal pellets”.
  
  New references :
  Pascual, M., Acuña, J., Sabatés, A., Raya, V., and Fuentes, V.: Contrasting diel vertical migration patterns in Salpa fusiformis populations, Journal of Plankton Research, 39, 836–842, 2017.
  Boyd, P. W., Claustre, H., Levy, M., Siegel, D. A., and Weber, T.: Multi-faceted particle pumps drive carbon sequestration in the ocean, Nature, 568, 327–335, 2019.
  Gorgues, T., Aumont, O., and Memery, L.: Simulated changes in the particulate carbon export efficiency due to diel vertical migration of zooplankton in the North Atlantic, Geophysical Research Letters, 46, 5387–5395, 2019.
  
  Generally the size of doliolids residing in the epi-pelagic layer is small except for one species. Therefore I consider that role of doliolids in their estimation was negligible. Is it possible to estimate the impact depending on the taxonomic group?
  
  The data do show a much lower median concentration for doliolids (0.01 mg C m^-3) than for salps (0.55 mg C m^-3) and pyrosomes (0.59 mg C m^-3) (taking the median of non-null biomass samples for each Order). This suggests that the role of doliolids on the carbon cycle would be negligible within large pelagic tunicates. For instance, assuming that carcasses and fecal pellets are identical within pelagic tunicates, and that the median of an Order divided by the sum of the medians of the Orders (Salpida + Pyrosomatida + Doliolida) is a good proxy of the contribution of the order to the biomass of large pelagic tunicates, doliolids would contribute to less than 1% of the production of carcasses and fecal pellets from FFGMs and thus to the associated carbon fluxes. These are very uncertain values based on very rough assumptions, but it would be difficult to assess the impact by taxonomic group using the PISCES-FFGM model, given the lack of constraints on the parameterization of the different groups, which led to the choice to represent only one group of "Large pelagic tunicates".
  
  Please add references to the CNP ratio of pelagic tunicates in line 103.
  
  Following the C:N:P ratio implemented for the other zooplankton groups modelled in PISCES-v2 (Aumont et al. 2015), and to avoid variable stoichiometry (which requires a substantial increase in computational resources), the C:N:P ratio of the two present macrozooplankton groups was set to the Redfield ratio (122:16:1). This will be clarified in the manuscript as follows:
  
  L103: “As with micro- and mesozooplankton in the standard version of PISCES, the C:N:P stoichiometric composition of the two macrozooplankton groups is assumed to be constant.”
  Will be replaced by: “As with micro- and mesozooplankton in the standard version of PISCES, the C:N:P stoichiometric composition of the two macrozooplankton groups is assumed to be constant and equal to the Redfield ratio (Aumont et al. 2015).”
  
  Citation: https://doi.org/10.5194/egusphere-2022-1282-AC2

Peer review completion

AR: Author's response | RR: Referee report | ED: Editor decision | EF: Editorial file upload

ED: Publish subject to minor revisions (review by editor) (19 Jan 2023) by Olivier Sulpis

AR by Corentin Clerc on behalf of the Authors (01 Feb 2023) Author's response Author's tracked changes Manuscript

ED: Publish as is (07 Feb 2023) by Olivier Sulpis

AR by Corentin Clerc on behalf of the Authors (07 Feb 2023) Author's response Manuscript

Journal article(s) based on this preprint

28 Feb 2023

Including filter-feeding gelatinous macrozooplankton in a global marine biogeochemical model: model–data comparison and impact on the ocean carbon cycle

Corentin Clerc, Laurent Bopp, Fabio Benedetti, Meike Vogt, and Olivier Aumont

Biogeosciences, 20, 869–895, https://doi.org/10.5194/bg-20-869-2023,https://doi.org/10.5194/bg-20-869-2023, 2023

Short summary

Corentin Clerc, Laurent Bopp, Fabio Benedetti, Meike Vogt, and Olivier Aumont

Viewed

Total article views: 301 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	BibTeX	EndNote
218	73	10	301	4	3

HTML: 218
PDF: 73
XML: 10
Total: 301
BibTeX: 4
EndNote: 3

Views and downloads (calculated since 30 Nov 2022)

Month	HTML	PDF	XML	Total
Nov 2022	35	5	1	41
Dec 2022	70	21	4	95
Jan 2023	76	33	5	114
Feb 2023	37	14	0	51
Mar 2023	0
Apr 2023	0
May 2023	0
Jun 2023	0
Jul 2023	0
Aug 2023	0
Sep 2023	0
Oct 2023	0
Nov 2023	0
Dec 2023	0
Jan 2024	0
Feb 2024	0
Mar 2024	0
Apr 2024	0
May 2024	0
Jun 2024	0
Jul 2024	0
Aug 2024	0
Sep 2024	0

Cumulative views and downloads (calculated since 30 Nov 2022)

Month	HTML	PDF	XML	Total
Nov 2022	35	5	1	41
Dec 2022	70	21	4	95
Jan 2023	76	33	5	114
Feb 2023	37	14	0	51
Mar 2023	0
Apr 2023	0
May 2023	0
Jun 2023	0
Jul 2023	0
Aug 2023	0
Sep 2023	0
Oct 2023	0
Nov 2023	0
Dec 2023	0
Jan 2024	0
Feb 2024	0
Mar 2024	0
Apr 2024	0
May 2024	0
Jun 2024	0
Jul 2024	0
Aug 2024	0
Sep 2024	0

Viewed (geographical distribution)

Total article views: 298 (including HTML, PDF, and XML) Thereof 298 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 02 Sep 2024

Download

The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.

Preprint (6787 KB)
Metadata XML

Short summary

Gelatinous zooplankton play a key role in the ocean carbon cycle. In particular, pelagic tunicates, which feed on a wide size range of preys, produce rapidly sinking detritus. Thus, they efficiently transfer carbon from the surface to the depth. Consequently, we added these organisms to a marine biogeochemical model (PISCES-v2) and evaluated their impact on the global carbon cycle. We found that they contribute significantly to carbon export, and that this contribution increases with depth.


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