Review of ms entitled "Biological production in two contrasted regions of the Mediterranean Sea during the oligotrophic period: An estimate based on the diel cycle of optical properties measured by BGC-Argo profiling floats" by Barbieux et al.

GENERAL COMMENTS

This study is an original contribution to the field of bio-optical oceanography. Though the approach used in this study is not new, the ms provide a detailed analysis of a unique dataset that includes high-resolution vertical measurements of biogeochemical properties acquired in two ecoregions of the Mediterranean Sea, during summer oligotrophy.

The application of the cp-based approach to estimate particles production from such data provides new findings, that significantly improve our understanding of particles dynamics in oligotrophic areas. The ms successfully addresses the challenge of filling typical observation gaps in a traditionally under-sampled ocean.

The introduction clearly introduces the general context and specific objectives of the present study.

The M&M is detailed, sound and robust. The structure of the M&M is sometimes a bit confusing, and there this some overlap between sections 2.3 – 2.5 and section 2.6. I have two major methodological comments (see below comment #2), and would suggest to display more examples cp and bbp measurements (see below comment #1).

The Results and Discussion, which are merged in a single section, are nicely structured and generally well supported.

The conclusion is too long, and there is some degree of redundancy with the discussion. Synthetizing the conclusion to the main results and “take-home messages” would improve the article.

SPECIFIC COMMENTS

1) As the whole study is built on cp and bbp measurements, it is necessary to show the data (an appendix could be used for that purpose). A schematic representation of the cp diel variability (Figure 2) is useful to explain the method, and the example provided in Fig. 5 (Ionian Sea) is interesting, but this is not sufficient.

2) Here, I raise two methodological comments concerning section 2.3. “Characterization of the diel cycle of the bio-optical properties”.

First comment: in the literature, the diel variability is generally defined as the change in cp between sunrise and sunset (Siegel et al., 1989; Cullen et al., 1992, etc.). Such daytime increase in cp has been previously associated to particle growth and production.

In the present study, the diel variability is computed as the relative variation between two sunrises (in the Ionian Sea, eq. 3) or two noons (in the Ligurian Sea). Did the authors also characterize the sunrise-to-sunset variability?

Second comment: the time reference used to characterize the daily changes in cp are not the same in the Ionian Sea (reference = sunrise) and in the Ligurian Sea (reference = noon).

This introduces a bias in the comparison between the results from both study sites, and should be discussed. Would it be possible to compute the diel variability using the same time reference (e.g. sunrise) in both cases?

The same two comments also apply to Section 2.6.3 “Calculation of the production rate”. In eqs. 9-12, the authors explain how a daily and depth-integrated production of particles, P can be inferred from diel changes in cp. Is the variable "P" a proxy of the net community production (NCP) as defined in Claustre et al 2008? Or is it something different? It would be useful to add this precision in subsection 2.6.3. It is not clear whether the production has been inferred using the day-to-day or the daytime (sunrise to sunset) change in the cp-derived POC.

How was the gross community production (GCP) estimated?

Here also, the time reference differs between the Ligurian and Ionian Sea (L343-345). How is it expected to influence the results? Would it be possible to standardize the method?

3) As the instruments deployed at BOUSSOLE provide high-frequency measurements, it would be very interesting to compare the surface cp time-series acquired by the fLig float with the BOUSSOLE data (accounting the lag of 2 days identified in L371-380). Such a comparison would be useful to assess if some information is missed by the lower (but still high) temporal resolution of the BGC Argo float, as well as to assess the spatial representativity of BOUSSOLE point-based measurements.

TECHNICAL COMMENTS

In the title, the term "optical properties" is maybe to general, and could be more detailed (e.g. beam attenuation coefficient).

The variables cp, bbp, Zeu, Zpd, etc. should be italicized throughout the ms

L167-175 how was the time period of the two floats selected?

L265 typo in "Loisel et al."

L287-300 and Table 1. The same cp-to-POC relationship is used in the Ligurian and Ionian Sea, despite the bio-optical differences between the two basins. Two different relationships have also been reported in the literature (Oubelkheir et al 2005, Loisel et al 2011). The authors decided to apply the results from Oubelkheir et al. (2005). Why not from Loisel et al (2011)? This could be added to the discussion in L502-508.

Figure1: it would be useful to display the summertime-averaged satellite-derived Chl concentration in Figure 1

Figure 2: adding grey vertical bar to represent nighttime would help reading the graph

Figure 3: I suggest to add "Ligurian Sea" and "Ionian Sea" at the top of the left and right panels, respectively (and same comment for Figures 4, 6 – 10). In Figure 3, displaying the cp-to-Chl ratio would also be useful to help reading section 3.1 (in particular L389 - 408)

L482-485 the results also compare with the "delta POC" estimates reported in Gernez et al. (2011; see their fig. 14)

L539-542 Here the authors assumed that negative values could be associated with particles transport. Please note that negative values could also occur if the losses exceed particles growth (which could occur if the community is dominated by heterotrophs)

L543-549 this is very interesting. It would be useful to provide the averaged cp/Chl (at the depth of the SCM) of the Ligurian Sea and Ionian Sea. Besides photo-acclimation, are there other ecophysiological hypothesis that would be consistent with this hypothesis? (e.g. composition and size distribution of the community of living particles, higher influence of nutrient stress in the Ionian Sea?)

L607-623 when discussing the range of variability of cp/Chl, please consider referring to Loisel & Morel (1998) 

L653 "it appears to result from changes in light conditions": light and/or nutrients?

L679-689 as acknowledged by the authors, these are very hypothetical statements that are not supported by the present study. I therefore suggest to remove the quantitative results of such crude estimations from the conclusion (i.e. remove or re-write L745-747)

Review Barbieux marie et al. Biogeosciences

General comments

The study aims to derive community production estimation from a dataset obtained by two BioGeoChemical-Argo (BGC-Argo) profiling floats deployed in the Ligurian and Ionian Seas. The authors mainly used the infradiel variability of c_p and b_bp measurements along the water column to investigate the diel variations of the bio-optical properties and thus of the production (estimated in particulate organic carbon). The authors compared the results between two contrasted areas of the Mediterranean Sea.

The study is well-done and convincing. The draft is well-written, well-structured and organized, starting an informative and well-documented introduction.

Below some more specific comments:

Abstract

Lines 18-19: “…marine biological production of organic carbon” … I suggest to be more precise: organic carbon is referred to particulate O.C.?; marine biological production is referred to phytoplankton, bacteria, zooplankton?

Line 30: SCM layer (16-42%): I imagine that it is also (mainly?) dependent on the depth, i.e. on light availability?

Lines 32-33: “the SCM is permanent induced by phytoplankton photoacclimation…” What does it mean?: does this SCM only the result of physiological acclimation/regulation of microalgae (increase of chl.a per cell)? Or/and to the fact that microalgae actively accumulate to this depth for many reasons (light, nutrient, stratification,…)

Introduction

Line 40: “Primary production is an essential component…” instead of component, process? Flux?

Lines 57-60: what about active fluorescence measurements (FRRF?, etc.)

Line 99: “…and found weak results for the diel b_bp cycle…” in which sense?, rephrase/explain

Line 136-137: “…with an SCM induced mostly by photoacclimation (e.g., Mignot et al. 2014; Barbieux et al. 2019)”. Please explain?

Methods

Lines 167-175: move to the section 2.2. (BGC-floats)

Lines 196-201: “fluorescence-to-Chl ratio”. If I understand well, the authors applied two successive types of correction: one related to NPQ and one on the differences between chl.a fluo and chl.a concentration. The last is considered basically from the chl.a fluo/chl.aratio = 2, applied on the two contrasted systems Ligurian and Ionian seas. I was wondering if using the same value for the two contrasted sites can represent an “error” for the interpretation of the results. The environmental (light, nutrients) and biological properties (microalgal communities) of the two sites are strongly different, that probably might affect the ratio fluo/chl.a.

The authors reported that they measured chl.a concentration with HPLC during the cruises they set-up for the deployment of the floats (lines 365-370). Can the authors use these data to retrieve a more precise chl.a fluo/chl.a ratio for the two sites?

Lines 242-247:

the difference of float cycle (noon-noon vs sunrise-sunrise) between the two sites might be a problem for the comparative interpretation of the data? Since Lig cycle was run every four days, while Lion did every day, one solution would be to use the cycle noon-noon also at Lion to be similar to Lig.

Did the authors try to modify the cycle to look at the potential differences in using another starting point?

Lines 267-277: “daily solar cycle”

268: “Abundance of microorganisms”. Which kind of microorganisms is referred to?: phytoplankton, bacteria? Do zooplankton affect the c_p or b_bp as well?

line 272: cell division. Since cell division is strongly dependent on the species and on the physiology of the species which is modulated by diel light cycle, I suggest adding some information on the physiological changes (e.g., photoacclimation) together with abundances as key-factors determining the success in term of production/division of a community (e.g., Doney, et al (1995) Photochemistry, mixing and diurnal cycles in the upper ocean. J. Mar. Res., 53; Litaker, R. W., et al. (2002) Effect of diel and interday variations in light on the cell division pattern and in situ growth rates of the bloom-forming dinoflagellate Heterocapsa triquetra. Mar. Ecol. Prog. Ser., 232, 63–74; Brunet C. et al. 2008. Phytoplankton diel and vertical variability in photobiological responses at a coastal station in the Mediterranean Sea. J. Plankton Res., 30: 645-654.

Line 277: what is the link between microorganism respiration and particle size/refractive index?

Line 308: what does “quasi 1-D framework” mean? It seems to me 1-D?

Line 325: “ZSCM the depth of the SCM”: does the ZSCM refer to the maximum chla fluo depth?

Results&discussion

Line 382: remove “in this section”

Line 402: “…the Ionian Sea SCM is located twice as deep (97 m) and is uncoupled from any cp and bbp maxima that occur at shallower depth.”  Did the authors have an explanation for this spatial uncoupling between cp and scm?  And why was the cause of the cp maximum in the surface layer?

I am also wondering if the chl.a maximum at 100 m depth might be called DCM rather than SCM ?

Line 407: “whereas the Ionian SCM is induced by photoacclimation of phytoplankton cells.” What does it mean? Is there a chl.a maximum? Did the authors suggest that this chl.a maximum is not due to accumulation of microalgae but to an intracellular increase of chl.a due to low light? Can it be a mix of the two (increase in phytoplankton as well as in chla per cell)?

Moreover, the difference between the two systems might be attributed to the fact that SCM depth was 40 m deep in one case and almost 100 m deep in the other. One belongs to the euphotic depth, while the other does not. Microalgal communities seem to be significantly different. All those environmental/ecological properties might explain the low productivity rate in the Ionian sea, compared to the highly productive Ligurian sea. Indeed, other studies (Combet et al., 2011, biogeosciences) reported an increase of POC at depth in the Ionian sea in correspondence of chl.a maximum in summer.

More generally, I think that light data/information is lacking sometimes. Light is the one of the main driver of primary production as well as of the diel variations of microalgal communities. I suggest adding a figure showing characteristic light profiles. the fig. 10 “only” reports the daily integrated light at the SCM in both sites.

Figure 4: I think that there is a mistake on the colors: chl.a is higher in surface than in scm?

Figure 8 is not very informative like this. Please, change the type of figures, we do not really appreciate the potential changes in chla in the three size classes. Also because the right panels report the time variations for only four days.

Table 3: some values are negative:

• Did the authors estimate the mean of these two parameters mixing positive and negative values ? does it really make sense?
• How can be explained the negative value found for ð¦ÌƒΔ of cp in the euphotic layer in the Ligurian sea, in correspondence with the huge variability (2603.1%)? It would reveal the very high spatial variability in this zone, but the ð¦ÌƒΔ of cp was positive for both the surface and SCM layers.

Another point: from the fig.3, for the Ligurian sea, euphotic depth includes SCM and surface, while for the Ionian sea, SCM is generally below than the euphotic depth. Does it useful to compare the relative daily variations (ð¦ÌƒΔ and 𝐫̃Δ, respectively) in the diel cycle of cp and bbp in the euphotic depth between the two systems?

Line 435: efficient?

Line 451: “We compare the cp- and bbp based estimates with primary production estimates computed with the model of Morel (1991).” Is it necessary to use the bbp estimate since the lack of correlation with scm for instance? It does not seem to be relied on phytoplankton production.

Line 544: an/a (correct)

Line 546: “the SCM reflects photoacclimation…”  the SCM is greatly deeper (below the euphotic depth and the mixing layer), Light is strongly lower than in the Ligurian system SCM, maybe temperature is lower, chl.a is less…consequently the production might be lower than in the other system.

Lines 595-596: no need to present the data in absolute and in %.

Lines 613-614: Please explain

Line 633: “…increase their intracellular Chl.” And/or the fluo/chl.a ratio in relation with the little (or absence) of high-light-induced chl.a quenching

Conclusions

To me, this section is too long, and is like a discussion. Many points raised by the authors in this section  were already presented/discussed. I suggest reducing the length of the conclusions.