The 3D biogeochemical marine mercury cycling model MERCY v2.0 – linking atmospheric Hg to methylmercury in fish

Bieser, Johannes; Amptmeijer, David J.; Daewel, Ute; Kuss, Joachim; Soerensen, Anne L.; Schrum, Corinna

doi:https://doi.org/10.5194/gmd-16-2649-2023

Articles | Volume 16, issue 9

https://doi.org/10.5194/gmd-16-2649-2023

© Author(s) 2023. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/gmd-16-2649-2023

© Author(s) 2023. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 16, issue 9

Model description paper

|

17 May 2023

Model description paper |

| 17 May 2023

The 3D biogeochemical marine mercury cycling model MERCY v2.0 – linking atmospheric Hg to methylmercury in fish

Johannes Bieser, David J. Amptmeijer, Ute Daewel, Joachim Kuss, Anne L. Soerensen, and Corinna Schrum

Download

Final revised paper (published on 17 May 2023)
Preprint (discussion started on 21 Oct 2022)

Interactive discussion

Status: closed

RC1:
'Comment on gmd-2021-427', Anonymous Referee #1, 25 Nov 2022

General comment:

The manuscript presents detailed description of formulation, development and evaluation of a new biogeochemical marine Hg cycling model MERCY v2.0 as a part of a multi-media modelling system. Developments of multi-media capabilities of Hg dispersion modelling is highly topical. The problem of Hg pollution on a global scale is well recognized and currently assessed under the effectiveness evaluation efforts of the Minamata Convention. Despite other pollutants Hg requires model evaluation in various environmental compartments. However, available developments of Hg modelling in the marine environment are still insufficient. The presented a model of Hg cycling in seawater including transport transformation and bioaccumulation processed. The model is applied as a part of a modelling complex in combination with atmospheric and oceanic transport models, and a seawater biogeochemical model to simulate Hg levels and dynamics in the North and Baltic seas. The results are thoroughly evaluated against observations to reveal the model uncertainties and propose ways for further improvement. For this purpose, a system of detailed statistical analysis is developed and applied based on methods used in atmospheric transport modelling. This statistical evaluation system could be useful for application by other marine chemistry modelers.

The subject of the manuscript is relevant to the scope of the journal and the work makes up a new and original contribution to the modelling science. The scientific approaches applied are adequate and explicitly stated. Description of the modelling methods is sufficiently complete and precise to allow reproduction. The manuscript will be suitable for publication after addressing comments mentioned below.

Specific comments:

Generally, the manuscript contains a large number of typos and misprints and requires careful editing.

Page 3, lines 84: “While there is a large number of emissions …”

Probably, there should be mentioned a large number of emission inventories.

Page 7, lines 182: “… change in concentration of Hg state variables over time δC/δt is estimated by the prognostic equation…”

δC/δt is unnecessary here. The partial derivative describes the change rate. The change itself requires integration of the equation over time.

Page 9, lines 220-227: “… Bioconcentration … remineralization rate (see Eq. 9 in Section 2.3.1). …”

Notations of variables and parameters used in this paragraph differ from those in Eq. 5. It complicates understanding.

Page 10, Figure 1: The oxidation pathway via formation of the intermediate oxidation product (Hg*) is not included to the model (page 13, line 280) but shown in the model scheme.

Page 12, Table 3: Reactions R5, R13, R18 and R20 are not shown in the model scheme (Fig. 1).

Page 13, line 282: “… oxidation (R5) …”. Should be R4.

Page 13, line 283: “… oxidation (R6) rates …”. Should be R5.

Page 13, line 287: “… of MeHg+ (R19), which …”. Should be R20.

Page 15, lines 339-341: Species HgOHCl(aq), Hg2+-POC(s) and MeHg+-POC(s). are absent in Table 2. MeHg+-POC(s) is also absent in Fig. 1.

Page 15, line 282: “… (Eqs. 10-13). …”. Should be (Eqs. 10-12).

Page 15, line 355 and hereafter: Units of non-dimensional parameters can be given as [1] or [n/d].

Page 16, line 369: “… (Table 2) …”. Should be (Table 1).

Page 42, line 953: “… Figure 14 …”. Should be Figure 15.

Figures 7, 9 and 17: The circles showing measured data in the figures are very small and not readable.

Figure 13: The upper and lower panels are not signed in the caption. The legend is not readable.

Figures 14, 15, 16, 19: The legends are not readable.

Figure 18: The panels are not signed in the caption.

Citation: https://doi.org/10.5194/gmd-2021-427-RC1
- AC1: 'Reply on RC1', Johannes Bieser, 01 Feb 2023
  
  The comment was uploaded in the form of a supplement: https://gmd.copernicus.org/preprints/gmd-2021-427/gmd-2021-427-AC1-supplement.pdf
  
  Citation: https://doi.org/10.5194/gmd-2021-427-AC1
RC2:
'Comment on gmd-2021-427', Yanxu Zhang, 08 Dec 2022
The Marine Hg cycle is an important component of its global biogeochemical cycling. Numerical models, especially 3D ones, are useful to reveal the interaction between transport and biogeochemical processes, interpret observations, and test hypotheses. However, the history of the 3D ocean Hg model is less than 10 years and there are only a handful of such models, which limits our ability to conduct multi-model intercomparison and adopt a model ensemble approach. This manuscript includes a detailed description of the processes and configuration of a numerical multi-compartment model for marine Hg cycling, MERCY v2.0. It presents the model evaluation from regional simulations of two shelf seas. The model also includes important and novel updates regarding Hg biogeochemistry, such as the S-Hg chemistry, larger Kd value consistent with field observations, sedimentation/resuspension, and the tentative inclusion of fish in 3D models. Overall, I consider it an important addition and advancement to existing ocean Hg modeling efforts, which merits publication in GMD. Congratulations to the author team.

Some suggestions and questions:

1.     The model mainly follows a prognostic equation, and the authors describe each sub-term of the transformation term in great detail. Among them the implementation of the sulfur chemistry of mercury is novel, yet an evaluation of the importance of this newly added process seems to be lacking. How does it compare with observations?

2.     The simulation of the bioconcentration process considers the biological uptake of Hg by organisms from higher trophic levels through the body parts exposed to seawater other than phytoplankton. Can you quantify the contribution from this pathway and via food consumption?

3.     Although the model is claimed to be improved by employing a high Kd, the authors do not seem to have explicitly considered the effect of the biological pump on the mercury species at different depths. A single sinking velocity wd is utilized to calculate the vertical transport but the association between this and the biological pump was not given in detail. Nevertheless, including the sedimentation and resuspension in the model makes it more complete than previous models.

4.     The quality criteria proposed in this manuscript entails sophisticated statistical analyses, and the elaborated presentation enables other ocean modelers to reproduce and apply. Also, the authors emphasize the importance of observational data and indicate that some processes are poorly constrained in the discussion. This can help field and laboratory studies address these issues.

Specific points (some may also be spotted by other reviewers):

1.     Line 81, the authors state “The only real sink for Hg in the environment is a burial in the lithosphere mainly as stable cinnabar (HgS) in anoxic marine sediments.” However, there are several data suggesting that the sedimentation of compounds to organic material is a major sink in coastal and open-ocean systems. This may need the authors to include some references to address.

Line 106, "red-dox chemistry" should be "red-ox chemistry".

Line 115 and line 116, Rosati 2022 paper was mentioned twice but they did not appear in the reference list.

Line 162, "en-to-end" should be "end-to-end"

Line 171 Table 1, “GOM”, “PBM” - it would be better if these abbreviations be written out in full on first use.

Line 270, "concentrationdependent" should be "concentration dependent". Line 271, "raction" should be "reaction". And it would be better to add the note on R12 about the remineralized organic matter concentration dependency.

Line 295, is there a literature-based argument to support the use of negative oxygen concentration to represent sulfur ions concentration?

Line 306, “chemistrcy” should be “chemistry”.

Line 320, the square symbol in the formula should be the superscript of "T_w" instead of "w".

Line 348, what does “R(C,B)” mean here? According to the previous introduction (line 188), it represents the transformation term in the prognostic equation?

Line 353-359, 392-399 and 463, 559-563, the units (even unitless or 1) are needed to be written out in the description of the variables.

Line 420 Figure 2, the labels of the color bar may have gone wrong. ‘60’ appears twice for different colors.

Line 436, “due to the comparably low surface areas of these species”, do the authors mean diffusive uptake by zooplankton is less important due to the low surface-to-volume ratio of zooplankton? Since zooplankton generally has a larger diameter, thus larger surface areas but a lower surface-to-volume ratio.

Line 455, readers may wish to know about the exact feeding relationship of “17 × Feeding rates for biological species (x) on species (y)” from Table 1, however, the predation related to fish is not mentioned here.

Line 628 Eq. 44, what is “k1” here?

Line 626 and 640, it seems that the stated “the measurement error to range from 20% (Hg0 and HgT) to 50% (MeHg)” is the same value as “U = measurement uncertainty” that is used to calculate MQO? It would be easier for the reader to understand if it could be phrased consistently.

Figure 12: the title of the x-axis is missing.

Figure 13: according to line 918, the lower panel should be the profile of the North Sea, but it is not listed in the caption.

Figure 16: the caption and axes are not clear.
Citation: https://doi.org/10.5194/gmd-2021-427-RC2
- AC2: 'Reply on RC2', Johannes Bieser, 01 Feb 2023
  
  The comment was uploaded in the form of a supplement: https://gmd.copernicus.org/preprints/gmd-2021-427/gmd-2021-427-AC2-supplement.pdf
  
  Citation: https://doi.org/10.5194/gmd-2021-427-AC2
CEC1:
'Comment on gmd-2021-427', Juan Antonio Añel, 12 Dec 2022

Dear authors,

Unfortunately, after checking your manuscript, it has come to our attention that it does not comply with our "Code and Data Policy".

https://www.geoscientific-model-development.net/policies/code_and_data_policy.html
There are many problems with the code included in your manuscript. First, in your manuscript, you state, "The MERCY v2.0 source code is available upon request"; however, the code is available openly on Zenodo.org. This is good news, but please, you should modify the statement in the text and remove the "upon request". Also, you mention in the README file that the code is released under the Apache License. Notwithstanding, the repository is under the CC-4.0 license. You should modify the license of the repository and include a copy of the Apache License with your code.
About the COSMO-CLM and CMAQ codes: We need that they are archived in a suitable repository too, and the webpages that you mention are not enough. You could want to check if they are already stored in one. For example, there are versions of both of them on Zenodo.org, although I do not know if they are exactly the ones that you have used.
The same applies to HAMSOM-ECOMSO. However, here the problem is worse, as the link that you provided in the manuscript does not work. Therefore, you must provide a suitable link.
Therefore, please, publish the mentioned codes in one of the appropriate repositories, and reply to this comment with the relevant information (link and DOI) as soon as possible, as it should be available for the Discussions stage. In this way, you must include in a potentially reviewed version of your manuscript the modified 'Code and Data Availability' section and DOIs for the codes, which must be available and open without the need to request access to anyone.
Juan A. Añel
Geosci. Model Dev. Exec. Editor

Citation: https://doi.org/10.5194/gmd-2021-427-CEC1
- AC3: 'Reply on CEC1', Johannes Bieser, 01 Feb 2023
  
  The comment was uploaded in the form of a supplement: https://gmd.copernicus.org/preprints/gmd-2021-427/gmd-2021-427-AC3-supplement.pdf
  
  Citation: https://doi.org/10.5194/gmd-2021-427-AC3

Peer review completion

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

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

ED: Publish as is (15 Feb 2023) by Andrew Yool

AR by Johannes Bieser on behalf of the Authors (15 Feb 2023) Manuscript

Short summary

MERCY is a 3D model to study mercury (Hg) cycling in the ocean. Hg is a highly harmful pollutant regulated by the UN Minamata Convention on Mercury due to widespread human emissions. These emissions eventually reach the oceans, where Hg transforms into the even more toxic and bioaccumulative pollutant methylmercury. MERCY predicts the fate of Hg in the ocean and its buildup in the food chain. It is the first model to consider Hg accumulation in fish, a major source of Hg exposure for humans.