Calibration of a simple and a complex model of global marine biogeochemistry

Kriest, Iris

doi:https://doi.org/10.5194/bg-14-4965-2017

Articles | Volume 14, issue 21

https://doi.org/10.5194/bg-14-4965-2017

© Author(s) 2017. This work is distributed under
the Creative Commons Attribution 3.0 License.

Special issue:

Progress in quantifying ocean biogeochemistry – in honour...

https://doi.org/10.5194/bg-14-4965-2017

© Author(s) 2017. This work is distributed under
the Creative Commons Attribution 3.0 License.

Articles | Volume 14, issue 21

Research article

|

08 Nov 2017

Research article |

| 08 Nov 2017

Calibration of a simple and a complex model of global marine biogeochemistry

Iris Kriest

Download

Final revised paper (published on 08 Nov 2017)
Supplement to the final revised paper
Preprint (discussion started on 06 Mar 2017)
Supplement to the preprint

Interactive discussion

Status: closed

AC: Author comment | RC: Referee comment | SC: Short comment | EC: Editor comment

- Printer-friendly version

- Supplement

RC1: 'Review of “Calibration of a simple and a complex model of global marine biogeochemistry” by I. Kriest', Anonymous Referee #1, 05 Apr 2017
- AC1: 'Reply to referee 1', Iris Kriest, 30 May 2017
RC2: 'Referee Comment', Friederike Hoffmann, 10 Apr 2017
- AC2: 'Reply to referee 2', Iris Kriest, 30 May 2017
RC3: 'Review of "Calibration of a simple and a complex model of global marine biogeochemistry"', Anonymous Referee #3, 12 Apr 2017
- AC3: 'Reply to referee 3', Iris Kriest, 30 May 2017
RC4: 'Calibration of a simple and a complex model of global marine biogeochemistry', Anonymous Referee #4, 24 Apr 2017
- AC4: 'Reply to referee 4', Iris Kriest, 30 May 2017

Peer-review completion

AR: Author's response | RR: Referee report | ED: Editor decision

ED: Reconsider after major revisions (06 Jun 2017) by Christoph Heinze

AR by Iris Kriest on behalf of the Authors (23 Jun 2017) Author's response Manuscript

ED: Referee Nomination & Report Request started (28 Jun 2017) by Christoph Heinze

RR by Anonymous Referee #1 (13 Jul 2017)

RR by Anonymous Referee #4 (14 Jul 2017)

RR by Anonymous Referee #3 (18 Jul 2017)

Suggestions for revision or reasons for rejection

My thanks to the author for responding at length to my comments. In general, I am happy with the responses to my original comments. I have a few remaining comments, but these are also relatively minor, and it is even possible that they have been addressed already. I also have a “major” comment – see immediately below this paragraph – but I think this is more for the journal than for the author. In general, my recommendation is publish following minor revision.

Per my remark above, my “major” comment is that it is not easy to properly understand how the author has modified their manuscript in response to referee comments. The author responses are very much in the form of “I will alter the text to address this” statements rather than “I have altered the text as follows to address this” statements. As a result, I cannot be entirely sure about how the manuscript has changed. To be fair, the author has included a tracked changes version of the manuscript but it is not a straightforward task to work out what exactly has changed in response to particular criticisms. I suspect that this stems from the journal’s preference for authors to respond to comments in a sort of dialogue with their referees before revising the manuscript, rather than the more conventional “response + revised manuscript” mode. In any case, I have generally taken the author at face value on the content of their changes in response to my criticisms, but have focused where changes are less clear / more contentious.

In the text below, I leave the “IK” designations in place to denote author comments. My new comments are marked up as “REF”

IK: This comment by the reviewer addresses two points: (a) the (lack of) introduction of RetroMOPS, as well as (b) its performance with respect to the skill metrics to biogeochemical tracers.

REF: I think I have misled the author as to what I was suggesting here. By drawing attention to the Kwiatkowski et al. (2014) paper, I was merely making the passing observation that increasing complexity does not necessarily lead to increasing performance. I certainly didn’t intend to suggest that a particular intercomparison was needed, nor that the authors of that paper presented anything like a definitive guide as to how this should be done (its choice of metrics is more a “lowest common denominator” analysis). As such, section 3.3 of the revised manuscript, while largely welcome, may over-emphasise the latter.

IK: In the model nitrogen fixation balances denitrification on large time and space scales. It depends on biogeochemical parameters and circulation, how fast the two processes are connected (see also Kriest and Oschlies, 2015). I will add "in the long term" in a revised version of the manuscript, and add some words on the potential spatial distinction.

REF: This is fine, but there’s an earlier mention of nitrogen fixation for MOPS that makes the comment below but does not qualify the time/space scales point:

“Loss of fixed nitrogen is balanced by a simple parameterisation of nitrogen fixation by cyanobacteria, which relaxes the nitrate-to-phosphate ratio to d with a time constant, μ⇤NFix.”

REF: Just to clarify, but when I previously said “Pg. 7, lines 24-25: While the normalisation to global concentrationshould help with N and P (since they are related linearly through the ocean)”, what I meant was that N and P in the real ocean have a relatively strong linear relationship (unlike, say, either element and silicon), not that they are (or should be) tightly linearly controlled in the model. Sorry for any confusion on this point.

IK: In MOPS’ optimisations, R_-O2:P did not show any significant deviations from its default value of 170, so I decided to keep this fixed, and only vary those parameters, that relate to the parts of RetroMOPS that are very different to MOPS (see above). I will try to explain this better in the revision.

REF: This line of argument sounds sensible. However, I can’t find it being mentioned explicitly (or clearly so) in the manuscript. While I can see that the value of RetroMOPS is that of the optimised MOPS, it isn’t clear enough to me that this has been deliberately done. Also, this approach also overlooks the faint possibility that while a parameter is insensitive in one model, that it might not be in another. An acknowledgement of this might be worth making.

IK: Because the reduced data set did not show any large changes in estimated parameters or misfit of MOPS, and because I assume that other issues might be more important to investigate in an optimisation context (circulation; components and form of the misfit function), I decided not to spend computational resources on RetroMOPS with reduced data set. This "reduced data" experiment was mainly aimed at investigating the potential effect of circulation in the equatorial Pacific on the parameter estimate. Possibly due to its small area, the effect is almost negligible.

REF: Is this acknowledged in the text at all?

REF: Pg. 16, line 19: correct “mesopelagial” to “mesopelagic”.

Hide

ED: Publish subject to minor revisions (Editor review) (20 Jul 2017) by Christoph Heinze

AR by Iris Kriest on behalf of the Authors (19 Aug 2017) Author's response Manuscript

ED: Publish subject to technical corrections (01 Sep 2017) by Christoph Heinze

AR by Iris Kriest on behalf of the Authors (08 Sep 2017) Manuscript

Download

Article (8904 KB)
Full-text XML

Short summary

Early biogeochemical ocean models were of a simple structure, with few biogeochemical components. I here investigate whether additional biological complexity improves the fit with respect to observed global climatologies of annual mean nutrients and oxygen. After optimisation against these tracers a simple model fits observations almost as well as a more complex one, also with respect to independent estimates of global biogeochemical fluxes.