Synthesizing long-term sea level rise projections – the MAGICC sea level model v2.0

Nauels, Alexander; Meinshausen, Malte; Mengel, Matthias; Lorbacher, Katja; Wigley, Tom M. L.

doi:https://doi.org/10.5194/gmd-10-2495-2017

Articles | Volume 10, issue 6

https://doi.org/10.5194/gmd-10-2495-2017

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

https://doi.org/10.5194/gmd-10-2495-2017

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

Articles | Volume 10, issue 6

Development and technical paper

|

30 Jun 2017

Development and technical paper |

| 30 Jun 2017

Synthesizing long-term sea level rise projections – the MAGICC sea level model v2.0

Alexander Nauels, Malte Meinshausen, Matthias Mengel, Katja Lorbacher, and Tom M. L. Wigley

Download

Final revised paper (published on 30 Jun 2017)
Preprint (discussion started on 05 Oct 2016)

Interactive discussion

Status: closed

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

- Printer-friendly version

- Supplement

SC1: 'Executive Editor Comment on "Synthesizing long-term sea level rise projections - the MAGICC sea level model"', Astrid Kerkweg, 24 Oct 2016
- AC1: 'Synthesizing long-term sea level rise projections – the MAGICC sea level model v2.0', Alexander Nauels, 24 Mar 2017
RC1: 'Solid model description paper, could use more discussion of related work', Anonymous Referee #1, 10 Feb 2017
- AC2: 'Author Responses to RC1', Alexander Nauels, 24 Mar 2017
RC2: 'Review Nauels et al, GMD, 2016', Anonymous Referee #2, 24 Feb 2017
- AC3: 'Author Responses to RC2', Alexander Nauels, 24 Mar 2017

Peer-review completion

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

AR by Alexander Nauels on behalf of the Authors (24 Mar 2017) Author's response Manuscript

ED: Referee Nomination & Report Request started (29 Mar 2017) by David Ham

RR by Anonymous Referee #2 (16 Apr 2017)

Suggestions for revision or reasons for rejection

General Comments:

The authors are to be applauded for a careful and overall quite responsive revision that addresses many of the concerns and points raised in the first reviews. Below, I discuss one still open issue that requires an editorial decision and make a few minor (mostly stylistic) suggestion.

Specific Points

1) Major

It is still not clear to me how the manuscript follows the data and code policy of GMD. As quoted in the last review, the policy (http://www.geoscientific-model-development.net/about/code_and_data_policy.html) states:

“Preferably, this section should contain the instructions for obtaining the model code and/or data, either from the supplement or from an archive with a digital object identifier (DOI). Suitable repositories can be found at the Registry of Research Data Repositories, e.g. ZENODO for model code. After the paper is accepted, a link to the GMD paper should be added to the metadata of the archive.”

From the gitlab location specified in the paper, one gets: “Please note that the MAGICC sea level model is not a stand-alone tool. It is fully integrated into the simple carbon-cycle climate model MAGICC. The MAGICC model is described in Meinshausen et al. (2011).” as well as: “Corresponding calibration input data as well as output from the conducted experiments is available on request. Component-specific additional reference datasets should be requested from the authors of the individual studies. ”

The main issues I see are that the code is not a stand-alone program and that the data and script required to reproduce the analysis are not supplied via the repository. There may be perfectly understandable reasons for this, but they are not explained in the manuscript as far as I can tell and they are still unclear to me. Put simply: Many researchers would have severe difficulties (certainly would need to spend a lot of time) to reproduce the results. Why not provide a wrapper that calls the currently provided subroutine, the data needed to force the model, as well as the scripts that plot the figures? This would drastically increase the ease at which the results could be reproduced and the use of the (neat, we seem to agree here) model by others. It would likely also safe the authors time in the long run.
In the end, this is the editor’s decision.

a. Font sizes in several figures are too small to read (e.g., Figure 2).

2) Minor
a. P1,L6: maybe shift from advice to analysis?
b. P1,L8: bound to -> by?
c. P2,L33: attempts -> approaches (attempt seems to be talking down)
d. P3,l3: the same -> constant?
e. P11, new text: It is not the models that show convergence but the calibration method. In addition, please address the question of convergence to what?
f. P13, l13: I would spell out numbers <= 10.
g. P16, l33: It is not the question whether the authors are confident. Rather, what are the fulfilled criteria that provides backup for the claim. Maybe just reword the statement, as you specify criteria in the previous sentence.
h. P17, l.29: Please explain what the “reference” response is.

Hide

ED: Publish subject to minor revisions (Editor review) (04 May 2017) by David Ham

Dear Alex,

Apologies for the delay in getting back to you. As you can see, the remaining reviewer is largely satisfied with the modifications to the paper, but still has concerns about code availability and reproducibility. While the release of MAGICC is clearly beyond your control, I think it ought at least to be possible to publish data and code such that someone who had a MAGICC licence could reproduce the work. For this a few things are needed, but you should be able to do them without too much difficulty:

1. Permanent archive of the exact version of your code. The GitLab URL has two issues: the first is that it's not permanent (the repository could be taken down tomorrow if the authors decided to), the second is that a GitLab URL does not indicate exactly which version of the code was used. The simplest way to achieve this is to upload a snapshot of the exact version of the code to an archival site, preferably one which issues DOIs. Zenodo is a good choice of archive, though others are available. You could either manually upload to Zenodo, or push a copy of your repository to GitHub and use the GitHub Zenodo integration.

2. Permanent archive of the configuration and inputs used in the tests. This could be another Zenodo archive. Where you have used published CMIP input data, there is no need to duplicate this, so long as there is sufficient detail that another scientist could easily retrieve exactly the same input data. At the moment, a potential reproducer would be unable to ascertain which data they need to ask for.

3. Document exactly the version of MAGICC used, so that a MAGICC licence holder could run the same version you ran.

While it's unfortunate that the full code base can't be published, I think the above would be a sufficient "best effort" at reproducibility in the circumstances.

Regards,

David

Hide

AR by Alexander Nauels on behalf of the Authors (09 May 2017) Author's response Manuscript

ED: Publish as is (12 May 2017) by David Ham

AR by Alexander Nauels on behalf of the Authors (15 May 2017) Manuscript

ED: Publish subject to minor revisions (Editor review) (20 Jun 2017) by David Ham

AR by Alexander Nauels on behalf of the Authors (21 Jun 2017) Author's response Manuscript

ED: Publish as is (22 Jun 2017) by David Ham

AR by Alexander Nauels on behalf of the Authors (24 Jun 2017)

Short summary

The MAGICC sea level model projects global sea level rise by emulating process-based estimates for all major sea level drivers and applying them to available climate scenarios and their extensions to 2300. The MAGICC sea level projections are well within the ranges of the fifth IPCC assessment report. Due to its efficient structure, this emulator is a powerful tool for exploring sea level uncertainties and investigating sea level responses for a wide range of climate mitigation pathways.