The Brazilian Earth System Model ocean–atmosphere (BESM-OA) version 2.5: evaluation  of  its CMIP5 historical simulation

Veiga, Sandro F.; Nobre, Paulo; Giarolla, Emanuel; Capistrano, Vinicius; Baptista Jr., Manoel; Marquez, André L.; Figueroa, Silvio Nilo; Bonatti, José Paulo; Kubota, Paulo; Nobre, Carlos A.

doi:https://doi.org/10.5194/gmd-12-1613-2019

Articles | Volume 12, issue 4

https://doi.org/10.5194/gmd-12-1613-2019

© Author(s) 2019. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/gmd-12-1613-2019

© Author(s) 2019. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 12, issue 4

Model evaluation paper

|

24 Apr 2019

Model evaluation paper |

| 24 Apr 2019

The Brazilian Earth System Model ocean–atmosphere (BESM-OA) version 2.5: evaluation of its CMIP5 historical simulation

Sandro F. Veiga, Paulo Nobre, Emanuel Giarolla, Vinicius Capistrano, Manoel Baptista Jr., André L. Marquez, Silvio Nilo Figueroa, José Paulo Bonatti, Paulo Kubota, and Carlos A. Nobre

Download

Final revised paper (published on 24 Apr 2019)
Preprint (discussion started on 22 Jun 2018)

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 "The Brazilian Earth System Model version 2.5: Evaluation of its CMIP5 historical simulation"', Astrid Kerkweg, 09 Jul 2018
- AC1: 'Comment: Executive Editor comment on "The Brazilian Earth System Model version 2.5: Evaluation of its CMIP5 historical simulation"'', Sandro Veiga, 11 Jul 2018
RC1: 'minor revisions', Anonymous Referee #1, 20 Jul 2018
- AC2: 'Response to Referee#1', Sandro Veiga, 30 Oct 2018
RC2: 'Interactive comment on “The Brazilian Earth System Model version 2.5: Evaluation of its CMIP5 historical simulation” by Sandro F. Veiga et al.', Anonymous Referee #2, 25 Jul 2018
- AC3: 'Response to Referee#2', Sandro Veiga, 30 Oct 2018

Peer-review completion

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

AR by Sandro Veiga on behalf of the Authors (31 Oct 2018)

ED: Referee Nomination & Report Request started (06 Nov 2018) by Qiang Wang

RR by Anonymous Referee #3 (30 Nov 2018)

Suggestions for revision or reasons for rejection

Comment on “The Brazilian Earth System Model version 2.5: Evaluation of its CMIP5 historical simulation”

This manuscript presents the general performance of the Brazilian Earth System Model (BESM) version 2.5 in a “historical” experiment. The authors evaluated the simulated modern climate mean states and major climate modes by comparing with the reanalysis/observation data. In general, the manuscript is well structured and well-written. However, it has also weaknesses, and substantial revisions are needed before I can recommend it for publication.

Major comments
1. The CMIP6 historical experiment forcing is released a longtime ago, and the BESM model is targeted at the CMIP6 project, why do not run the BESM with CMIP6 forcing data? Moreover, the GHG forcing is only one aspect of the historical forcing, what is the consideration to ignore other forcing? It is unfair to compare the GHG-historical simulation to the real-world observation, how to clarify the role of other forcing (e.g. aerosols)?

2. The model suffers a large TOA energy imbalance (about -4 W m-2) and surface imbalance (about 1.2 W m-2) from the beginning of historical simulation (e.g. 1850-1900). The authors should discuss the possible reason and causing of the energy bais. Is the imbalance due to the non-conservation in the AGCM or coupling process?

3. Figure 2 shows the 2-m air temperature is less response to the GHG forcing in BESM during 1850s-1960s, while it isn’t the case for any CMIP5 model. Any explanation for this unique feature? The simulated SST is much warmer than the ERSST v4 during the evaluation period (Fig. 10), how is the land surface temperature? How is the sea ice simulation under such a warm climate, please provide the figure of sea ice performance? It is also necessary to clarify whether the 1.5K SST warming affect the model climate variability or not.

4. As point out by reviewer 2, the coupled model spin-up period is very short. It is unclear, from the manuscript, whether the decadal variability is affected by model spin-up or not. Is the weakening of AMOC strength (Fig. 14) due to the model adjustment?

5. Some experience on coupled model tuning would be desirable.

Hide

RR by Anonymous Referee #2 (11 Dec 2018)

Suggestions for revision or reasons for rejection

Interactive comment on “The Brazilian Earth System Model version 2.5: Evaluation of its CMIP5 historical simulation” by Sandro F. Veiga et al. (Revised version)

The paper documents the present state of a new version of the Brazilian Earth System Model (BESM) that is supposed to be used in the CMIP6 Model Intercomparison. The authors describe mean state and variability characteristics comparing the simulated results from 20th century observations and reanalyses.
I appreciate that the authors have taken up the reviewer’s comments and have improved considerably the content and the readability of the manuscript. In particular, they clarified that there are accompanying papers describing the pre-industrial control simulation and respective stand-alone experiments with the ocean and atmosphere components. They also more precisely state that the model version described here is a somewhat intermediate step in the development toward a more complete Earth System Model. Understanding the manuscript as such makes it a useful contribution and reference document. Overall the presented analyses are carried out and interpreted in a competent way and reflect particular aspects that are important for the Brazilian region (South Atlantic/ Southern Hemisphere variability).
I suggest that the manuscript should be accepted after some minor revisions (see below).

Minor issues:

The authors have responded to the reviewer’s comment at length and explained in detail the spin-up procedure and the problem of model drift. Some of the new information has entered the revised manuscript, but some important issues are still missing in the main text, or is misleading.

The issue of model drift is still somewhat “put under the carpet”. The authors state that the surface quantities assume a stable state after 13 years in the PiControl run and suggest that drift is therefore no issue. However, the more than 1 W/m2 excess heat that the ocean receives over the entire simulation leads to strong heat uptake. The authors have shown that to the reviewer in their response (Fig 2). Howver, comparing this figure with the Hovmueller plot in the main text (Figure 13 revised version) suggest that a considerable part of that warming is simply drift and not, as the authors claim in the text, a response to global warming.
Since the historical and PiControl runs have been run in parallel, a solution would be to include the control run figure. Another possibility would be to discuss the model drift in terms of an integrated quantity, eg heat content or steric (thermo-haline) sea-level change, where the “real” changes in the historical run is estimated after the control run drift is subtracted.

I am also still confused by the description of the spin-up procedure. In the authors’ response it is explained that there has been a 100-year spin up (with a slightly different version of the coupled model) “as initial conditions for the piControl”. I don’t understand, why this is not mentioned in the main text and in the schematic and there is just talk of a 13 year spin-up. Was that done in addition to the 100 years?

Other minor issues:
Page 7, lns 3ff: How is river runoff and transfer to the ocean treated?

Page 20, lns 3ff: “appropriate depth”: Not sure if that is true, e.g. the zero crossing in the North Atlantic is below 2500 m. (Compare for example, with the recent paper by Menary et al.: https://agupubs.onlinelibrary.wiley.com/doi/abs/10.1029/2018MS001495).

Hide

ED: Reconsider after major revisions (13 Dec 2018) by Qiang Wang

AR by Sandro Veiga on behalf of the Authors (25 Jan 2019)

ED: Publish subject to minor revisions (review by editor) (30 Jan 2019) by Qiang Wang

AR by Sandro Veiga on behalf of the Authors (01 Mar 2019) Author's response Manuscript

ED: Publish as is (04 Mar 2019) by Qiang Wang

AR by Sandro Veiga on behalf of the Authors (20 Mar 2019) Manuscript

Download

The requested paper has a corresponding corrigendum published. Please read the corrigendum first before downloading the article.

Article (16628 KB)
Full-text XML

Short summary

This study evaluates the Brazilian Earth System Model with coupled ocean–atmosphere version 2.5 (BESM-OA2.5) and the effectiveness of reproducing the main characteristics of the atmospheric and oceanic variability in a real-life-based scenario of greenhouse gas increase (the CMIP5 historical protocol). The evaluation specifically focuses on how the model simulates the mean climate state, as well as the most important large-scale climate patterns.