Recent development of the Met Office operational ocean forecasting system: an overview and assessment of the new Global FOAM forecasts

Blockley, E. W.; Martin, M. J.; McLaren, A. J.; Ryan, A. G.; Waters, J.; Lea, D. J.; Mirouze, I.; Peterson, K. A.; Sellar, A.; Storkey, D.

doi:https://doi.org/10.5194/gmd-7-2613-2014

Articles | Volume 7, issue 6

https://doi.org/10.5194/gmd-7-2613-2014

© Author(s) 2014. This work is distributed under
the Creative Commons Attribution 3.0 License.

Special issue:

Nucleus for European Modelling of the Ocean - NEMO

https://doi.org/10.5194/gmd-7-2613-2014

© Author(s) 2014. This work is distributed under
the Creative Commons Attribution 3.0 License.

Articles | Volume 7, issue 6

Model description paper

|

10 Nov 2014

Model description paper |

| 10 Nov 2014

Recent development of the Met Office operational ocean forecasting system: an overview and assessment of the new Global FOAM forecasts

E. W. Blockley, M. J. Martin, A. J. McLaren, A. G. Ryan, J. Waters, D. J. Lea, I. Mirouze, K. A. Peterson, A. Sellar, and D. Storkey

Download

Final revised paper (published on 10 Nov 2014)
Supplement to the final revised paper
Preprint (discussion started on 29 Nov 2013)

Interactive discussion

Status: closed

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

- Printer-friendly version

- Supplement

RC C2363: 'FOAM v12', Gary B. Brassington, 15 Jan 2014
- AC C2965: 'Response to Review #1 (Gary Brassington)', Ed Blockley, 08 Apr 2014
RC C2435: 'Review of Manuscript gmd-2013-158', Anonymous Referee #2, 24 Jan 2014
- AC C2984: 'Response to Review #2 (Anonymous)', Ed Blockley, 08 Apr 2014

Peer-review completion

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

AR by Ed Blockley on behalf of the Authors (08 Apr 2014) Author's response

ED: Reconsider after major revisions (24 Apr 2014) by Olivier Marti

AR by Ed Blockley on behalf of the Authors (20 Jun 2014) Author's response Manuscript

ED: Referee Nomination & Report Request started (30 Jun 2014) by Olivier Marti

RR by Anonymous Referee #2 (25 Jul 2014)

RR by Anonymous Referee #3 (13 Aug 2014)

Suggestions for revision or reasons for rejection

Review of the revised manuscript of "Recent development of the Met Office operational ocean forecasting system: an overview and assessment of the new Global FOAM forecasts" by E. W. Blockley et al.

This manuscript offers a detailed description on the latest operational version of the UKMO global ocean data assimilation system, FOAM, and also evaluates the accuracy of the data assimilative fields and the short-range ocean forecasts. I admit that the detailed description on FOAM and information of its accuracy is beneficial to the readers of the Geoscientific Model Development.
I was nominated as a new reviewer instead of the first reviewer (Dr. Gary Brassington), who is not available for reviewing the revised version. Since the editor did not give me enough time for a complete review, I basically took over the comments and concerns made by the first reviewer.
In my opinion, the manuscript is revised adequately according to the first and second reviewers. However, I found that a couple of concerns made by the first reviewer are still not resolved. Therefore I recommend accepting this manuscript after a minor revision considering the following comments.

Comments

Line 23: I suggest adding “short-range ocean” before “forecasting accuracy” in order to clarify the assessment is performed only for the short-range ocean forecasts although FOAM is also used for seasonal predictions.

I think Figure 8 indicates that the data assimilation is not effective for the sea ice reproduction in seasonal forecasts, especially in the Arctic Ocean, because the sea ice extent quickly goes back to the model climatology in the 5-day forecasts although the drift of v12 is smaller than that of v11. Authors may want to make some comments about this problem.

Line 639: The first reviewer suggested quantifying the statement “The largest improvements at high latitudes can be mainly attributed to the NEMOVAR data assimilation scheme fitting small-scale features better than the old OCNASM scheme” with a metric. Authors did not answer to this comment. It may be difficult to quantify this with a metric. However, is it possible to state some basis for the statement. Do Waters et al. (2014) examine about this?

Section 4.2.1: The first reviewer noted that the SST observations are not completely independent in the forecast assessment. Although the authors did not answer to the comment, it should be noted in the text. Actually, this may important for analyzing the reason why the subsurface tracer accuracy has a gap between T-12h and T+12h although the SST accuracy has no gap.

Section 4.2.2: I believe that the gap of the subsurface tracer accuracy between T-12h and T+12h mainly caused by the fact that the forecasts are compared with almost independent observation although observations used for the assessment of analysis values (T-12h) are assimilated in the assimilative field itself and therefore not independent. The observations used for the forecast assessment are almost independent because subsurface observations are mainly performed by Argo floats, which make observation generally every 10 days. It should be noted that the situation is different from the assessment of the SST forecasts (see the comment above.) The authors are also aware about this and describe it in the response to the first reviewer (the last reply on the summary section.) I think it is almost inevitable that the score becomes better if non-independent data are used as reference because data assimilation generally nudges the assimilative field to the observation data. In particular, it necessarily happens when the horizontal scale of the sampling is larger than the scale that the effect of an observation extends. I think this should not be regarded as over-fitting as far as the scale of the effect is enough larger than the scale of the representativeness error and the physical balance in the model is not severely destroyed. I think that a similar situation must also occur for SST although this does not revealed in the assessment because SST data is not independent even in the forecast assessment. Authors should discuss the gap between T-12 and T+12 more carefully in the text.

Section 5: The first reviewer recommended emphasizing that the surface velocities are independent observation in the summary. I support this comment, although it seems to be ignored in the author response and the revised manuscript.

Appendix A “The update to FOAM v11 also saw the extension of the operational FOAM system to include an additional 24 hour data assimilation cycle ...”: Does it mean that the assimilation cycle from T-48h to T+24h is not re-performed in the operation in v10? I think the description is still not clear. It would be clear if it is noted that data not arriving in 24 hours are ignored in v10. And the expression “the 48 hour observation window” is very confusing. How about revising to “the retrospective assimilation cycle enables the system to use more late arriving observations.”?

Hide

ED: Publish subject to minor revisions (Editor review) (01 Sep 2014) by Olivier Marti

AR by Ed Blockley on behalf of the Authors (10 Sep 2014) Author's response Manuscript

ED: Publish as is (15 Sep 2014) by Olivier Marti