Forecasting experiments of a dynamical–statistical model of the sea surface temperature anomaly field based on the improved self-memorization principle

Hong, Mei; Chen, Xi; Zhang, Ren; Wang, Dong; Shen, Shuanghe; Singh, Vijay P.

doi:https://doi.org/10.5194/os-14-301-2018

Articles | Volume 14, issue 2

https://doi.org/10.5194/os-14-301-2018

© Author(s) 2018. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/os-14-301-2018

© Author(s) 2018. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 14, issue 2

Research article

|

24 Apr 2018

Research article |

| 24 Apr 2018

Forecasting experiments of a dynamical–statistical model of the sea surface temperature anomaly field based on the improved self-memorization principle

Mei Hong, Xi Chen, Ren Zhang, Dong Wang, Shuanghe Shen, and Vijay P. Singh

Download

Final revised paper (published on 24 Apr 2018)
Preprint (discussion started on 07 Nov 2017)

Interactive discussion

Status: closed

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

- Printer-friendly version

- Supplement

RC1: 'Review of “Forecasting experiments of a dynamical-statistical model of the sea surface temperature anomaly field based on the improved self-memorization principle” by Hong et al..', Anonymous Referee #1, 26 Nov 2017
- AC1: 'Responses to reviewer#1:', Xi Chen, 06 Feb 2018
RC2: 'Review of Hong et al "Forecasting experiments of a dynamical-statistical model of sea surface temperature anomaly field based on self-memorization principle. 14 Dec 2017', Anonymous Referee #2, 14 Dec 2017
- AC2: 'Responses files to reviewer#2', Xi Chen, 06 Feb 2018

Peer-review completion

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

AR by Xi Chen on behalf of the Authors (06 Feb 2018) Author's response Manuscript

ED: Referee Nomination & Report Request started (07 Feb 2018) by Neil Wells

RR by Anonymous Referee #1 (12 Feb 2018)

RR by Anonymous Referee #3 (11 Mar 2018)

Suggestions for revision or reasons for rejection

With the objective of tackling the problem of inaccurate long-term El Niño Southern Oscillation (ENSO) forecast, this study develops a new dynamical-statistical forecast model of sea surface temperature anomaly (SSTA) field. A self-memorization principle is introduced to improve the dynamic reconstruction model, making the model more appropriate for describing such chaotic systems like ENSO. Compared with six mature models published previously, the present model has an advantage in prediction precision and time length, and is a novel exploration of the ENSO forecast method. Thus, I recommend its possible publication in the OS after some minor revisions.

Major:
1. Line633-663: I think there is no need to spend space here to introduce the definition of overfitting and you just have to give a reference. The discussion whether the model is overfitting or not is redundant and the discussion should focus on the good forecast results of new examples. This part of the discussion can be properly streamlined.
2. Line371: Why value was in the range 5 to 15? Is this the experience of the predecessors? Or is it the result of the author's own experiment?
3. Line597-615：The purpose of this section is to show that a certain impact on the pattern of the 2nd EOF will not make our forecast unstable. But this section here occupied the space too much and I think there is no need to make a list of table5 alone. The author just says the difference among forecast results of both TC and MAPE of five different sample data are less. Because the main purpose of this paper is to prove that the prediction accuracy of SST and ENSO of the model. Therefore, it is not necessary for table5 to list separately. It is recommended that this part should be simplified.
4. Line243-249: The author introduced the standardization of the data before modelling and it is too wordy here. The author just list the standardized formula here. So the section should be suggested to be simplified.
5. Line269 and 298: The abbreviation appears for the first time, and there is no need to repeat the Pearson correlation coefficient (CC).
6. Dynamic model recommendations are consistent, the topic is dynamical-st, followed by several dynamic text in a number of tense inconsistency, it is recommended to be consistent.

Minor:
1.Line649: “the forecast results of the Cross-validated” should be revised as” the forecast results of the cross-validated”.
2.Line 30, “dynamical reconstruction”is more appropriate
3.Line 35, ”pearson” should be ‘’Pearson”
4.Line 36, “approximately” should be ‘’approximate” or “about”
5.Line 38,”but” should be “but also”
6.Line 52,”influences” should be “influence”
7.Line 57,”gradually improved” should be “gradual improvement”
8.Line 67,”large a amount of” should be “ a large amount of ”
9.Line 75,”advanced” should be “ been advanced ”
10.Line 91,”This is because” should be “It is because that”

Referee Report: PDF

Hide

ED: Publish subject to minor revisions (review by editor) (12 Mar 2018) by Neil Wells

AR by Xi Chen on behalf of the Authors (18 Mar 2018) Author's response Manuscript

ED: Publish as is (19 Mar 2018) by Neil Wells

AR by Xi Chen on behalf of the Authors (23 Mar 2018)

Short summary

With the objective of tackling the problem of inaccurate long-term ENSO forecasts, a new forecasting model of the SSTA field was proposed based on a dynamic system reconstruction idea and the principle of self-memorization. The improved model was used to forecast the SSTA field. The forecasted SSTA fields of three types of events are accurate. The improved model also has good forecasting results of the ENSO index. So our model has an advantage in ENSO prediction precision and length.