Last Millennium Reanalysis with an expanded proxy database and seasonal proxy modeling

Tardif, Robert; Hakim, Gregory J.; Perkins, Walter A.; Horlick, Kaleb A.; Erb, Michael P.; Emile-Geay, Julien; Anderson, David M.; Steig, Eric J.; Noone, David

doi:https://doi.org/10.5194/cp-15-1251-2019

Articles | Volume 15, issue 4

https://doi.org/10.5194/cp-15-1251-2019

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

https://doi.org/10.5194/cp-15-1251-2019

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

Articles | Volume 15, issue 4

Research article

|

05 Jul 2019

Research article |

| 05 Jul 2019

Last Millennium Reanalysis with an expanded proxy database and seasonal proxy modeling

Robert Tardif, Gregory J. Hakim, Walter A. Perkins, Kaleb A. Horlick, Michael P. Erb, Julien Emile-Geay, David M. Anderson, Eric J. Steig, and David Noone

Download

Final revised paper (published on 05 Jul 2019)
Supplement to the final revised paper
Preprint (discussion started on 18 Sep 2018)
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 cp-2018-120 “Last Millennium Reanalysis with an expanded proxy database and seasonal proxy modeling” by Robert Tardif et al.', Anonymous Referee #1, 19 Oct 2018
- AC1: 'Response to Referee #1', Robert Tardif, 18 Jan 2019
RC2: 'Review of Last Millennium Reanalysis with an expanded proxy database and seasonal proxy modeling by Tardif et al.', Anonymous Referee #2, 18 Nov 2018
- AC2: 'Response to Referee #2', Robert Tardif, 18 Jan 2019
RC3: 'Review of Tardif et al.', Anonymous Referee #3, 27 Nov 2018
- AC3: 'Response to Referee #3', Robert Tardif, 18 Jan 2019

Peer-review completion

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

ED: Reconsider after major revisions (04 Feb 2019) by Hans Linderholm

AR by Robert Tardif on behalf of the Authors (18 Mar 2019) Author's response Manuscript

ED: Referee Nomination & Report Request started (01 Apr 2019) by Hans Linderholm

RR by Anonymous Referee #1 (23 Apr 2019)

Suggestions for revision or reasons for rejection

Review of “Last Millennium Reanalysis with an expanded proxy database and seasonal proxy modeling” by Robert Tardif et al. after major revisions

The manuscript has improved significantly due to the major revisions that have been conducted by the authors. Reconstruction skill is now evaluated before the 20th century. Additionally, multi-decal variability is discussed with regard to the selected input data set and a comparison with previous statistical reconstructions has been added. Therefore, I suggest publication after a few minor improvements and clarifications.

What I would like to see most is, how spatial reconstruction skill changes over time, i.e. which regions are corrected by data assimilation with a sparse paleodata. This could be shown for instance with maps that indicate the reductions of ensemble spread at a few time slices, e.g. 250, 750, 1250 and 1750 CE. Something similar is presented in Fig. 2b but just for global mean temperature. Looking at the time series in Fig. 2c it appears like the reconstruction is very close to climatology in the first millennium due to static offline assimilation approach and few paleodata, which very available at this time. This tendency towards climatology, inherent to this data assimilation approach, should be discussed. The main conclusion from this paper is the improvement compared to Hakim et al, 2016 mostly based on validation in the 20th century. In Fig. 2b it appears like the major improvement compared to the prototype version would be just in the 20th century, which is mostly used for evaluation in this study of not of scientific interest. This needs to be discussed, too.

Abstract:
• Line 5: Are citations allowed in the abstract?
• I suggest adding one sentences about the important result that the choice of the input data set determines the amplitude of multi-decadal to centennial variability in the reconstruction.

2.2. Climate proxies:
• Are possible duplicate records in the PAGES2k and Breitenmoser et al. 2014 collection somehow identified/treated?
• Tree-ring density records discussed here do not appear in Fig. 1 but only in a separate Fig. 8. I suggest merging Fig. 8 into a third panel of Fig. 1 and having an overview of all three tested proxy networks this way.

2.3. Climate model prior information
• Does the CCSM ensemble really only cover the period 850 to 1850 CE? If so, it would be worth discussing why an ensemble without 20th century greenhouse gas forcing performs so well in reconstructing 20th century climate.

2.4.1 Seasonality
• Line 21: in neither of the cited papers I find indications that the growing season could last until November as in the used season (JJASON).

2.4.2
• Line 31: It would be helpful for the reader to know a little more details about how the withheld proxy data with low signal-to-noise ratios will be used to measure reconstruction error.

3. The updated analysis:
• Page 11, line 1: “isn’t” -> is not
• Page 11, line 14: Is the abbreviation BE really needed? It is used very rarely.
• Page 11, line 17: define or explain CE briefly. Either with an equation or as “a RMSE based skill score with climatology as a reference”.

4.1 Proxy system models
• Page 13, line 2: the so-called “expert metadata” is also mainly derived by identification of highest correlation coefficients and hence not really different from the so-called “objective” method. The difference is rather that more variations of growing season months are possible and only correlation with temperature was used.

4.2 Covariance localization
• Page 14, line 24: add a reference to the Fig. S4

Hide

RR by Anonymous Referee #3 (06 May 2019)

ED: Publish subject to minor revisions (review by editor) (07 May 2019) by Hans Linderholm

AR by Robert Tardif on behalf of the Authors (28 May 2019) Author's response Manuscript

ED: Publish subject to technical corrections (04 Jun 2019) by Hans Linderholm

AR by Robert Tardif on behalf of the Authors (11 Jun 2019) Author's response Manuscript

Short summary

An updated Last Millennium Reanalysis is presented, using an expanded multi-proxy database, and proxy models representing the seasonal characteristics of proxy records, in addition to the dual sensitivity to temperature and moisture of tree-ring-width chronologies. We show enhanced skill in spatial reconstructions of key climate variables in the updated reanalysis, compared to an earlier version, resulting from the combined influences of the enhanced proxy network and improved proxy modeling.