Identifying robust bias adjustment methods for European extreme precipitation in a multi-model pseudo-reality setting

Schmith, Torben; Thejll, Peter; Berg, Peter; Boberg, Fredrik; Christensen, Ole Bøssing; Christiansen, Bo; Christensen, Jens Hesselbjerg; Madsen, Marianne Sloth; Steger, Christian

doi:https://doi.org/10.5194/hess-25-273-2021

Articles | Volume 25, issue 1

https://doi.org/10.5194/hess-25-273-2021

© Author(s) 2021. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/hess-25-273-2021

© Author(s) 2021. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 25, issue 1

Research article

|

18 Jan 2021

Research article |

| 18 Jan 2021

Identifying robust bias adjustment methods for European extreme precipitation in a multi-model pseudo-reality setting

Torben Schmith, Peter Thejll, Peter Berg, Fredrik Boberg, Ole Bøssing Christensen, Bo Christiansen, Jens Hesselbjerg Christensen, Marianne Sloth Madsen, and Christian Steger

Download

Final revised paper (published on 18 Jan 2021)
Preprint (discussion started on 07 Jul 2020)

Interactive discussion

Status: closed

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

- Printer-friendly version

- Supplement

RC1: 'Review of - Identifying robust bias adjustment methods for extreme precipitation in a pseudo-reality setting', Anonymous Referee #1, 21 Jul 2020
- AC1: 'Authors' response to referee #1', Torben Schmith, 14 Aug 2020
RC2: 'Review of 'Identifying robust bias adjustment methods for extreme precipitation in a pseudo-reality setting'', Jorn Van de Velde, 14 Sep 2020
- AC2: 'Reply to referee #2', Torben Schmith, 14 Oct 2020
SC1: 'Comment on `Identifying robust bias adjustment methods for extreme precipitation in a pseudo-reality setting’', Timo Kelder, 15 Sep 2020
- AC3: 'Author reply to short comment #1', Torben Schmith, 14 Oct 2020
RC3: 'Review of "Identifying robust bias adjustment methods for extreme precipitation in a pseudo-reality setting"', Anonymous Referee #3, 18 Sep 2020
- AC4: 'Author reply to referee #3', Torben Schmith, 14 Oct 2020

Peer-review completion

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

ED: Publish subject to revisions (further review by editor and referees) (16 Oct 2020) by Nadav Peleg

AR by Torben Schmith on behalf of the Authors (11 Nov 2020)

ED: Referee Nomination & Report Request started (16 Nov 2020) by Nadav Peleg

RR by Anonymous Referee #1 (16 Nov 2020)

Suggestions for revision or reasons for rejection

I went through the authors response to my earlier comments and generally agree with the modifications they have made. The paper is more complete and better suited for publication.

My only other query is whether the authors noted any systematic changes in the relative errors for the two durations for the 12.5km runs compared to the 50km runs? This is an issue that comes up with RCM simulations often as the computational expense in the higher resolution runs is considerable. Any comments could be useful to others.

Other than this, I noted down some specific issues in the order I read the paper. these are:

l65 - Authors should note the following two papers which are relevant here
Kim, Y., et al. (2020). "Impact of bias correction of regional climate model boundary conditions on the simulation of precipitation extremes." Climate Dynamics 55(11): 3507-3526.
Kim, S., et al. (2020). "Quantification of Uncertainty in Projections of Extreme Daily Precipitation." Earth and Space Science 7(8).
The first paper assesses the impact on extreme precipitation simulations once lateral and lower boundary biases are corrected. This is directly relevant to the present study. The second presents estimates of uncertainty in extreme precipitation simulations after bias correction for a range of models. I suggest the second one as it may give some guidance to authors to contrast their uncertainty estimates against those presented in the paper to gauge the extent of improvements made.

l72 and 73 - the referenced studies are not directly relevant to the present work because of the psuedoCC run. A bias corrected run would be a more appropriate comparison as noted in the above two papers.

l97 - I have long felt that in hydrological modelling settings, the issue of persistence is more important than simply correcting means.The Nesting Bias Correction papers and its variants in multivariate settings are my recommended alternatives where a water nalance simulation is needed. I urge the authors to exapnd their discussion about bias correction to mention this line of thought. The paper below given details of the software and how it should be used.
Mehrotra, R., et al. (2018). "A software toolkit for correcting systematic biases in climate model simulations." Environmental Modelling and Software 104: 130-152.

l205 - In my experience the fitting of the GPD creates instability in the shape and scale parameters, especially given the short record that are typically used. It would be good for the authors to discuss whether unstable parameter values resulted and whether the RE estimates are sensitive to such instability in a systematic manner.

l328 - The 24h results show 3 cases where there is a decrease into the future. No such decrease is present in the hourly simulations. Is this a pattern that was common across the other regions the study focussed on? Would there be any reason for this? Perhaps linked to the instability in the GPD parameters?

Hide

ED: Publish subject to minor revisions (review by editor) (16 Nov 2020) by Nadav Peleg

AR by Torben Schmith on behalf of the Authors (23 Nov 2020) Author's response Manuscript

ED: Publish as is (25 Nov 2020) by Nadav Peleg

AR by Torben Schmith on behalf of the Authors (27 Nov 2020) Manuscript

Short summary

European extreme precipitation is expected to change in the future; this is based on climate model projections. But, since climate models have errors, projections are uncertain. We study this uncertainty in the projections by comparing results from an ensemble of 19 climate models. Results can be used to give improved estimates of future extreme precipitation for Europe.