CO Editor

Interactive discussion

Status: closed

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

- Printer-friendly version

- Supplement

RC1: 'Review of the article by Paprotny et al.', Anonymous Referee #1, 19 Jan 2017
RC2: 'Review of nhess-2017-4', Anonymous Referee #2, 11 Mar 2017
AC1: 'Reply to Referees', Dominik Paprotny, 11 Apr 2017

Peer-review completion

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

ED: Reconsider after major revisions (further review by Editor and Referees) (18 Apr 2017) by Paolo Tarolli

AR by Dominik Paprotny on behalf of the Authors (16 May 2017) Author's response Manuscript

ED: Referee Nomination & Report Request started (16 May 2017) by Paolo Tarolli

RR by Anonymous Referee #1 (16 May 2017)

RR by Anonymous Referee #2 (30 May 2017)

Suggestions for revision or reasons for rejection

The manuscript aims to derive a pan-European flood map using the statistical approach of Bayesian Network to derive flood discharge and a 1-D model to propagate flood discharge and identify flood prone areas. The same procedural chain is applied in order to capture the effect of climate change on the extension of flooded areas. The proposed method is compared with another pan-European flood map and with different local reference maps produced by different European countries.
This is the second time I read the paper, and the authors improved a little bit their manuscript but my main concerns remain unchanged with regard to some important points:
1) I suggested describing in detail in the paper the structure of Bayesian network used and to remove equations (1) and (2) since these are well known to every hydrologist and are parts of a commercial model (SOBEK). The authors have left the DSV equations (absolutely not useful for the general reader) and do not add any explanation (even synthetic) about the Bayesian network.
2) About the model calibration process again my observation is that they cannot calibrate their model by comparing their results to JRC map (which was no calibrated). They are trying to reproduce another map maybe created with a similar model. It is very strange that they use a single value of roughness (or a couple of values) for all the test sites and it is strange that their approach is low sensitive to the choice of a key parameter like roughness.
3) Honestly I do not understand why they cannot apply K-index (“The Kappa index is not applicable in this situation, as it would requires four cases, including a case when both datasets agree that a given area will not be flooded”). If they have both dataset, it is very simple to derive the four cases even the case not flooded in both dataset.

Hide

ED: Publish subject to minor revisions (further review by Editor) (02 Jun 2017) by Paolo Tarolli

AR by Dominik Paprotny on behalf of the Authors (10 Jun 2017) Author's response Manuscript

ED: Publish as is (23 Jun 2017) by Paolo Tarolli

AR by Dominik Paprotny on behalf of the Authors (24 Jun 2017)