Estimating exposure of residential assets to natural hazards in Europe using open data

Paprotny, Dominik; Kreibich, Heidi; Morales-Nápoles, Oswaldo; Terefenko, Paweł; Schröter, Kai

doi:https://doi.org/10.5194/nhess-20-323-2020

Articles | Volume 20, issue 1

https://doi.org/10.5194/nhess-20-323-2020

© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.

Special issue:

Global- and continental-scale risk assessment for natural...

https://doi.org/10.5194/nhess-20-323-2020

© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 20, issue 1

Research article

|

28 Jan 2020

Research article |

| 28 Jan 2020

Estimating exposure of residential assets to natural hazards in Europe using open data

Dominik Paprotny, Heidi Kreibich, Oswaldo Morales-Nápoles, Paweł Terefenko, and Kai Schröter

Download

Final revised paper (published on 28 Jan 2020)
Supplement to the final revised paper
Preprint (discussion started on 30 Sep 2019)
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 nhess-2019-313', Anonymous Referee #1, 14 Oct 2019
- AC1: 'Reply to Reviewer #1', Dominik Paprotny, 11 Nov 2019
RC2: 'Referee report on manuscript "Estimating exposure of residential assets to natural hazards in Europe using open data"', Anonymous Referee #2, 18 Oct 2019
- AC2: 'Reply to Reviewer #2', Dominik Paprotny, 11 Nov 2019

Peer-review completion

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

ED: Reconsider after major revisions (further review by editor and referees) (27 Nov 2019) by Sven Fuchs

AR by Dominik Paprotny on behalf of the Authors (28 Nov 2019) Author's response Manuscript

ED: Referee Nomination & Report Request started (02 Dec 2019) by Sven Fuchs

RR by Anonymous Referee #2 (04 Dec 2019)

RR by Anonymous Referee #1 (20 Dec 2019)

Suggestions for revision or reasons for rejection

My thanks to the authors for their efforts to take on board the recommendations made by myself and the other reviewer. I think that the quality of the manuscript and its accessibility has improved clearly.

I have only some minor remarks:

(1) Equation 1: In the original manuscript, c was 0.7. In the revised version, c=30%. Please double check this value.

(2) p.11, l.326: I suggest to simply write "low bias" instead of "negligible bias". Negligible might be a little bit too judgemental for a results section. Also, I think that the bias might be negligible from a practical point of view, but from a purely statistical assessment it seems that low building heights are overestimated and tall buildings are underestimated, entailing that a simple linear regression line will be systematically shifted compared to a 1:1 fit (45° line). It might be a good idea to add these two lines to Fig. 3.

(2) p.11, l.328: "An out-of-sample was also carried out ..." - There is a noun missing after sample. Alternatively, I would just refer to this procedure as a standard leave-one-out cross-validation?

(3) p. 14, l.33f: Just a thought: Isn't "For comparative purposes" a better motivation than "Due to the moderate accuracy"?

(4) p.14, l.33: I suggest to replace "automated data-mining method" with a term clearer describing random forests, such as e.g. "ensemble learning method". The term "data mining" can be misleading, since it is mainly used in a pattern extraction context.

(5) p.14. l.36ff: One reason for the low performance of the random forest might be that hyperparameters (such as the number of variables to possibly split at in each node, minimal node size, etc) were not tuned properly, and the rather small number of trees. However, since this is not the main focus of this paper, this is probably not much of an issue.

(6) Figure 2: x and y axis labels are still missing.

Hide

ED: Publish subject to minor revisions (review by editor) (20 Dec 2019) by Sven Fuchs

AR by Dominik Paprotny on behalf of the Authors (22 Dec 2019) Author's response Manuscript

ED: Publish as is (07 Jan 2020) by Sven Fuchs

Download

Article (4879 KB)
Full-text XML

Short summary

Houses and their contents in Europe are worth trillions of euros, resulting in high losses from natural hazards. Hence, risk assessments need to reliably estimate the size and value of houses, including the value of durable goods kept inside. In this work we show how openly available or open datasets can be used to predict the size of individual residential buildings. Further, we provide standardized monetary values of houses and contents per square metre of floor space for 30 countries.