Automatic detection of snow avalanches in continuous seismic data using hidden Markov models

Heck, Matthias; Hammer, Conny; van Herwijnen, Alec; Schweizer, Jürg; Fäh, Donat

doi:https://doi.org/10.5194/nhess-18-383-2018

Articles | Volume 18, issue 1

https://doi.org/10.5194/nhess-18-383-2018

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

https://doi.org/10.5194/nhess-18-383-2018

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

Articles | Volume 18, issue 1

Research article

|

25 Jan 2018

Research article |

| 25 Jan 2018

Automatic detection of snow avalanches in continuous seismic data using hidden Markov models

Matthias Heck, Conny Hammer, Alec van Herwijnen, Jürg Schweizer, and Donat Fäh

Download

Final revised paper (published on 25 Jan 2018)
Preprint (discussion started on 10 Jul 2017)

Interactive discussion

Status: closed

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

- Printer-friendly version

- Supplement

RC1: 'Review of: “Automatic detection of snow avalanches in continuous seismic data using hidden Markov models” by M. Heck et al.', Anonymous Referee #1, 20 Jul 2017
- AC1: 'Response to reviewer #1', Matthias Heck, 27 Sep 2017
RC2: 'RC2', Naomi Vouillamoz, 17 Aug 2017
- AC2: 'Response to reviewer #2', Matthias Heck, 27 Sep 2017

Peer-review completion

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

ED: Publish subject to minor revisions (review by editor) (20 Nov 2017) by Margreth Keiler

Dear Matthias Heck and colleagues,
Thank you for submitting your manuscript ‘Automatic detection of snow avalanches in continuous seismic data using hidden Markov models’ to NHESS.
I read the two reports from the reviewers and your interactive response on their comments. Both reviewers highlighted that the manuscript is sound, the data and approach of hidden Markow models (HMMs) to detect snow avalanches is clearly described. The comments of the reviewers asked for at least two topics more detailed information/discussion on a) model performance and b) the cons of the HMM in regard of pre-detection and post-processing. Additional comments include mainly suggestions to considering additional clarification on some methods aspects and results, to reach a higher readability of the figures, a few more aspects in the discussion. I therefore ask you to consider all the comments of the reviewers and include this in the revision of your manuscript. Moreover, please check all equations in your manuscript and adapt them to the guidelines of NHESS.
Regarding the comments of the reviewers I decided a minor revision. I will decide afterwards if only the editor will review the revised version of the manuscript or also additionally one of the reviewer.

The next steps are to prepare one response file (as .pdf) including (1) comments from referees/public, (2) author's response to each comment one by one, and (3) author's changes in manuscript. Regarding author's changes, a marked-up manuscript version (track changes in Word, latexdiff in LaTeX) converted into a *.pdf including the author's response must be submitted. Additionally, you have to upload the new version of the manuscript (see also please consider therefore the guidelines of the NHESS procedure again (http://www.natural-hazards-and-earth-system-sciences.net/peer_review/interactive_review_process.html#interactivepublicdiscussion).

If you have any question regarding the procedure or interpretation of the comments, please do not hesitate to contact me for clarification.

Regards, Margreth Keiler
[NHESS Editor]
[Associate Professor of Geomorphology, Natural Hazards and Risk Research, University of Bern]

Hide

AR by Matthias Heck on behalf of the Authors (21 Nov 2017) Author's response Manuscript

ED: Publish subject to technical corrections (02 Dec 2017) by Margreth Keiler

AR by Matthias Heck on behalf of the Authors (07 Dec 2017) Manuscript

Short summary

In this study we use hidden Markov models, a machine learning algorithm to automatically identify avalanche events in a continuous seismic data set recorded during the winter 2010. With additional post processing steps, we detected around 70 avalanche events. Although not every detection could be confirmed as an avalanche, we clearly identified the two main avalanche periods of the winter season 2010 in our classification results.