Snow water equivalents exclusively from snow depths  and their temporal changes: the Δ<span style="" class="text smallcaps">snow</span> model

Winkler, Michael; Schellander, Harald; Gruber, Stefanie

doi:https://doi.org/10.5194/hess-25-1165-2021

Articles | Volume 25, issue 3

https://doi.org/10.5194/hess-25-1165-2021

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

https://doi.org/10.5194/hess-25-1165-2021

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

Articles | Volume 25, issue 3

Research article

|

05 Mar 2021

Research article |

| 05 Mar 2021

Snow water equivalents exclusively from snow depths and their temporal changes: the Δsnow model

Michael Winkler, Harald Schellander, and Stefanie Gruber

Download

Final revised paper (published on 05 Mar 2021)
Preprint (discussion started on 17 Apr 2020)

Interactive discussion

Status: closed

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

- Printer-friendly version

- Supplement

RC1: 'Referee Comment on hess-2020-152: Snow Water Equivalents exclusively from Snow Heights and their temporal Changes: The ΔSNOW.MODEL', Anonymous Referee #1, 28 Apr 2020
- AC1: 'Answer to Anonymous Referee #1's comment from 28 April', Michael Winkler, 24 May 2020
SC1: 'Nice work, but some statistical concerns', Michael Matiu, 09 May 2020
- AC4: 'Answer to Michael Matiu's comment from 09 May', Michael Winkler, 05 Jun 2020
  - SC2: 'Re:', Michael Matiu, 05 Jun 2020
RC2: 'Snow Water Equivalents exclusively from Snow Heights and their temporal Changes: The ΔSNOW.MODEL', Anonymous Referee #2, 12 May 2020
- AC2: 'Answer to Anonymous Referee #2's comment from 12 May', Michael Winkler, 05 Jun 2020
RC3: 'Winkler et al: deltaSnow.Model', Anonymous Referee #3, 25 May 2020
- AC3: 'Answer to Anonymous Referee #3's comment from 25 May', Michael Winkler, 05 Jun 2020

Peer-review completion

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

ED: Reconsider after major revisions (further review by editor and referees) (23 Jun 2020) by Markus Weiler

AR by Michael Winkler on behalf of the Authors (06 Aug 2020) Author's response Manuscript

ED: Referee Nomination & Report Request started (26 Aug 2020) by Markus Weiler

RR by Anonymous Referee #1 (09 Sep 2020)

Suggestions for revision or reasons for rejection

Dear authors,

the manuscript has profited from the improvements you have realized. However, several important issues still deserve attention, some of which have been addressed in the first round already, but they are not sufficiently solved. These issues should be seriously elaborated on prior to publication of the manuscript as full paper.
Line numbers refer to the annotated version of the improved manuscript (hess-2020-152-author_response-version1.pdf).

(1): it is still not entirely clear what the ultimate aim of your presented work is. Main points: (i) You say you „present a new model, a new method to simulate SWE.“ Usually a snow modeller would expect a "simulation of SWE" to be based on meteorological forcings (whichever, depending on the empirical or more physical type of the model). What you present to the reader effectively is a layered snow density evolution parameterization for ephemeral snow packs, from which - of course - swe can be derived. Maybe this could be expressed more precisely. (ii) There is still no strong argument to resign from using available meteorological observations. Even if your goal is to develop an assimilation scheme that uses nothing more than recorded HS, a comparison of this scheme to a model using at least temperature records (generally available, extendable with humidity for phase detection) would be scientifically comprehensible approach. There should be locations where even precipitation is available, allowing for extending the classical temperature index approach (and, e.g., including rain-on-snow). However, I do agree that this is an option - but a crucial one, as long as you call your model a snow model. (iii) As the paper is now, it seems that the ultimate goal of your numerical experiment was the snow load map, to be developed without meteorological recordings (because they are not available everywhere for the entire period). If this is the case, the paper should better put this fact in the focus and follow a respective structure and argumentation. If your goal is the model development and comparison with its peers, as you present it in the validation/comparison section, you should focus on the latter and remove much of the rest that does not contribute to this goal. Please make the paper as much as possible clear in its intention.

(2) I still strongly recommend to separate any mathematical formulation of empirical/physical relations from the implementation of the model (code, modules etc.). Swapping two different types of elements in the model along the model description is more puzzling than helpful. Better present the equations of the model in the text of the manuscript, and the code (with its structure – „modules" in your particular implementation) as flow process chart in the appendix.

(3) The language still is common speech like in many formulations, e.g. „a bit of dexterity“ (39), "with the help of De Michele (2013)“ (212) or „For the ∆SNOW.MODEL this kind of high error tolerance of ρ0 is a rather feeble argument to use a power law“ (202/203). Also better avoid all formulations where the model, the code or a process „does“ or „decides“ or „ignores“ something; It is the modeller who does or decides or ignores. I generally recommend professional native English language support.

(4) The manuscript should be significantly shortened. There are entire paragraphs with long explanations that could be replaced by just presenting the mathematical formulas with the respective references (e.g., 384-404). And, if you have shaped the focus of the experiments, much of the rest can be omitted. Examples: (i) there is no need to argue the advantages of HS measurements against anything else (chapter 1.1) - long-term observations are always valuable. (ii) 292-299. (iii) 349-351: it seems that what you call „scaling“ is sort of an assimilation of the HS observations into the current model state by means of an replacement/adjustment of respective variables. Why not say this this with few sentences and skip the rest?

Other details I came across (exemplary, several occurences of the respective type in the text):
- 89ff: your division of SWE models in the „thermodynamic“ and „empirical regression“ type of model is unusual (more usual are energy balance vs. temperature-index). Maybe it should clearly be indicated here that you use the expressions for density models?
- entire chapter 1.2: this chapter should lead to the next one, the motivation. However, a clear argumentation (following the Rango/Martinec model description) of what is missing there, or what could/should be improved, is not given. The last sentence seems to loosly tight to what was stated before
- 194: „For non-zero snow depth observations“: how many?
- 220: „constituted by the sum of loads overlying layer i“: something is missing here to make the sentence complete
- 217, 224: avoid „today’s …“ (better „the most current“ or the „actual“)
- 230: „cannot be observed“: why? Did you make an attempt to illustrate density recordings and analyze if and when these reach a maximum value?
- 237: better formulate that the age of the layer is a surrogate for the physical processes and their evolution and effect over time
- 287-290: this certainly depends on the time step of the model
- 325: „in the model world“: what is a „world“ here?
- 335: „a small “stretching” of the snowpack is necessary“: better avoid such formulations: there is no stretching, but a modification of state variables
- 387: „Sub-daily means of new snow densities are lower“: when and why? Please explain
- 398: „7.62237 × 106 Pa s“: the number of decimal places should be reduced to one
- table A2: correct „valibration“
- etc. …

Hide

RR by Anonymous Referee #2 (11 Sep 2020)

ED: Publish subject to revisions (further review by editor and referees) (22 Nov 2020) by Markus Weiler

Dear Authors,
as you can see from the two reviews of your revised version of the paper, they are quite different. One reviewer only provides some minor points to be revised, the other reviewer is much more critical and points out several issues with the paper. Therefore, I had a much closer look into the paper myself. I think the paper has certainly potential and in general, I like the approach of the paper. But I also think as reviewer #1 that the paper still needs many points to be revised before it can be considered for publication. Based on my and the reviewers assessment, i would highlight the following points to be considered in the revision:
1) The language is still difficult - please try to find a native English speaker to improve the language.
2) Length: the paper and in particular the method section is too long. The method section extends over 8 pages - very detailed, but too detailed for a scientific paper. This should not be a report to an agency. Please constrain the method section much more and refrain from writing the section like a code. Describe what the models does, what physical equation you use, but not how these are taken in the code. If you think this is really necessary, a flow chart with the main model parts would be much more helpful.
3) No selling of the model: there is no need to sell the model to the scientific community. Either it is good, provides new ideas and is superior to other models or it is not. So please refrain from sentences like this: "seems to be the first model since long that ..... Not particularly innovative, but remarkably successful...." This is not a good scientific style.
4) I also think that including temperature or other meteorological variables into the model framework may further improve the model. Measurement of precipitation is certainly difficult and hence a model driven by snow depth makes a lot of sense. But how much could the model improve if these observations would be considered. This would be scientifically very interesting. If you can show, that this would not improve the model much (at least showing it at one location), this would certainly strengthen the realist of your study.
I hope you can consider these points so I can reconsider the paper for publication in Hess.
Best regards
Markus Weiler

Hide

AR by Michael Winkler on behalf of the Authors (08 Jan 2021) Author's response Author's tracked changes Manuscript

ED: Publish subject to minor revisions (further review by editor) (16 Jan 2021) by Markus Weiler

AR by Michael Winkler on behalf of the Authors (25 Jan 2021) Author's response Author's tracked changes Manuscript

ED: Publish as is (25 Jan 2021) by Markus Weiler

AR by Michael Winkler on behalf of the Authors (29 Jan 2021)

Short summary

A new method to calculate the mass of snow is provided. It is quite simple but gives surprisingly good results. The new approach only requires regular snow depth observations to simulate respective water mass that is stored in the snow. It is called ΔSNOW model, its code is freely available, and it can be applied in various climates. The method is especially interesting for studies on extremes (e.g., snow loads or flooding) and climate (e.g., precipitation trends).