the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
openAMUNDSEN v 0.8.3: an open source snow-hydrological model for mountain regions
Abstract. openAMUNDSEN (= the open source version of the Alpine MUltiscale Numerical Distributed Simulation ENgine) is a fully distributed model, designed primarily for calculating the seasonal evolution of a snow cover and melt rates in mountain regions. It resolves the mass and energy balance of snow covered surfaces and layers of the snowpack beneath, thereby including the most important processes that are relevant in such regions. The potential model applications are very versatile; typically, it is applied in areas ranging from the point scale to the regional scale (i.e., up to some thousands of square kilometers), using a spatial resolution of 10–1000 m and a temporal resolution of 1–3 h or daily. Temporal horizons may vary between single events and climate change scenarios. The openAMUNDSEN model has been applied for manyfold applications already which are referenced herein. It features a spatial interpolation of meteorological observations, several layers of snow with different density and liquid water content, wind-induced lateral redistribution, snow-canopy interaction, glacier ice response to climate and more. The model can be configured according to each specific application case. A basic consideration for its development was to include a variety of process descriptions of different complexity to set up individual model runs which best match a compromise between physical detail, transferability, simplicity as well as performance for a certain region in the European Alps, typically a (preferably gauged) hydrological catchment. The Python model code and example data are available for the public as open source project (Hanzer et al., 2023).
-
Notice on discussion status
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
-
Preprint
(9404 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(9404 KB) - Metadata XML
- BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2024-193', Anonymous Referee #1, 10 Apr 2024
General comments
The paper, titled “openAMUNDSEN v 0.8.3, an open-source snow-hydrological model for mountain regions”, describes a fully distributed snow-hydrological model optimized for mountain regions. The model provides a detailed simulation of the snow cover's seasonal evolution, including mass and energy balance across the snowpack. It's designed for varied spatial scales (from point scale to thousands of square kilometers) and temporal scales (from single events to climate change scenarios), incorporating features such as spatial interpolation of meteorological data, multi-layer snow simulation, and glacier ice response to climate change. The model's flexibility allows customization for specific applications, backed by a Python codebase and available as an open-source project for public use. The paper is concisely and clearly written, but I believe it requires some revisions. I am convinced that making these revisions will improve the paper. The comments on major revisions are as follows.
- The gridding of meteorological elements through interpolation of weather station data in this paper is conceptually similar to the Micromet model (Liston and Elder, 2006). Therefore, I would like the paper to discuss the advantages of its methodology by comparing it with Micromet. Additionally, please explain the temperature lapse rate with elevation and the elevation dependence of precipitation using equations. I would also like you to describe how these values differ when compared to Micromet.
- The method of determining the Snow Redistribution Factor (SRF) should be explained in figures such as Figure 1 or Figure 3. Furthermore, it seems that calculating SRF requires a fairly detailed DEM, so there should be a discussion on the maximum grid size for which SRF can be calculated. Additionally, expressing how SRF is used in a formula would allow for a better understanding, so I would like you to show the utilization of SRF in an equation.
- I believe the merit of this model lies in the estimation of the spatiotemporal distribution of snow water equivalent (SWE). On the other hand, the validation data consists of snow cover fraction and snow depth, and I think a comparison with SWE is essential to demonstrate the model's validity. I would like you to show the reproducibility of point SWE measurements. By doing so, it would be possible to verify to what extent the model can reproduce the spatiotemporal distribution of SRF and precipitation, so I would like to request additional validation.
- I do not fully understand the meaning of the sentences in lines 355-357, so I would like you to rewrite them more clearly.
Citation: https://doi.org/10.5194/egusphere-2024-193-RC1 - AC1: 'Reply on RC1', Ulrich Strasser, 26 Apr 2024
-
RC2: 'Comment on egusphere-2024-193', Richard L.H. Essery, 14 Apr 2024
AMUNDSEN is a well-known snow model with multiple capabilities and applications, and this description of the open source version is a useful reference. There is little demonstration of model performance, but that is OK in this model description paper, and the model has been extensively evaluated elsewhere. There is no demonstration at all of the method for generating climate scenarios, and I wonder if so much description is warranted when it is not subsequently used in the paper. Otherwise, there are a few places where I would like to see some more detail, and I have noted some minor corrections.
Line 17
“manyfold applications” – many64
How long are the ”longer time horizons”, and why is lateral snow distribution then especially important?86
“v0.9” – the title and text otherwise refer to v0.8.3. (what is it going to take to commit to v.1.0?)Figure 4
It would be useful to see the station locations (the same as Figure 2a?). What is the resolution of the interpolated grid?271
I guess that Liston and Elder (2006) is used for (the vast majority of) catchments that are less well gauged than Rofental. How does this compare with the dynamic lapse rates?272
How are the precipitation thresholds chosen?282
The method for calculating multiple reflections from clouds and slopes is not described. These reflections contribute to measured radiation, so does the model not end up double counting?353
It is not clear what it means that “different length scales” are used in Figure 5; none are specified.Figures 5 and 7 (especially 7d)
It is counterintuitive for the areas with more snow to be darker.
The Figure 7 caption does not mention that the pink blobs are clouds (it is not a big problem, but there were better Sentinel-2 views on several other days in June 2019).422
“orby satellites” – orbital? Or just “satellites” (ones that are not orbital, such as CryoSat-1, are of limited value).630
I can’t tell what the missing early snowfall event is in Figure 9.
There are no metrics given, but I might judge from Figures 7 and 9 that the most sophisticated EB + Multi + SRF configuration has the worst performance in comparison with observations.Citation: https://doi.org/10.5194/egusphere-2024-193-RC2 - AC2: 'Reply on RC2', Ulrich Strasser, 26 Apr 2024
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2024-193', Anonymous Referee #1, 10 Apr 2024
General comments
The paper, titled “openAMUNDSEN v 0.8.3, an open-source snow-hydrological model for mountain regions”, describes a fully distributed snow-hydrological model optimized for mountain regions. The model provides a detailed simulation of the snow cover's seasonal evolution, including mass and energy balance across the snowpack. It's designed for varied spatial scales (from point scale to thousands of square kilometers) and temporal scales (from single events to climate change scenarios), incorporating features such as spatial interpolation of meteorological data, multi-layer snow simulation, and glacier ice response to climate change. The model's flexibility allows customization for specific applications, backed by a Python codebase and available as an open-source project for public use. The paper is concisely and clearly written, but I believe it requires some revisions. I am convinced that making these revisions will improve the paper. The comments on major revisions are as follows.
- The gridding of meteorological elements through interpolation of weather station data in this paper is conceptually similar to the Micromet model (Liston and Elder, 2006). Therefore, I would like the paper to discuss the advantages of its methodology by comparing it with Micromet. Additionally, please explain the temperature lapse rate with elevation and the elevation dependence of precipitation using equations. I would also like you to describe how these values differ when compared to Micromet.
- The method of determining the Snow Redistribution Factor (SRF) should be explained in figures such as Figure 1 or Figure 3. Furthermore, it seems that calculating SRF requires a fairly detailed DEM, so there should be a discussion on the maximum grid size for which SRF can be calculated. Additionally, expressing how SRF is used in a formula would allow for a better understanding, so I would like you to show the utilization of SRF in an equation.
- I believe the merit of this model lies in the estimation of the spatiotemporal distribution of snow water equivalent (SWE). On the other hand, the validation data consists of snow cover fraction and snow depth, and I think a comparison with SWE is essential to demonstrate the model's validity. I would like you to show the reproducibility of point SWE measurements. By doing so, it would be possible to verify to what extent the model can reproduce the spatiotemporal distribution of SRF and precipitation, so I would like to request additional validation.
- I do not fully understand the meaning of the sentences in lines 355-357, so I would like you to rewrite them more clearly.
Citation: https://doi.org/10.5194/egusphere-2024-193-RC1 - AC1: 'Reply on RC1', Ulrich Strasser, 26 Apr 2024
-
RC2: 'Comment on egusphere-2024-193', Richard L.H. Essery, 14 Apr 2024
AMUNDSEN is a well-known snow model with multiple capabilities and applications, and this description of the open source version is a useful reference. There is little demonstration of model performance, but that is OK in this model description paper, and the model has been extensively evaluated elsewhere. There is no demonstration at all of the method for generating climate scenarios, and I wonder if so much description is warranted when it is not subsequently used in the paper. Otherwise, there are a few places where I would like to see some more detail, and I have noted some minor corrections.
Line 17
“manyfold applications” – many64
How long are the ”longer time horizons”, and why is lateral snow distribution then especially important?86
“v0.9” – the title and text otherwise refer to v0.8.3. (what is it going to take to commit to v.1.0?)Figure 4
It would be useful to see the station locations (the same as Figure 2a?). What is the resolution of the interpolated grid?271
I guess that Liston and Elder (2006) is used for (the vast majority of) catchments that are less well gauged than Rofental. How does this compare with the dynamic lapse rates?272
How are the precipitation thresholds chosen?282
The method for calculating multiple reflections from clouds and slopes is not described. These reflections contribute to measured radiation, so does the model not end up double counting?353
It is not clear what it means that “different length scales” are used in Figure 5; none are specified.Figures 5 and 7 (especially 7d)
It is counterintuitive for the areas with more snow to be darker.
The Figure 7 caption does not mention that the pink blobs are clouds (it is not a big problem, but there were better Sentinel-2 views on several other days in June 2019).422
“orby satellites” – orbital? Or just “satellites” (ones that are not orbital, such as CryoSat-1, are of limited value).630
I can’t tell what the missing early snowfall event is in Figure 9.
There are no metrics given, but I might judge from Figures 7 and 9 that the most sophisticated EB + Multi + SRF configuration has the worst performance in comparison with observations.Citation: https://doi.org/10.5194/egusphere-2024-193-RC2 - AC2: 'Reply on RC2', Ulrich Strasser, 26 Apr 2024
Peer review completion
Journal article(s) based on this preprint
Data sets
Operational and experimental snow observation systems in the upper Rofental: data from 2017 to 2023 M. Warscher, T. Marke, E. Rottler, and U. Strasser https://doi.org/10.5194/essd-2024-45
Viewed
HTML | XML | Total | BibTeX | EndNote | |
---|---|---|---|---|---|
340 | 185 | 32 | 557 | 19 | 19 |
- HTML: 340
- PDF: 185
- XML: 32
- Total: 557
- BibTeX: 19
- EndNote: 19
Viewed (geographical distribution)
Country | # | Views | % |
---|
Total: | 0 |
HTML: | 0 |
PDF: | 0 |
XML: | 0 |
- 1
Cited
1 citations as recorded by crossref.
Ulrich Strasser
Michael Warscher
Erwin Rottler
Florian Hanzer
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(9404 KB) - Metadata XML