the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 22 Sep 2023
Automated Avalanche Terrain Exposure Scale (ATES) mapping – Local validation and optimization in Western Canada
Håvard Toft
Pascal Haegeli
Grant Statham
Abstract. The Avalanche Terrain Exposure Scale (ATES) is a system for classifying mountainous terrain based on the degree of exposure to avalanche hazard. The intent of ATES is to improve backcountry recreationist’s ability to make informed risk management decisions by simplifying their terrain analysis. Access to ATES has been largely limited to manually generated maps in high use areas due to the cost and time to generate ATES maps. Automated ATES (AutoATES) is a chain of geospatial models which provides a path towards developing ATES maps on large spatial scales for relatively minimal cost compared to manual maps. This research validates and localizes AutoATES using two ATES benchmark maps which are based on independent ATES maps from three field experts. We compare the performance of AutoATES in two study areas with unique snow climate and terrain characteristics; Connaught Creek in Glacier National Park, British Columbia, Canada and Bow Summit in Banff National Park, Alberta, Canada. Our results show that AutoATES aligns with the ATES benchmark maps in 74.5 % of the Connaught Creek study area and 84.4 % of the Bow Summit study area. This is comparable to independently developed manual ATES maps which on average align with the ATES benchmark maps in 76.1 % of Connaught Creek and 84.8 % of Bow Summit. We also compare a variety of DEM types (LiDAR, stereo photogrammetry, Canadian National Topographic Database) and resolutions (5 m–26 m) in Connaught Creek to investigate how input data type affects AutoATES performance. Overall, we find that DEM resolution and type are not strong indicators of accuracy for AutoATES, with map accuracy of 74.5 % ± 1 % for all DEMs. This research demonstrates the efficacy of AutoATES compared to expert manual ATES mapping methods and provides a platform for large-scale development of ATES maps to assist backcountry recreationists in making more informed avalanche risk management decisions.
- Preprint
(3284 KB) - Metadata XML
- BibTeX
- EndNote
John Sykes et al.
Status: final response (author comments only)
-
RC1: 'Comment on nhess-2023-112', Zachary Miller, 01 Nov 2023
The manuscript titled “Automated Avalanche Terrain Exposure Scale (ATES) Local validation and optimization in Western Canada” explores a relevant comparison of automated and human derived avalanche hazard mapping. It leverages the updated AutoATESv2 processing workflow with specific reverse-engineering techniques to produce maps of two Canadian domains and quantifies the accuracy of those maps against human created maps that leverage expert knowledge of specific terrain contained within the two domains. The work also quantifies the effects of DEM resolution on ATES classification accuracy for the first time. This research is a valuable contribution to the snow avalanche natural hazards research, forecasting, and communications communities.
The primary concern I have with the manuscript is the color palette chosen for mapping not being red-green (Protanopia) colorblind friendly. I realize this color choice is previously established in ATES mapping, but differentiating between the “Simple” and “Extreme” polygon colors (and Figure 11 bars) is not easy and could lead to considerable confusion in the interpretation of this very valuable tool and the manuscript. I appreciate that the authors utilized differentiated line types in their grid search results figures (7 & 8 and appendix). Please consider adjusting this before final publication.Specific technical corrections and comments:
- Line 102 – “Figures 2 & 3” should be “Figures 1 & 2.”
- Line 116 – The authors mention Rogers Pass without any clarification as to where it is in relation to study site (Connaught Creek) or marking on study site maps. Please clarify for readers without prior knowledge.
- Line 119 – What years are the DEM/DSM data from? The fine-resolution surface elevations and vegetation cover may change frequently in such a dynamic environment and be relevant to the outputs of the ATES analysis.
- Line 153-154 – The numbering of basic “processing” doesn’t align with the referenced plot (Figure 3). Update to include “Runout simulation” within step 1 as shown in figure and include “Validation data generation” as step 2, or adapt figure?
- Figure 3 – Great plot but imagery and text boxes/lines are blurry. Perhaps re-produce at higher dpi.
- Line 245 – Please briefly define "F1 score," "precision," and "recall" to clarify their relevance as quality metrics.
- Line 287 & 289 – “Figures 8 & 9” should be “Figures 7 & 8.”
- Line 304 – “Figure 7,8” should be “Figures 7 & 8” to keep consistent with style.
- Figures 7 & 8 – “Optimized value” symbol in legend is a grey vertical line and is a light blue vertical line in the plots – fix to match.
- Line 433 – remove “26m” from description of the ALOS DEM to match other referenced DEM/DSMs in sentence.
I reiterate that with minor updates this paper will be a valuable addition to the avalanche hazard and communication communities.
Citation: https://doi.org/10.5194/nhess-2023-112-RC1 -
RC2: 'Comment on nhess-2023-112', Marc Adams, 17 Nov 2023
Review of NHESS-2023-112 manuscript entitled "Automated Avalanche Terrain Exposure Scale (ATES) mapping - Local validation and optimization in Western Canada" by Sykes et. al.
General comments
In this contribution, an automised procedure is evaluated, which allows classifying mountainous terrain based on the degree to which backcountry recreationists are exposed to avalanche hazard. The Automated Avalanche Terrain Exposure Scale (ATES) goes back to 2006, when first maps were manually drafted by (local) avalanche experts with the aim of providing decision-support to folks traveling in wintry mountainous terrain. Against the backdrop of an increasing demand for ATES-maps from the community and the necessity to be able to map larger areas at lower costs, an automated routine was developed. This routine uses GIS tool chains and open-source avalanche simulation models. In this manuscript, the routine is validated against manually drafted maps and a site-specific sensitivity study of the key parameters for map generation carried out and analysed.
Overall, the manuscript is clearly structured, succinct and very well written - a joy to read! Figures and Tables are well presented and organised, captions are clearly written and comprehensible.
Specific comments
Apart from a few minor typos, consistency issues and glitches in some figures (see line comments below), I was hard-pressed to find any points to critique in the manuscript. One thing that did strike me, was that I would have expected the impact of using different types of DEM (DTM and DSM) and different resolution DEMs to be much higher on the calculated results. Did I understand correctly that for Connaught Creek both DSM (10m, 17m & 26m from different sources) and DTM (LiDAR) were used in the AutoATES routine, i.e. one bare-earth dataset and three datasets physically including trees? I can see how differing grid resolutions would not impact the result much (especially as forest input data remained the same, as pointed out in the manuscript), but the physical presence of the trees in the DSM, I would have thought to impact for example the FlowPy simulations and thus overhead exposure. This point is however more of a general observation on my part rather than a shortcoming of the manuscript.
To better grasp the implications of certain parameters and accuracy scores on the output of the AutoATES routine, it would have been somewhat helpful to have access to the Toft et al. (2023) manuscript – I take it this is still in preparation, as stated in the reference list, or has it become available in the meantime?
(for line comments see attached PDF)
John Sykes et al.
Data sets
Automated Avalanche Terrain Exposure Scale (ATES) mapping - Local validation and optimization in Western Canada Data and Cpde John Sykes, Håvard Toft, Pascal Haegeli https://osf.io/zxjw5/
Model code and software
AutoATES-v2.0 Håvard Toft, John Sykes, Andrew Schauer https://github.com/AutoATES/AutoATES-v2.0
John Sykes et al.
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 292 | 64 | 8 | 364 | 3 | 4 |
- HTML: 292
- PDF: 64
- XML: 8
- Total: 364
- BibTeX: 3
- EndNote: 4
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1