Supplementary material to "Data-driven automated predictions of the avalanche danger level for dry-snow conditions in Switzerland"

2021 ◽  
Author(s):  
Cristina Pérez-Guillén ◽  
Frank Techel ◽  
Martin Hendrick ◽  
Michele Volpi ◽  
Alec van Herwijnen ◽  
...  
Keyword(s):  
Data Driven ◽  
Supplementary Material ◽  
Snow Conditions ◽  
Danger Level ◽  
Avalanche Danger

scholarly journals Data-driven automated predictions of the avalanche danger level for dry-snow conditions in Switzerland

2021 ◽  
Author(s):  
Cristina Pérez-Guillén ◽  
Frank Techel ◽  
Martin Hendrick ◽  
Michele Volpi ◽  
Alec van Herwijnen ◽  
...  
Keyword(s):  
Decision Support ◽  
Snow Cover ◽  
Heterogeneous Data ◽  
Swiss Alps ◽  
Data Driven ◽  
Weather Data ◽  
Data Set ◽  
Snow Conditions ◽  
Danger Level ◽  
Avalanche Danger

Abstract. Even today, the assessment of avalanche danger is by large a subjective, yet data-based decision-making process. Human experts analyze heterogeneous data volumes, diverse in scale, and conclude on the avalanche scenario based on their experience. Nowadays, modern machine learning methods and the rise in computing power in combination with physical snow cover modelling open up new possibilities for developing decision support tools for operational avalanche forecasting. Therefore, we developed a fully data-driven approach to predict the regional avalanche danger level, the key component in public avalanche forecasts, for dry-snow conditions in the Swiss Alps. Using a large data set of more than 20 years of meteorological data measured by a network of automated weather stations, which are located at the elevation of potential avalanche starting zones, and snow cover simulations driven with these input weather data, we trained two random forest (RF) classifiers. The first classifier (RF #1) was trained relying on the forecast danger levels published in the avalanche bulletin. Given the uncertainty related to a forecast danger level as a target variable, we trained a second classifier (RF #2), relying on a quality-controlled subset of danger level labels. We optimized the RF classifiers by selecting the best set of input features combining meteorological variables and features extracted from the simulated profiles. The accuracy of the danger level predictions ranged between 74 % and 76 % for RF #1, and between 72 % and 78 % for RF #2, with both models achieving better performance than previously developed methods. To assess the accuracy of the forecast, and thus the quality of our labels, we relied on nowcast assessments of avalanche danger by well-trained observers. The performance of both models was similar to the accuracy of the current experience-based Swiss avalanche forecasts (which is estimated to 76 %). The models performed consistently well throughout the Swiss Alps, thus in different climatic regions, albeit with some regional differences. A prototype model with the RF classifiers was already tested in a semi-operational setting by the Swiss avalanche warning service during the winter 2020-2021. The promising results suggest that the model may well have potential to become a valuable, supplementary decision support tool for avalanche forecasters when assessing avalanche hazard.

scholarly journals How is avalanche danger described in textual descriptions in avalanche forecasts in Switzerland? Consistency between forecasters and avalanche danger

2021 ◽  
Vol 21 (12) ◽  
pp. 3879-3897
Author(s):  
Veronika Hutter ◽  
Frank Techel ◽  
Ross S. Purves
Keyword(s):  
Data Driven ◽  
Extreme Conditions ◽  
Avalanche Size ◽  
Future Studies ◽  
The Public ◽  
Starting Point ◽  
Efficient Communication ◽  
Level 2 ◽  
Danger Level ◽  
Avalanche Danger

Abstract. Effective and efficient communication of expected avalanche conditions and danger to the public is of great importance, especially where the primary audience of forecasts are recreational, non-expert users. In Europe, avalanche danger is communicated using a pyramid, starting with ordinal levels of avalanche danger and progressing through avalanche-prone locations and avalanche problems to a danger description. In many forecast products, information relating to the trigger required to release an avalanche, the frequency or number of potential triggering spots, and the expected avalanche size is described exclusively in a textual danger description. These danger descriptions are, however, the least standardized part of avalanche forecasts. Taking the perspective of the avalanche forecaster and focusing particularly on terms describing these three characterizing elements of avalanche danger, we investigate first which meaning forecasters assign to the text characterizing these elements and second how these descriptions relate to the forecast danger level. We analyzed almost 6000 danger descriptions in avalanche forecasts published in Switzerland and written using a structured catalogue of phrases with a limited number of words. Words and phrases representing information describing these three elements were labeled and assigned to ordinal classes by Swiss avalanche forecasters. These classes were then related to avalanche danger. Forecasters were relatively consistent in assigning labels to words and phrases with Cohen's kappa values ranging from 0.64 to 0.87. Avalanche danger levels were also described consistently using words and phrases, with for example avalanche size classes increasing monotonically with avalanche danger. However, especially for danger level 2 (moderate), information about key elements of avalanche danger, for instance the frequency or number of potential triggering spots, was often missing in danger descriptions. In general, the analysis of the danger descriptions showed that extreme conditions are described in more detail than intermediate values, highlighting the difficulty of communicating conditions that are neither rare nor frequent or neither small nor large. Our results provide data-driven insights that could be used to refine the ways in which avalanche danger could be communicated. Furthermore, through the perspective of the semiotic triangle, relating a referent (the avalanche situation) through thought (the processing process) to symbols (the textual danger description), we provide an alternative starting point for future studies of avalanche forecast consistency and communication.

scholarly journals On the importance of snowpack stability, its frequency distribution, and avalanche size in assessing the avalanche danger level: a data-driven approach

2020 ◽  
Author(s):  
Frank Techel ◽  
Karsten Müller ◽  
Jürg Schweizer
Keyword(s):  
Frequency Distribution ◽  
Large Data ◽  
Relevant Information ◽  
Data Driven ◽  
Data Set ◽  
Avalanche Size ◽  
Quantitative Evidence ◽  
Avalanche Hazard ◽  
Danger Level ◽  
Avalanche Danger

Abstract. Consistency in assigning an avalanche danger level when forecasting or locally assessing avalanche hazard is essential, but challenging to achieve, as relevant information is often scarce and must be interpreted in the light of uncertainties. Furthermore, the definitions of the danger levels, an ordinal variable, are vague and leave room for interpretation. Decision tools, developed to assist in assigning a danger level, are primarily experience-based due to a lack of data. Here, we address this lack of quantitative evidence by exploring a large data set of stability tests (N = 10,125) and avalanche observations (N = 39,017) from two countries related to the three key factors that characterize avalanche danger: snowpack stability, its frequency distribution and avalanche size. We show that the frequency of the most unstable locations increases with increasing danger level. However, a similarly clear relation between avalanche size and danger level was not found. Only for the higher danger levels the size of the largest avalanche per day and warning region increased. Furthermore, we derive stability distributions typical for the danger levels 1-Low to 4-High using four stability classes (very poor, poor, fair and good), and define frequency classes (none or nearly none, a few, several and many) describing the frequency of the most unstable locations. Combining snowpack stability, its frequency and avalanche size in a simulation experiment, typical descriptions for the four danger levels are obtained. Finally, using the simulated snowpack distributions together with the largest avalanche size in a step-wise approach, as proposed in the Conceptual Model of Avalanche Hazard, we present an example for a data-driven look-up table for avalanche danger assessment. Our findings may aid in refining the definitions of the avalanche danger scale and in fostering its consistent usage.

Data-driven automatic predictions of avalanche danger in Switzerland

2021 ◽  
Author(s):  
Cristina Pérez-Guillén ◽  
Martin Hendrick ◽  
Frank Techel ◽  
Alec van Herwijnen ◽  
Michele Volpi ◽  
...  
Keyword(s):  
Real Time ◽  
Winter Season ◽  
Data Driven ◽  
Target Variable ◽  
Test Set ◽  
Avalanche Forecasting ◽  
The Public ◽  
Data Driven Approach ◽  
Danger Level ◽  
Avalanche Danger

<p>Avalanche forecasting implies predicting current and future snow instability in time and space. In Switzerland, avalanche bulletins are issued daily during the winter season to warn the public about the avalanche hazard, described by region with one of five danger levels. Assessing avalanche danger is by large a data-driven, yet experience-based decision-making process. It involves analysing a multitude of data diverse in scale – time and space, and concluding by expert judgment on the avalanche scenario. Numerous statistical models were developed in the past, but rarely applied due to limited usefulness in operational forecasting. Modern machine learning techniques open up new possibilities for developing support tools for operational avalanche forecasting. With this aim, we developed a data-driven approach based on the supervised Random Forest (RF) classifier to automatically predict the danger level for dry-snow avalanche conditions in the Swiss Alps. A large database of more than 20 years of meteorological data and modelled snow stratigraphy data obtained with the numerical snow cover model SNOWPACK were used to train the RF algorithm. We optimized the model and selected the best set of input features that combine meteorological variables and features extracted from the simulated profiles, resampled at the same daily resolution as the forecasts. Our target variable was the regional danger level forecast in the public bulletin. We evaluated the predictive performance of the RF model with an independent test set with data of two winter seasons (2018-2019 and 2019-2020). The test set accuracy was 72 %, which is slightly lower than the accuracy estimate of the public forecasts (about 76 %). Given this uncertainty in our target variable, we trained an optimized RF model on a subset containing so-called verified avalanche danger levels. The test set accuracy then increased to 80 %. During the winter season 2020-2021, both RF models were tested in operational setting and automatically predicted a ‘nowcast’ and a ‘forecast’ in real-time.  In parallel, we also tested a deep recurrent neural network model, which used a 7-days time series with 3-hours time resolution as input and also predicted the avalanche danger level. We present a comparison of the performance of the three models. This is one of the first times that a data-driven approach is tested in real-time as a feasible tool for operational avalanche forecasting.</p>

scholarly journals On the importance of snowpack stability, the frequency distribution of snowpack stability, and avalanche size in assessing the avalanche danger level

2020 ◽  
Vol 14 (10) ◽  
pp. 3503-3521
Author(s):  
Frank Techel ◽  
Karsten Müller ◽  
Jürg Schweizer
Keyword(s):  
Frequency Distribution ◽  
Large Data ◽  
Relevant Information ◽  
Key Factors ◽  
Data Set ◽  
Avalanche Size ◽  
Quantitative Evidence ◽  
Decision Tools ◽  
Danger Level ◽  
Avalanche Danger

Abstract. Consistency in assigning an avalanche danger level when forecasting or locally assessing avalanche hazard is essential but challenging to achieve, as relevant information is often scarce and must be interpreted in light of uncertainties. Furthermore, the definitions of the danger levels, an ordinal variable, are vague and leave room for interpretation. Decision tools developed to assist in assigning a danger level are primarily experience-based due to a lack of data. Here, we address this lack of quantitative evidence by exploring a large data set of stability tests (N=9310) and avalanche observations (N=39 017) from two countries related to the three key factors that characterize avalanche danger: snowpack stability, the frequency distribution of snowpack stability, and avalanche size. We show that the frequency of the most unstable locations increases with increasing danger level. However, a similarly clear relation between avalanche size and danger level was not found. Only for the higher danger levels did the size of the largest avalanche per day and warning region increase. Furthermore, we derive stability distributions typical for the danger levels 1-Low to 4-High using four stability classes (very poor, poor, fair, and good) and define frequency classes describing the frequency of the most unstable locations (none or nearly none, a few, several, and many). Combining snowpack stability, the frequency of stability classes and avalanche size in a simulation experiment, typical descriptions for the four danger levels are obtained. Finally, using the simulated stability distributions together with the largest avalanche size in a stepwise approach, we present a data-driven look-up table for avalanche danger assessment. Our findings may aid in refining the definitions of the avalanche danger scale and in fostering its consistent usage.

scholarly journals How to create an interactive dashboard using R: the example of the Queensland COVID-19 tracker

2020 ◽  
Vol 4 (2) ◽  
pp. 39-47
Author(s):  
Julia Loginova ◽  
Pia Wohland
Keyword(s):  
Low Cost ◽  
Data Driven ◽  
Web Page ◽  
Software Environment ◽  
Online Data ◽  
Interactive Tools ◽  
Queensland Health ◽  
Dissemination Of Research ◽  
Supplementary Material ◽  
Interactive Data

Background  Interactive tools like data dashboards enable users both to view and interact with data. In today’s data-driven environment it is a priority for researchers and practitioners alike to be able to develop interactive data visualisation tools easily and where possible at a low cost. Aims  Here, we provide a guide on how to develop and create an interactive online data dashboard in R, using the COVID-19 tracker for Health and Hospital Regions in Queensland, Australia as an example. We detail a series of steps and explain choices made to design, develop, and easily maintain the dashboard and publish it online. Data and methods  The dashboard visualises publicly available data from the Queensland Health web page. We used the programming language R and its free software environment. The dashboard webpage is hosted publicly on GitHub Pages updated via GitHub Desktop. Results  Our interactive dashboard is available at https://qcpr.github.io/. Conclusions  Interactive dashboards have many applications such as dissemination of research and other data. This guide and the supplementary material can be adjusted to develop a new dashboard for a different set of data and needs.

scholarly journals Supplementary material to "Data-driven distinction between convective, frontal and mixed extreme rainfall events in radar data"

2020 ◽  
Author(s):  
Emma Dybro Thomassen ◽  
Hjalte Jomo Danielsen Sørup ◽  
Marc Scheibel ◽  
Thomas Einfalt ◽  
Karsten Arnbjerg-Nielsen
Keyword(s):  
Extreme Rainfall ◽  
Radar Data ◽  
Data Driven ◽  
Extreme Rainfall Events ◽  
Rainfall Events ◽  
Supplementary Material

scholarly journals A field method for avalanche danger-level verification

1998 ◽  
Vol 26 ◽  
pp. 343-346 ◽  
Author(s):  
A. Cagnati ◽  
M. Valt ◽  
G. Soratroi ◽  
J. Gavaldà ◽  
C. G. SelléS
Keyword(s):  
Regional Scale ◽  
Weather Conditions ◽  
Field Method ◽  
Actual Condition ◽  
Verification Method ◽  
Verification Methods ◽  
Activity Survey ◽  
Further Development ◽  
Danger Level ◽  
Avalanche Danger

Even though the danger-level verification indicated in a bulletin should be a priority aim of avalanche-forecast services, there are no easily applicable verification methods available today. The main difficulty lies in the fact that avalanche observation is no longer sufficient. Therefore, it is necessary to verify the actual condition of the snow-pack stability, particularly concerning low danger levels. This work introduces a procedure for “a posteriori” field verification of danger level, both in space and time (24–72 hours). The method is based on the following elements: avalanche-activity survey, observation of cross-country skiers’ activity, snow profiles and “Rutschblock” tests. These elements, relating both to time and the examination zone, are combined to provide an objective danger degree according to the European avalanche-danger scale. The method was used experimentally in the winter of 1993–94 in the Dolomites and subsequently, in the winter of 1995–96 in the Catalan Pyrenees. As far as 24 hour forecasts are concerned, the method has shown a forecast reliability of 93% in the Dolomites and 76% in the Catalan Pyrenees, while 48 hour forecasts have given values of 89% and 64%, respectively. The lower degree of forecast reliability in the Catalan Pyrenees is accounted for by the unusual weather conditions of winter 1995–96, which was very snowy and characterized by few foreseeable avalanche conditions. The practical application of the proposed verification method has given encouraging results, thus allowing experts to find the main errors in order to improve future forecasts. However, simpler survey procedures are necessary in order to operate on a regional scale. The method is suitable for further development relating to verification of both degree of danger and danger localization.

Sign in / Sign up

Export Citation Format

Share Document