Get in touch
Just type your contacts
Developing hardware and software platform for avalanche hazard assessment at mountain ski resorts
In the 2019/2020 season, some of the top ski destinations worldwide registered over 200 snow avalanches that claimed the lives of 62 people. According to the long-term statistics collected by EAWS, Europe alone accounts for around 100 fatalities from snowslides every year. Cutting-edge techniques for analyzing and forecasting hazardous meteorological conditions help to prevent avalanches at alpine resorts and within their boundaries, keeping their guests safe and reputation flawless.
In the 2019/2020 season, some of the top ski destinations worldwide registered over 200 snow avalanches that claimed the lives of 62 people. According to the long-term statistics collected by EAWS, Europe alone accounts for around 100 fatalities from snowslides every year. Cutting-edge techniques for analyzing and forecasting hazardous meteorological conditions help to prevent avalanches at alpine resorts and within their boundaries, keeping their guests safe and reputation flawless.
part 1
Background
Fighting a large body of snow sliding down a steep slope at a high speed is impossible, but certain steps can be taken to mitigate the danger. To this end, ski patrol restricts access to dangerous terrain until the natural loss of snow mass occurs or deliberately triggers an avalanche when no one is on the slope.
The job of ski patrol becomes especially strenuous amid heavy snowfall, high wind, and sudden temperature changes. Such extreme weather may result in repeated and significant variations of meteorological characteristics that differ greatly even in normal conditions based on the slope height and exposure. Hence, the more areas are being tracked, the more precise is the avalanche forecast. Also, the more frequent are the data readings, the more efficient are the risk management efforts taken by ski patrollers.
The job of ski patrol becomes especially strenuous amid heavy snowfall, high wind, and sudden temperature changes. Such extreme weather may result in repeated and significant variations of meteorological characteristics that differ greatly even in normal conditions based on the slope height and exposure. Hence, the more areas are being tracked, the more precise is the avalanche forecast. Also, the more frequent are the data readings, the more efficient are the risk management efforts taken by ski patrollers.
Digital means of collecting and analyzing data will become a reliable component of the safety monitoring infrastructure for mountainous areas. Providing precise data for forecasting instead of average values, the system will help ski resorts ensure proactive assessment of hazardous situations and avoid unnecessarily closed pistes.
— Aleksei Zakusov, Chief Business Development Officer at Skyward
Fatal avalanches in season 2019/2020
Italy
35
13 Dead
Austria
8
13 Dead
Switzerland
6
7 Dead
Norway
96
3 Dead
France
46
12 Dead
USA: Colorado
20
6 Dead
Canada
7
8 Dead
part 2
Optimization task
Krasnaya Polyana Resort is one of the most popular ski resorts in Russia built prior to the 2014 Winter Olympics in Sochi. Located in the Western Caucasus, it attracts hundreds of thousands of tourists from all over the world every year.
Avalanche hazard forecasting at Krasnaya Polyana Resort is based on the evaluation of snow and various meteorological parameters. The three main data sources for the avalanche forecast utilized by the resort are:
Avalanche hazard forecasting at Krasnaya Polyana Resort is based on the evaluation of snow and various meteorological parameters. The three main data sources for the avalanche forecast utilized by the resort are:
Weather stations
01
A network of three stations by SensAlpin (Switzerland), located at different altitudes. The more stations are installed within the resort area, the more complete is the data acquired by the ski patrol. If a controversial situation occurs due to the lack of data, they might decide to close off certain slopes just to be on the safe side, which affects the resort’s attendance and earnings.
Visual observation
02
Every day, prior to the opening, the ski patrol goes on a field trip to monitor the slope condition. During the two hours allocated for the procedure, one patroller has enough time to examine the off-piste area within the resort’s territory. A detailed forecast covering every piste of the resort requires all employees on shift.
Snow stability tests
03
If the data collected at the previous stages does not allow for an informed decision, ski patrol performs field tests to evaluate snow stability. Such tests are usually not planned in advance and may take from 15 to 90 minutes to execute. As a result, the resort’s opening time may be postponed, which will lead to queues.
The work process executed by the ski patrol of Krasnaya Polyana Resort
Daily slope examination, execution of field tests
Manual data processing by members of the ski patrol
Decisions about the slopes accessibility based on the collected observations and weather data
Manual data collection
Data analysis
Avalanche forecasting
part 3
New avalanche hazard forecasting technique
To avoid the above limitations, Maxim Pankov, a weather observer at Krasnaya Polyana Resort, came up with an in-house method of predicting avalanche-prone situations.
This is a simpler, faster and more precise system for snow stability testing. Apart from issuing a reliable avalanche forecast, it will allow us to create our own set of patterns to predict less obvious but dangerous snow and weather conditions.
— Maxim Pankov, a weather observer from the avalanche rescue team at Krasnaya Polyana Resort
Since 2016, the resort’s area is equipped with self-made snow poles
That take snow depth measurements, collect additional meteorological parameters, and partially automate the analysis of the accumulated data. Despite the low prime cost, the solution is not suitable for scaling up due to a number of reasons:
!
Data readings are limited to 2 times a day because of the high energy consumption and low independence of operation.
!
!
Occasional shutdowns occur during operation in extreme weather conditions.
Lack of automated data analysis and visualization system.
part 4
Hardware and software platform Snowburst
Taking into account the limitations of the system currently utilized at the resort, Skyward engineers developed Snowburst, a hardware-software platform consisting of a weather station and a software product. Designed to be the central part of an updated avalanche hazard forecasting process, the solution can accumulate and analyze data, build reports, and many more.
Updated work process executed by the ski patrol of Krasnaya Polyana Resort
Daily data collection with the help of the snow poles installed in avalanche-prone areas by the ski patrol. Execution of field tests.
Snowburst automatically analyzes the collected data, builds reports and converts the statistics into graphs and diagrams.
Decisions about the slope accessibility. Communicating the avalanche forecast insights to the guests of the resort.
Automated data collection
Data analysis performed by software means
Avalanche forecasting
Characteristics of the weather station:
Low energy consumption
01
A specially designed control unit allows for uninterrupted operation and low energy consumption by the station. The circuit board with two STM microcontrollers functions as a communication channel: obtains the station’s sensor readings and sends them to the collection point.
Increased data collection frequency
02
Due to low energy consumption, the weather station sends out updated data readings every hour. For reference, the same characteristic for the snow pole designed by the ski patrol is once per day, for the weather station by SensAlpin — once every half an hour.
High independence of operation
03
A solar panel combined with a battery pack eliminates the need for maintenance throughout the ski season, reduces the possibility of failures and sudden shutdowns of the system.
Reliable communication channel
04
The station has a reserved communication channel for data transmission via low-power wide-area network protocol LoRa. The system is ideal for mountainous areas as it does not depend on GSM or any other external infrastructure.
Operation in extreme weather conditions
05
The station is equipped with industrial-grade components that guarantee flawless operation at low temperatures. Parts with moving elements such as a wind spinner or an anemometer are replaced by ultrasonic sensors. A heating system is used to prevent icing.
Affordable price
06
The station can be customized in accordance with the needs and goals of every project. It can be designed to have a minimum set of components collecting the relevant data, which reduces the cost by several times compared to the Swiss station.
No installation costs
07
Elbow joints allow for easy assembly and high portability of the station. Its installation in mountainous areas does not require helicopter assistance, which makes the product even more cost-effective.
One of the possible station’s designs
Data collected by the weather station:
Snowpack temperature by layers
Snowpack height
Wind speed and direction
Solar flux level
Air temperature
Air humidity
Analytical system
The software part of Snowburst is a proprietary solution developed by Skyward. The visualization is featured with the use of Grafana, an open-source web application that allows to store the data collected from weather stations as well as calculate and display statistics based on it.
Analytics provided by the system:
-
Minimum and maximum air temperature for the last 3 hours - Wind direction statistics (windrose) for the last 6 hours
- Air temperature variations dynamics
- Delta temperature for the last 3 hours
- Wind-blown transfer for the last 6 hours
- Average wind speed for the last 6 hours
- Snow accumulation during the season
- Snow accumulation during a snowfall
- Hourly snow accumulation
- Snowfall statistics during the season
- Precipitation forecast via API of external sources
Snowburst system interface
The solution is integrated with the alerting feature, allowing snow patrollers to receive work-related notifications via a communication channel of their choice: Telegram, Slack, and others.
The following alerts indicate an increase of avalanche hazard at the resort:
High wind and wind-blown snow
Approaching snowfall
Critical snow accumulation
Temperature change over 5℃ for the last 3 hours
part 5
Customer benefits
Automated avalanche hazard forecasting platform to be implemented at Krasnaya Polyana Resort is a unique work method for ski patrollers in Russia and all over the world. Customized settings and robust design combined with a reasonable price turn Snowburst into an indispensable solution for enlarging the scale of observation in mountainous regions. The solution utilization at ski resorts will have a positive impact on their financial performance, professional reputation and, most importantly, guests' safety.
The system capable of automating part of the avalanche team’s routine tasks and forecasting hazardous situations may trigger a new approach to the analysis of environmental conditions at ski resorts and populated mountainous areas by making the forecast more precise, prompt and widespread.
— Maxim Pankov, a weather observer from the avalanche rescue team at Krasnaya Polyana Resort
Being the central part of an updated avalanche hazard forecasting process, the Snowburst software automates complex analytical processes and converts data into graphs and diagrams. Flexible customization allows users to set a list of the monitored parameters and receive real-time alerts in case of deviations. Skyward’s engineering team will undertake all issues related to technical maintenance, new feature development and technical support.
Low energy consumption
Increased data collection frequency
High independence of operation
Reliable communication channel
Operation in extreme weather conditions
Affordable price, no installation costs
Low energy consumption
Prospects for development
Manual data input feature
If required, the solution may be extended to receive and store manual data inputs from field tests. Having access to both environmental parameters and data about the slope stability in the system will make it easier for the ski patroller to see the big picture. With this feature, the avalanche rescue team will be able to make use of the information transmitted in real-time by authorized specialists in the region like ski tourists, mountain guides, etc. This will allow the ski patrollers to enlarge the scale of observation by collecting additional statistics from the territory adjacent to the ski resort.
Avalanche danger patterns recognition system
The possibility to accumulate historical data throughout the time the resort has been in operation is another advantage of the analytical system. Leveraging Machine Learning algorithms and using the data collected by weather stations, it is possible to build and train a model that will forecast the probability of hazardous avalanche situations in different sectors of the resort. The concept of using avalanche patterns to predict dangerous situations was for the first time proposed by a soviet weather observer Vasiliy Akkuratov and, a half-century later, modern technology can help put his idea to practice.
About Skyward
The company was founded in 2011 by a group of enthusiasts who wanted to channel their skills and energy into cre- ating something of value. Now we are a recognized company in the field of telecom and digital technologies. Being relatively small in size, we possess the required agility and willingness to constantly improve for the benefit of our clients. And yet, our expertise combines a broad range of technical skills with an exceptional managerial talent, which is reflected in our motto: driven by people and technology.
Want to inquire about prices or schedule a presentation?
Drop us a line. We will get back to you shortly