Annotation. Avalanche hazard forecasts are of great importance for ensuring the safety of residents and tourists in mountainous areas. All snow avalanches’ forecasting methods used in the work of the snow avalanche service in Kazakhstan were developed in the 80s of the XX century. They need a significant modernization. The Natural Hazards Laboratory of the Institute of Geography and Water Safety studies the advanced scientific methods for assessing and forecasting avalanche danger. In the winter period from December 2020 to April 2021, an experiment was carried out to assess the avalanche danger level in the vicinity of Almaty city and to issue a weekly avalanche bulletin. The main objective of the project is to adapt the experience of the Swiss Institute for Snow and Avalanches (SLF) for the conditions of Kazakhstan. In the course of the work, the meteorological information received from the meteorological stations of international exchange was analyzed. The avalanche danger level was assessed on an international five-point scale. The expert assessment was carried out taking into account the recommendations of the SLF Institute. In order to facilitate the forecaster's work, an automatic expert system using neural network technologies was developed. The neural network was trained on the long-term data on avalanche conditions in the Kishi Almaty River basin. The European center weather forecasts were used to increase the lead time of avalanche warnings. The results of avalanche danger level’s assessing were published weekly on the website of the Institute of Geography and Water Safety, as well as on the Telegram social network. This method is recommended to be used in the operational work of the Kazakhstan Hydrometeorological Service and the Kazakhstan Mudflow Protection Service.
Read full abstract