This study aimed to combine avalanche statistics with climate change models in orderto assess how a change in precipitation patterns, snow depth and snow density canimpact the avalanche danger in Jämtland, Sweden. Existing climate model reportsfrom SMHI and the Swedish county administration offices were used, and avalanchestatistics were compiled using data from SEPA from 2017 to 2020.It was found that days with moderate avalanche danger are most common (56 %) andthat a lot of days the danger is considerable (33%). The most common avalancheproblem is wind-drifted snow. The results show that wind velocity of 8 m/s isconnected to considerable danger in over 80 % of cases and for 10 m/s even 90 %. Dailyprecipitation of 3 mm or more is also connected to considerable danger on 81% of days;independently of wind. Towards the end of the 21st century, precipitation in Jämtland in winter and spring isexpected to increase by up to 50 % whereas snow depth is likely to decrease so muchthat many places will not reach 100 cm anymore (under the conservative RCP8.5scenario). While the snow depth comes with shortened winter seasons, increasedprecipitation is shown to increase the danger level. It is therefore likely that theavalanche forecasting period will be shortened but intensified in terms of danger.In conclusion, this study confirms again that avalanches are difficult to predict, andthat climate change will not make this easier. This makes it essential to keep updatingthe avalanche information that is available not just in Sweden but across the globe.However, the results are inconclusive due to the shortage of data and due to thecomplex combinations of factors that can impact avalanche danger. Further researchis required. / <p>2021-07-02</p>

Denna forskningsansats undersöker genomförbarheten i att prediktera lavinfara med hjälp av ma-skininlärning i form avXGBoostoch väderdata. Lavinprognoser och meterologisk vädermodelldata harsamlats in för de sex svenska fjällområden där Naturvårdsveket genomlavinprognoser.sepublicerar lavin-prognoser. Lavinprognoserna har hämtats frånlavinprognoser.seoch den vädermodelldata som användsär hämtad från prognosmodellen MESAN, som produceras och tillhandahålls av Sveriges meteorologiskaoch hydrologiska institut. 40 modeller av typenXGBoosthar sedan tränats på denna datamängd, medsyfte att prediktera olika aspekter av en lavinprognos och den övergripande lavinfaran. Resultaten visaratt det möjligt att prediktera den dagligalavinfaranunder säsongen 2018/19 i Södra Jämtlandsfjällenmed en träffsäkerhet på 71% och enmean average errorpå 0,295, genom att applicera maskininlärningpå väderleken för det området. Värdet avXGBoosti sammanhanget har styrkts genom att jämföradessa resultat med resultaten från den enklare metoden logistisk regression, vilken uppvisade en sämreträffsäkerhet på 56% och enmean average errorpå 0,459. Forskningsansatsens bidrag är ett ”proof ofconcept” som visar på genomförbarheten av att med hjälp av maskininlärning och väderdata predikteralavinprognoser. / This research project examines the feasibility of using machine learning to predict avalanche dangerby usingXGBoostand openly available weather data. Avalanche forecasts and meterological modelledweather data have been gathered for the six areas in Sweden where Naturvårdsverket throughlavin-prognoser.seissues avalanche forecasts. The avanlanche forecasts are collected fromlavinprognoser.seand the modelled weather data is collected from theMESANmodel, which is produced and providedby the Swedish Meteorological and Hydrological Institute. 40 machine learning models, in the form ofXGBoost, have been trained on this data set, with the goal of assessing the main aspects of an avalan-che forecast and the overall avalanche danger. The results show it is possible to predict the day to dayavalanche danger for the 2018/19 season inSödra Jämtlandsfjällenwith an accuracy of 71% and a MeanAverage Error of 0.256, by applying machine learning to the weather data for that region. The contribu-tion ofXGBoostin this context, is demonstrated by applying the simpler method ofLogistic Regressionon the data set and comparing the results. Thelogistic regressionperforms worse with an accuracy of56% and a Mean Average Error of 0.459. The contribution of this research is a proof of concept, showingfeasibility in predicting avalanche danger in Sweden, with the help of machine learning and weather data.