Predicting avalanche danger in northern Norway using statistical models

Abstract. Snow avalanches are one of the most impactful natural hazards in mountainous areas. Thus, the assessment and forecasting of avalanche danger are of great importance for the protection of life and property. A changing climate may lead to changes in avalanche danger, although the manifestation is unclear. Since climate change is regionally different, an assessment of potential avalanche-danger changes should be conducted on a regional basis. Here the focus is on avalanche danger in the Troms region in northern Norway, i.e. a region in the Arctic. To estimate the linkage between avalanche danger and weather conditions in this region, we utilise expert assessments of regional avalanche-danger level (ADL), the 3 km Norwegian Reanalysis (NORA3), and snow-cover information from the snow model seNorge. Random forest (RF) models are trained and optimised for a binary case and for a four-level case. The binary-case RF model exhibits a much higher overall accuracy (76 %) than the four-level case RF model (57 %), which is due to the latter model often misclassifying ADL 1 as ADL 2 and ADL 4 as ADL 3. Still, the misclassification difference is seldom larger than one ADL, and the distribution of the frequencies of the different ADLs is reproduced. The most important predictive features are related to new snow and wind accumulated and averaged over several days. The binary-case RF model is used to hindcast avalanche-day frequency (ADF) from 1970 to 2024. In this period, the spring season (March–May) shows a small increase in ADF, whereas the winter season (December–February) exhibits negative trends. Moreover, the ADF is found to be correlated with the Arctic Oscillation (AO) index especially in winter, although this correlation appears to have deteriorated in recent years. Given recent advances in skill of representing the AO in decadal prediction systems, this is an encouraging result for the predictability of future avalanche-danger tendencies in northern Norway.

Authors

Institutions

Publication Details

Journal
˜The œcryosphere
Published
2025-05-14
DOI
https://doi.org/10.5194/tc-19-1849-2025
Primary Topic
Landslides and related hazards
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Predicting avalanche danger in northern Norway using statistical models

Kai‐Uwe Eiselt, Rune Grand Graversen
˜The œcryosphere
Landslides and related hazards
article

Predicting avalanche danger in northern Norway using statistical models

Kai‐Uwe Eiselt, Rune Grand Graversen
article en

Abstract

Abstract. Snow avalanches are one of the most impactful natural hazards in mountainous areas. Thus, the assessment and forecasting of avalanche danger are of great importance for the protection of life and property. A changing climate may lead to changes in avalanche danger, although the manifestation is unclear. Since climate change is regionally different, an assessment of potential avalanche-danger changes should be conducted on a regional basis. Here the focus is on avalanche danger in the Troms region in northern Norway, i.e. a region in the Arctic. To estimate the linkage between avalanche danger and weather conditions in this region, we utilise expert assessments of regional avalanche-danger level (ADL), the 3 km Norwegian Reanalysis (NORA3), and snow-cover information from the snow model seNorge. Random forest (RF) models are trained and optimised for a binary case and for a four-level case. The binary-case RF model exhibits a much higher overall accuracy (76 %) than the four-level case RF model (57 %), which is due to the latter model often misclassifying ADL 1 as ADL 2 and ADL 4 as ADL 3. Still, the misclassification difference is seldom larger than one ADL, and the distribution of the frequencies of the different ADLs is reproduced. The most important predictive features are related to new snow and wind accumulated and averaged over several days. The binary-case RF model is used to hindcast avalanche-day frequency (ADF) from 1970 to 2024. In this period, the spring season (March–May) shows a small increase in ADF, whereas the winter season (December–February) exhibits negative trends. Moreover, the ADF is found to be correlated with the Arctic Oscillation (AO) index especially in winter, although this correlation appears to have deteriorated in recent years. Given recent advances in skill of representing the AO in decadal prediction systems, this is an encouraging result for the predictability of future avalanche-danger tendencies in northern Norway.

˜The œcryosphereVol. 19(5)
Norwegian Meteorological Institute (NO), Norwegian University of Science and Technology (NO)
Universitetet i Tromsø, Strong, Horizon 2020
Openalex Percentile: Top 85%
Landslides and related hazards
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.