Avalanche forecasting increasingly relies on numerical predictions of snow stability. We present the first comparison of traditional as well as AI-based stability metrics and validate them against avalanche observations. The metrics show a similar level of performance. Combining both process-based and data-driven metrics provides the most reliable forecasts.

We examined how avalanche forecasters across Europe use the EAWS (European Avalanche Warning Services) Matrix, a decision-support tool for determining regional avalanche danger levels. Although warning services apply the Matrix differently, we identified both consistent patterns and notable inconsistencies in its application. Our findings highlight where the Matrix works well and where clarification is needed, supporting more consistent and transparent avalanche information for the public.

This paper presents the updated EAWS (European Avalanche Warning Services) Matrix, developed to support consistent avalanche danger assessments across Europe. It links snowpack stability, its frequency, and avalanche size to the five danger levels. Based on expert surveys and operational testing, the Matrix supports expert judgment and aligns with the Conceptual Model of Avalanche Hazard while addressing known ambiguities in practice.