Impact-based approaches are reshaping mountain hazard research by shifting focus from what the weather will be to what the weather will do. This shift is underpinned by the concept of the warning value chain, which integrates observations, forecasts, hazard assessments, and impact analyses to enable effective, risk-oriented early warning.

This session focuses on conceptual and applied contributions that cover one or more aspects along the warning value chain, with emphasis on integrative studies that link relevant risk drivers (e.g., hydrometeorological trigger information, geo-environmental conditions, and socio-economic factors) to support informed decision-making in mountain regions. We consider multiple time scales, from real-time forecasting to trends in losses under climate change.

Topics of interest include:

1. Coupled frameworks that link hydro-meteorological drivers, terrain and land cover, exposure, vulnerability, and runout dynamics to deliver spatially explicit estimates and scenario-based maps.
2. Geomorphic plausibility, interpretability, uncertainty quantification, and reproducibility, including physics-informed and hybrid machine learning approaches.
3. Evaluation of thresholds, predictive skill, and communication of uncertainty, with attention to true/false alarm trade-offs and decision relevance.
4. Use of long-term event records and loss databases to disentangle climate change, exposure dynamics, and mitigation effects.
5. Studies from regional forecasting to process-based local management, emphasizing transferability and generalizability across basins and ranges, and multi-hazard interactions.

The session aims to connect scientists, practitioners, and policymakers to guide adaptation under evolving climate and land-use pressures.