Ice and Snow Mechanics Day

9.00 Welcome 

9.15 Introduction of Frameglow project, Nicola M. Pugno, with Federico Bosia (Politecnico di Torino) 

9.30 Gianmarco Vallero, Development and FE implementation of a visco-plastic constitutive model for snow

Snow exhibits complex mechanical behaviour arising from its porous structure, temperature sensitivity, and rate-dependent deformation. This contribution presents a novel elasto-visco-plastic constitutive model for snow, formulated within the framework of continuum mechanics. The model introduces new yield and potential functions, and incorporates key physical mechanisms such as viscosity, sintering, and mechanical degradation through time- and deformation-dependent internal variables. It is numerically implemented in the Abaqus/Standard finite element code via a fully implicit backward Euler integration scheme and Powell’s hybrid method for system linearization. The model demonstrates strong predictive capability across various experimental conditions, accurately reproducing confined and unconfined compression, creep, and relaxation tests. In addition, it effectively captures critical deformation patterns in snow, notably the initiation and propagation of compaction bands, which are localised compressive regions governed by the material’s strain-rate sensitivity.

10.30 Edoardo Raparelli, Applying artificial neural networks to satellite and in situ data to produce sub-kilometre meteorological forcing for snow cover properties simulation

The seasonal snowpack in mountainous terrain represents a critical water resource and serves as a key indicator of climate change. The ability to accurately model its evolution in complex terrain is challenging but also fundamental for hydrological applications and natural hazard assessment. Indeed, standard snowpack simulations, driven by coarse[1]resolution meteorological model outputs, often fail to capture the high spatial variability of snow cover properties at the local and slope scales. This study presents an integrated framework to address this limitation. Initially, a modelling chain has been developed by coupling a regional numerical weather prediction model with a land surface snow cover model to simulate key snowpack physical properties. To improve the accuracy of this approach, we explore the use of Artificial Intelligence to downscale and interpolate the fundamental near-surface meteorological variables. This method enhances the resolution of the meteorological forcing for the snowpack model, enabling more realistic simulations of snow[1]atmosphere interactions at a sub-kilometre scale. The entire framework is calibrated and rigorously validated using a comprehensive, long-term dataset of in-situ observations, which includes both automatic weather station data and manual snow pit measurements. The integration of numerical modelling, AI-based data enhancement, and robust ground-truth validation provides a powerful and replicable methodology for high-resolution snowpack estimation. This work demonstrates that such an integrated approach can significantly improve the reliability of snowpack characterisation in complex mountain terrain, with important implications for water resource management and risk mitigation. 

11.30 Bastian Bergfeld, Fracture Mechanics of Snow

Early approaches to assess snowpack instability relied on shear strength–stress ratios, but these indices failed to predict avalanches reliably because they neglected crack propagation within weak layers. Pioneering work by McClung in the late 1970s laid the foundation for applying fracture mechanics to avalanche release. This has fundamentally reshaped our understanding of avalanche release by demonstrating that locally initiated cracks can propagate over large distances even when average stress remains below weak-layer strength. A further milestone was the introduction of the Propagation Saw Test, which enabled direct field observation of the onset and propagation of cracks, including weak-layer collapse and unexpectedly low initial crack speeds (~20 m s⁻¹). Today, avalanche release is widely recognised as a fracture process, with different modes—shear cracks, mixed-mode anticracks, and supershear cracks—playing a central role. This seminar reviews the development of fracture mechanical concepts in snow science over the past twenty years, with particular emphasis on experimental evidence, field and laboratory test methods, and their implications for estimating mechanical properties. I will also present insights from my recent research, focusing on the quantification of fracture properties and the characterisation of fracture modes in snow and avalanches.

12.30 Igor Chiambretti, AINEVA open data 

The Italian regional/provincial avalanche warning services affiliated with AINEVA collect, through their network of observers, an average of 1,200-1,500 snowpack profiles per season in the Italian Alps and the Marche Apennines, which are associated with stability tests (ECT, RB, PST) and other parameters helpful in defining the instability/stability of the snowpack. This database can be consulted, including for research purposes, through a dedicated platform with webGIS functionality (currently under development) or via API queries. The services themselves have repeatedly produced research applied to avalanche forecasting and monitoring, analysing the characteristics of critical layers and fracture types for both spontaneous and accident-triggered avalanches, mainly for sports and recreational users. 

13:30-14.30 Lunch (Orostube)

15 Marin Carlo, Impact of Liquid Water Content in Snow on SAR Backscattering 

Accurate monitoring of snowmelt processes is essential for water-resource management in mountainous regions and for forecasting snow-related hazards such as wet-snow avalanches. This seminar will explore the snowpack energy balance and the key phases of melting, i.e., warming, ripening, and output, which govern the snowmelt. Within this framework, particular emphasis will be placed on the liquid water content (LWC) of snow, a fundamental parameter controlling energy exchange, metamorphism, and the electromagnetic properties that influence radar backscattering during melt. Measuring LWC remains a major challenge due to the transient nature of melt dynamics and the strong spatial variability within the snowpack. The seminar will review the principal techniques developed to quantify LWC—from melting and freezing calorimeters to dielectric methods—and will discuss their field applications and relevance for interpreting Synthetic Aperture Radar (SAR) observations of wet snow.

16 Riccardo Parin, Icing dynamics in aeronautical applications: experiments, modelling and ice protection strategies

Atmospheric icing represents one of the most critical environmental hazards for flight safety and system performance in aeronautics and beyond. Ice accretion on rotating components such as UAV propellers induces strong aerodynamic penalties, vibrations and potential loss of control, making this phenomenon a key driver of experimental and numerical research over the recent years. Starting from the development and commissioning of a dedicated icing wind tunnel (IWT), experimental campaigns were carried out at terraXcube to investigate ice accretion on UAV propellers and to test innovative Passive Ice Protection Systems (PIPSs) designed to extend the time of flight (ToF) in adverse weather conditions. The last part of the talk focuses on data modelling, where ice accretion is analysed on simplified geometries using the established numerical tool FENSAP-ICE. 

Online presentations

17 Jacopo Borsotti, online, Bridging the Gap Between Theoretical Models and Operational Avalanche Forecasting

The transition from theoretical to operational snow avalanche forecasting presents several scientific and practical challenges. Although theoretical models, which are usually validated through field experiments and numerical simulations, provide a robust framework for understanding avalanche release mechanisms, their direct application in operational avalanche forecasting is usually limited by model (mathematical and computational) complexity and the difficulty of measuring key input variables. This seminar examines the differences between the approaches of avalanche forecasters and snow scientists, and explores the main issues arising when moving from theory to operational avalanche forecasting. Particular attention will be given to recently developed skier–snowpack stability models designed to bridge this gap. These models, which are currently used operationally by different avalanche centres, integrate techniques from various scientific disciplines to better assess avalanche triggering likelihood and potential avalanche size. The seminar will conclude by presenting some open problems in operational avalanche forecasting, highlighting the importance of interdisciplinary collaborations to develop effective solutions.

18 David McClung, online, Alpine snow properties relevant to dry snow slab avalanche release

Dry snow slab avalanches initiate from propagating shear fractures within a relatively thin weak layer under a thicker cohesive slab. In shear, alpine snow is a pressure-sensitive, dilatant strain-softening material with significant rate and temperature dependence. In avalanche release, the material is highly porous, and it is typically at 95% of the melt temperature on the Kelvin scale. It follows both quasi-brittle and quantised fracture mechanics. In this presentation, I shall outline the important physical properties, both from laboratory and field measurements, relevant to both dry slab avalanche initiation and dynamic fracture mechanics related to avalanche release.

This event is part of a project that has received funding from the European Research Council (ERC) under the European Union’s Horizon 1.1; kick-off meeting of the ERC AdG 2024 FRAMEGLOW, PI Nicola M. Pugno