Stress wave propagation through cohesive snow using viscoelastic analysis

Abstract

Many avalanches are triggered by dynamic loads. Previous avalanche-focused studies of snow's response to dynamic loading consisted of field-based observations and static, elastic models. In this dissertation, snow was modeled dynamically as a Maxwell-viscoelastic material using parameters determined from laboratory experiments which resembled the Compression Test (CT) and Extended Column Test (ECT) -- common stability tests used by avalanche practitioners. First, 1D homogenous tests, akin to the CT, determined relationships between observed snow properties and ascertained elastic moduli and viscosities. The snow was loaded with both shorter duration (~1 ms) and longer duration (~10 ms) impacts from a dropped mass. The model was then expanded to 2D and validated using laboratory tests which resembled the ECT. Lastly, layered snow was investigated. The cohesive snow, with densities ranging from 135 to 428 kg m -3 , exhibited elastic moduli between 1 and 100 MPa and viscosities between 5 and 40 kPa s. The shorter duration impacts resulted in higher wave speeds and greater damping coefficients. Furthermore, the laboratory's substantial concrete floor caused reflection and amplification of vertical-normal compressive stress -- a phenomenon both observed and modeled. This reflection had a dominating effect in the layered laboratory studies. The modeling effort was extended to infinite and semi-infinite domains. These simulations revealed that isolating a block of snow, as is done during a CT or ECT, creates a wave guide which leads to deeper transmission and different distribution of shear and normal stresses compared to a 2D half space. How cohesive layers of snow are positioned within the simulated snowpack are modeled to affect the dynamic stress distribution. In a softer-over-harder configuration, both vertical- normal compressive stress and shear stress are modeled to penetrate deeper below the layer interface. Above the interface, the vertical-normal compressive stress is modeled to be greater in a softer-over-harder configuration, while the shear stress is reduced. This result is attributed to how dilatational and distortional waves travel through layered materials. In conclusion, this study enhances our understanding of stress wave propagation through snow by dynamically modeling it as a Maxwell-viscoelastic material and validating the model with laboratory experiments.