Influence of ice properties on wave propagation characteristics in partially frozen soils and rocks: A temperature-dependent rock-physics model

A better understanding of the temperature effects on the propagation characteristics of elastic waves in frozen soils and rocks is imperative for accurately quantifying their freezing degrees. Although the existing rock-physics models based on the three-phase Biot (TPB) theory adeptly interpret observed velocity variation with temperature (VVT) curves, they often lack a comprehensive understanding of the mechanisms underlying attenuation variation with temperature (AVT) curves. In this study, we first extend the TPB theory to incorporate the temperature-dependent properties of ice, such as changes in volumetric fraction, morphology, and viscoelasticity, by integrating the relevant thermodynamic laws. Model parameters related to ice properties and interactions, such as rigidity, shear moduli, density, and friction, are redefined. Then, using a numerical rock-physics modeling approach, we examine influential factors and modes of the wave VVT and AVT responses. Our results indicate that the P- and S-wave velocities increase with source frequency, consolidation degree, and frame-supporting ice content, whereas they decrease with temperature and pore-floating ice content. The P- and S-wave attenuation factors increase with the frame-supporting ice content and decrease with the consolidation degree. Rising temperatures tend to amplify the peak magnitude of the P-wave attenuation factors and shift the central frequency of the S-wave attenuation factors. Finally, within a temperature-controlled laboratory environment, we conduct ultrasonic wave transmission testing on brine-saturated sediment and rock specimens. The results demonstrate that as the temperature increases from −15°C to −3°C, the P- and S-wave velocities decrease, whereas the P-wave attenuation factors decrease and the S-wave attenuation factors initially rise before declining. Our viscoelastic TPB theory outperforms the existing ones in interpreting the S-wave AVT observations. This temperature-dependent rock-physics model holds promise for interpreting the sonic logging data in the time-lapse monitoring of permafrost, glaciers, and Antarctica.