10:15 〜 10:30
[SCG45-42] Variability of the Velocity-Acceleration Scaling Relationship in Accelerating Creep
キーワード：Landslide, Acceleration, Creep, Friction, Failure-time forecast
Predicting the timing of natural hazards is inherently complex due to the interplay of various physical processes and parameters. The empirical power-law velocity-acceleration scaling relationship, known as the Voight model (Fukuzono, 1985; Voight, 1988), is widely recognized as an effective and reliable tool for forecasting laboratory creep failure and natural events such as landslides and volcanic eruptions.
The exponent in this power-law relationship plays a key role in characterizing precursory accelerating creep behavior of instabilities. Field observations and experiments indicate that this exponent typically ranges from 1 to 2 (e.g., Segalini et al., 2018) but can be significantly lower (e.g., Bozzano et al., 2014) and may evolve over time (e.g., Crosta & Agliardi, 2003; Chang et al., 2024). Rate- and state-dependent friction (RSF) laws, extensively used in fault mechanics, have also been applied to model accelerating landslide creep as a slider block under constant loading. These models predict either a constant α = 2 (Helmstetter et al., 2004; Noda & Chang, 2023) or an evolving α from 1 to 2 (Chang et al., 2024), depending on the frictional properties.
In this presentation, we analyze an extensive dataset of laboratory experiments and field observations to examine the statistical distribution of the exponent. Building on previous theoretical analyses (Noda & Chang, 2023; Chang et al., 2024), we further explore the physical basis for the observed variability through numerical modeling.
The exponent in this power-law relationship plays a key role in characterizing precursory accelerating creep behavior of instabilities. Field observations and experiments indicate that this exponent typically ranges from 1 to 2 (e.g., Segalini et al., 2018) but can be significantly lower (e.g., Bozzano et al., 2014) and may evolve over time (e.g., Crosta & Agliardi, 2003; Chang et al., 2024). Rate- and state-dependent friction (RSF) laws, extensively used in fault mechanics, have also been applied to model accelerating landslide creep as a slider block under constant loading. These models predict either a constant α = 2 (Helmstetter et al., 2004; Noda & Chang, 2023) or an evolving α from 1 to 2 (Chang et al., 2024), depending on the frictional properties.
In this presentation, we analyze an extensive dataset of laboratory experiments and field observations to examine the statistical distribution of the exponent. Building on previous theoretical analyses (Noda & Chang, 2023; Chang et al., 2024), we further explore the physical basis for the observed variability through numerical modeling.