3:30 PM - 3:45 PM
[SIT04-02] Experimental study on polycrystal anelasticty with implications for upper mantle seismic structure
Keywords:anelasticity, seismic attenuation, upper mantle
Rock anelasticity causes dispersion and attenuation of seismic waves. Therefore, for the quantitative interpretation of seismic low velocity and/or low Q regions in the upper mantle, understanding of rock anelasticity is necessary. Recent experimental studies have shown that anelasticity of polycrystalline materials is subject to the Maxwell frequency fM scaling: Q-1(f/fM). However, the applicability of this scaling to the seismic waves has not been guaranteed because experimental frequencies normalized to fM of the laboratory samples are usually much lower than the seismic frequencies normalized to fM in the upper mantle (106 ≤ f/fM≤ 109). In this study, by using polycrystalline organic borneol as an analogue to mantle rock, we measured anelasticity up to f/fM G 108 and found that the Maxwell frequency scaling is not fully applicable at f/fM > 104. A closer examination of our data showed that each of the relaxation spectra obtained under various temperature, grain size, and chemical composition can be represented by the superposition of a background dissipation which is subject to the Maxwell frequency scaling and a peak dissipation which is always centered at f/fM = 103. Significant increases of the peak amplitude and width with increasing temperature, grain size, and impurity (dyphenylamine) content result in failure of the Maxwell frequency scaling at f/fM > 104, where the peak dissipation dominates over the background dissipation. To quantitatively estimate the dispersion and attenuation of seismic waves, it is important to understand the behavior of the peak dissipation.
The addition of impurity (diphenylamine) to borneol significantly reduces the melting (solidus) temperature from Tmelt = 477 K to Tmelt = 316 K. Therefore, we have speculated that the observed variation of the peak dissipation with impurity and temperature can be scaled by the normalized temperature T/Tmelt, such that the peak amplitude and width increase with increasing T/Tmelt. The significant broadening of the peak observed near (but below) the solidus temperature (T/Tmelt = 0.93) means that seismic velocity and Q are considerably lowered even without melt and has important implications for upper mantle seismic structure. We further investigated the detailed behavior of the peak dissipation at near solidus temperatures (0.88 ≤ T/Tmelt ≤ 1.01), and found that the peak amplitude saturates at about T/Tmelt = 0.95, but that the peak width continuously increases up to the supersolidus temperature T/Tmelt = 1.01. The obtained result was formulated in terms of the two nondimensional parameters f/fM and T/Tmelt and preliminarily applied to the seismic waves in the upper mantle. The result shows that low V and low Q occur at near solidus temperatures even without melt. At the onset of melting, seismic wave velocity shows a discrete reduction due to the poroelastic effect of melt, but the seismic attenuation does not show a discontinuous change.
The addition of impurity (diphenylamine) to borneol significantly reduces the melting (solidus) temperature from Tmelt = 477 K to Tmelt = 316 K. Therefore, we have speculated that the observed variation of the peak dissipation with impurity and temperature can be scaled by the normalized temperature T/Tmelt, such that the peak amplitude and width increase with increasing T/Tmelt. The significant broadening of the peak observed near (but below) the solidus temperature (T/Tmelt = 0.93) means that seismic velocity and Q are considerably lowered even without melt and has important implications for upper mantle seismic structure. We further investigated the detailed behavior of the peak dissipation at near solidus temperatures (0.88 ≤ T/Tmelt ≤ 1.01), and found that the peak amplitude saturates at about T/Tmelt = 0.95, but that the peak width continuously increases up to the supersolidus temperature T/Tmelt = 1.01. The obtained result was formulated in terms of the two nondimensional parameters f/fM and T/Tmelt and preliminarily applied to the seismic waves in the upper mantle. The result shows that low V and low Q occur at near solidus temperatures even without melt. At the onset of melting, seismic wave velocity shows a discrete reduction due to the poroelastic effect of melt, but the seismic attenuation does not show a discontinuous change.