HR: 1330h
AN: S22B-0452    [PDF]
TI: Constraints on the Mechanism of Attenuation and Thermal Structure in Subduction Zones: Results from BEAAR
AU: * Abers, G A
EM: abers@bu.edu
AF: Boston University, Department of Earth Sciences 685 Commonwealth Av., Boston, MA 02215 United States
AU: Stachnik, J C
EM: josh@giseis.alaska.edu
AF: University of Alaska Fairbanks, Geophysical Institute, Fairbanks, AK 99775-7320
AU: Christensen, D H
EM: doug@giseis.alaska.edu
AF: University of Alaska Fairbanks, Geophysical Institute, Fairbanks, AK 99775-7320
AB: In the sub-lithospheric mantle, seismic attenuation is likely controlled by temperature. Global attenuation models, when compared with estimates of upper-mantle temperature, show consistency with laboratory measurements and indicate that $1/Q\mu$ (shear attenuation) could be a viable proxy for temperature. In most models, $Q\mu$ increases with depth in the upper mantle at a rate similar to that inferred from laboratory studies. Thermally-activated relaxation affects $1/Q\mu$ much more than bulk attenuation ($1/Q\kappa$). However, at low temperatures such as present in subducting slabs and in crust, other processes could dominate the apparent absorption of seismic energy, and $1/Q\kappa$ could be significant. Inversion of regional body-wave spectra from the recent Broadband Experiment Across the Alaska Range (BEAAR), a 2.5-year PASSCAL array, provides some insight into these processes. The 10-15 km station spacing of the 17-36 BEAAR stations allows resolution at scales comparable to travel-time tomography to depths of 150 km, and allows attenuation to be measured separately for wedge, overlying crust, and slab. From body wave spectra (0.5 to 20 Hz) of over 2000 upper-mantle paths, we estimate $1/Q$ for P and S waves. These data are then inverted for spatial variations in $1/Q$. Two aspects of the results provide constraints on the mechanism of attenuation: they require frequency dependence, and they constrain the relative significance of bulk attenuation ($Q\kappa$) to shear. In the mantle wedge, $1/Q\kappa$ is negligible, and $1/Q\mu$ varies as $(frequency)^{-\alpha}$ with $\alpha = 0.2$ to $0.6$. The low end of this range is consistent with laboratory estimates for thermally-controlled attenuation in high-temperature peridotites. In lower temperature regions of the slab and overlying crust, $1/Q\kappa \sim 1/Q\mu$ and $\alpha = 0.6$ to $0.7$, indicating that a different physical mechanism operates, perhaps a combination of scattering and thermoelasticity. The mantle wedge values for attenuation ($Q \sim 100$ to $150$ at 1 Hz) then can be used to constrain temperature. The $Q$ values suggest that near-solidus conditions prevail in the wedge, at least at resolution of tens of kilometers. Similar studies in Japan and the Andes indicate slightly higher temperatures beneath the volcanic arc.
DE: 3909 Elasticity and anelasticity
DE: 5144 Wave attenuation
DE: 7203 Body wave propagation
DE: 7218 Lithosphere and upper mantle
DE: 8180 Tomography
SC: Seismology [S]
MN: 2003 Fall Meeting