Viscosity of the upper mantle from laboratory creep and anelasticity measurements in peridotite at high pressure and temperature

• Published in: Geophysical Journal International
• Main Author: Sato, Hiroki
• Format: Text
• Language: English
• Published: Oxford University Press 1991
• Subjects: Articles; Iceland Plateau; Pacific; Iceland
• Online Access: http://gji.oxfordjournals.org/cgi/content/short/105/3/587; https://doi.org/10.1111/j.1365-246X.1991.tb00798.x

Recent laboratory results on anelasticity and viscoelasticity of mantle rocks at high pressure and temperature show the homologous temperature dependence of seismic Q and creep rate, i.e., both depend on the mantle solidus. These laboratory results enable us to estimate viscosity of the mantle from seismic anelasticity structure. We present the relation between viscosity and Q , and estimate the rheology of the mantle. Derived viscosities are in the range of 3 × 1018 to 3 × 1022 Pa s in the upper mantle beneath the Pacific Ocean. In the Iceland Plateau, the low viscosities of 1016 Pa s determined in 0–5 Myr asthenosphere and 1018 Pa s in 5–10 Myr are consistent with partial melting of 3–7 per cent volume and a high subsolidus temperature in those regions, respectively. These viscosity values also agree well with those determined from observations of Earth's deformation. Seismic Q structure depends on depth and plate age; therefore, both vertical and lateral viscosity variations exist. The difference in plate age between ∼40 and ∼125 Myr beneath the Pacific Ocean yields about one order of magnitude difference in viscosity values. The lateral viscosity gradient is much larger (two orders of magnitude per ∼5 Myr ≃ 100 km) in younger asthenosphere (0–10 Myr) beneath the Iceland Plateau, which is comparable to the vertical viscosity gradient.