HR: 1330h
AN: S42A-0140 [PDF]
TI: Why is the Cascadia subduction zone backarc hot? Numerical tests of mantle wedge flow
AU: Currie, C A
EM: currie@uvic.ca
AF: School of Earth and Ocean Sciences, University of Victoria, PO Box 3055, Victoria, BC V8W 3P6
Canada
AU: Currie, C A
EM: currie@uvic.ca
AF: Pacific Geoscience Centre, Geological Survey of Canada, PO Box 6000, Sidney, BC V8L 4B2
Canada
AU: * Wang, K
EM: kwang@nrcan.gc.ca
AF: School of Earth and Ocean Sciences, University of Victoria, PO Box 3055, Victoria, BC V8W 3P6
Canada
AU: * Wang, K
EM: kwang@nrcan.gc.ca
AF: Pacific Geoscience Centre, Geological Survey of Canada, PO Box 6000, Sidney, BC V8L 4B2
Canada
AU: Hyndman, R D
EM: rhyndman@nrcan.gc.ca
AF: School of Earth and Ocean Sciences, University of Victoria, PO Box 3055, Victoria, BC V8W 3P6
Canada
AU: Hyndman, R D
EM: rhyndman@nrcan.gc.ca
AF: Pacific Geoscience Centre, Geological Survey of Canada, PO Box 6000, Sidney, BC V8L 4B2
Canada
AU: He, J
EM: jhe@nrcan.gc.ca
AF: Pacific Geoscience Centre, Geological Survey of Canada, PO Box 6000, Sidney, BC V8L 4B2
Canada
AB:
Understanding mantle wedge processes is critical for constraining thermal and petrological controls on in-slab earthquakes
and the behaviour of the deep subduction thrust fault. Observational constraints indicate that the mantle wedge at the
northern Cascadia subduction zone is extremely hot. Below the volcanic arc, temperatures greater than 1300$\deg$C are
required for magma generation. In the backarc, surface heat flow, seismic velocities, thermal isostasy and xenolith studies
suggest temperatures of 1200$\deg$C at 60 km depth for a distance of 500 km. The landward limit of the backarc is the abrupt
contact with the thick, cold North America craton, making high backarc temperatures even more surprising. An initial
compilation of thermal data shows that most other backarcs are similarly hot.
Finite element thermal models are used to investigate the backarc mantle flow structure that maintains these high
temperatures. Two principle driving forces for flow are: traction along the top of the subducting slab and buoyancy forces
due to lateral thermal heterogeneities, such as cooling by the slab and Rayleigh instabilities. In this study, we primarily
deal with traction-driven flow, using Cascadia subduction parameters. A thick ($>$200 km) lithosphere was introduced at the
landward backarc boundary, consistent with the presence of the North America craton root.
For an isoviscous mantle, the craton deflects hot material from depth into the wedge, resulting in a warmer wedge than models
without a craton, although the temperatures are 150-300$\deg$C lower than inferred from observations. Decoupling of the
wedge from the over-riding plate increases the backarc Moho temperature by over 100$\deg$C; temperatures below the arc are
relatively unaffected. With a more realistic stress- and temperature-dependent viscosity, high velocity flow originates from
great depths along the landward boundary, even without a craton. Flow is strongly focussed into the wedge corner, leading
to much higher sub-arc temperatures ($>$1250$\deg$C). However, the flow pattern produces low backarc mantle temperatures and
heat flow, inconsistent with observations. For all models, full coupling between the subducting plate and wedge was
assumed. If partial decoupling occurs, due to the presence of hydrated mantle or serpentine/talc, wedge flow velocities will
be slower, and the wedge will be cooler than predicted by the models.
None of the traction-driven flow models simultaneously produced high temperatures below the volcanic arc and a uniformly hot
backarc. Numerous studies indicate that the mantle wedge viscosity is less than 10$^{19}$ Pa s. Initial tests suggest that
at such low viscosities, wedge flow driven by thermal buoyancy may be more important than traction-driven flow.
DE: 3210 Modeling
DE: 5134 Thermal properties
DE: 8120 Dynamics of lithosphere and mantle--general
SC: Seismology [S]
MN: 2003 Fall Meeting