HR: 13:40h
AN: T13F-01 INVITED     [Abstracts]
TI: Constraints and Simple Models for Arc and Subduction Geotherms
AU: Parmentier, E
T13F-01 AF: Brown, Univ, Providence, RI 02912 United States
AU: * Kelemen, P
EM: peterk@ldeo.columbia.edu
AF: Lamont, LDEO, Palisades, NY 10964 United States
AU: Hacker, B
T13F-01 AF: UC, SB, Santa Barbara, CA 93106 United States
AU: Hirth, G
T13F-01 AF: WHOI, GG, Woods Hole, MA 02543 United States
AB: There are many constraints on geotherms beneath arcs [1]. Metamorphic PT estimates for arc lower crustal rocks yield Moho temperatures > 800 C at 1 GPa, consistent with estimated mantle equilibration for some primary arc magmas at 1300 C and 1.5 GPa, and with regionally high heat flow in the Oregon and NE Japan arcs. However, none of these data indicate whether high T is steady state, or the result of transient heating due to ascending magma. Slow P- and S-wave velocities immediately beneath arc Moho suggest partial melt is common in the shallow mantle. Topography and gravity constraints for arcs indicate that the uppermost 100 km of the mantle wedge beneath arcs is >10x weaker than the surrounding mantle at the same depth. The geophysical observations suggest that high T(P) is regionally extensive, and so probably continues for long periods. Many recent thermal models for subduction zones emphasize the important effect of temperature dependent viscosity, the possible consequences of stress dependent viscosity and 3D flow, and the importance of accurately tracking the subduction interface [2]. However, most models incorporate a rigid upper layer. Omitting this rigid layer and modeling the entire upper plate with a temperature dependent viscosit, yields a variable thickness thermal boundary layer [1,3]. Where the thermal boundary layer is thinnest, the PT constraints in the previous paragraph are satisfied at steady state. The predicted distance of the highest heatflow from the trench is uncertain, however, because the models do not fully treat the shallow geometry of the subducting plate. There are fewer constraints on subduction geotherms. Low heat flow in forearcs precludes substantial shear heating of the top of the subducting plate while it is colder than about 500 C [4]. However, recent work on viscous heating in narrow, pre-existing shear zones suggests that instabilities arising at temperatures > 500 C can heat a region by a few hundred C over hundreds of meters around the shear zone [5]. This could be important along the subduction zone, and within pre-existing shear zones in both footwall and hanging wall, beneath arcs where heat flow data do not constrain the amount of shear heating. High P, blueschist facies rocks, thought to record subduction conditions, generally record T(P) higher than in most thermal models. Many ultra-high pressure (UHP) terranes, subducted to > 2 GPa, record peak T at or above aqueous fluid saturated solidii for metabasalt and metasediment, hotter than in most thermal models. Some terranes record T(P) higher than fluid undersaturated solidii as well. Because exhumed high P and UHP rocks crossed the subduction interface to transfer from footwall to hanging wall, they may have undergone nearly isobaric, conductive heating at peak P. Transfer from footwall to hanging wall limits the utility of high P and UHP samples for tracking Benioff zone conditions, but illustrates possibilities for partial melting and/or diapirism of subducted metasediment and metabasalt [6]. 1. Kelemen et al AGU03 2. Van Keken et al G302, Conder et al GRL02, Conder PEPI 04, Kincaid and Griffiths G304 3. Rowland and Davies GRL99 4. Peacock AGU03 5. Kelemen and Hirth EOS04 6. Kelemen et al TOG03, Gerya and Yuen EPSL03, Gerya et al Geol04
DE: 1031 Subduction zone processes (3060, 3613, 8170, 8413)
SC: Tectonophysics [T]
MN: Fall Meeting 2005