HR: 14:45h
AN: T53D-05    [Abstracts]
TI: Syntectonic Melt Distribution in Deep Crust Inferred From Residual Granulites: Implications for the Rheology of Orogenic Crust
AU: * Brown, M
EM: mbrown@geol.umd.edu
AF: University of Maryland, Lab. for Crustal Petrology, Dept. of Geology, College Park, MD 20742
AB: Anatectic systems are heterogeneous, nonlinear and characterized by multiphase flow. Each process contributing to melt extraction has a characteristic length and time scale, and it is the nonlinear interactions and feedback among them that give rise to the patterning observed in lower crust. Melting and melt ascent and emplacement modify the physical properties of the crust and generate viscosity contrasts, which lead to a heterogeneous response and localization. The characteristic length scale of deformation is important, since reducing the length scale for the same velocity of deformation increases the strain rate (e.g. whole crust to orogenic channel to shear zone). Whether patterns of localization are stable with increasing strain and what effect localization has on rheology are poorly understood. Geophysical imaging suggests 6-20 vol.% interconnected melt in the crust of active orogens, numerical modeling of the transition from coupled doubly-vergent wedge structure to plateaux formation requires a viscosity drop of up to 4 orders of magnitude, and deformation experiments on rock undergoing melting indicate a 4-10-fold drop in strength due to wetting of most grain boundaries as melt volume approaches 10%. The magnitude of weakening accompanying melting suggests deformation dominantly by melt-assisted diffusion creep with melt segregation and extraction rather than magma behavior (i.e. bulk rheology of anatectic crust is solid-dominated rather than melt-dominated). However, fertile rocks generate 10-50 vol.% melt (at 1 GPa, 1,173K), which suggests melt is extracted to maintain a solid-dominated rheology. Melting occurs at multiphase grain boundaries around hydrate phases, whereas plutons represent 1,000-10,000 km3 of crystallized magma; this requires focusing the flow of segregated melt to channels that allow ascent through subsolidus crust. Studies of residual granulites suggest that melt has migrated from grain boundaries to networks of (leucosome-filled) structures to ascent conduits (now steeply-inclined rod or tabular granites). However, the common assumption that leucosome vol. equals melt vol. is precluded by the mineralogy and chemistry of leucosomes, which indicates both cumulate (early-crystallized solids) and fractionated (late-crystallized residual liquids) varieties. At the grain scale, location of melt is controlled by fabric and strain. At outcrop in residual granulites leucosome occurs in fabric parallel and transverse stromata, along foliation planes and in dilation/shear bands, forming networks analogous to ideal deformation band networks. Leucosome networks commonly are elongate parallel to lineation, so melt flow is inferred to have been primarily in the plane of the foliation and along the lineation to developing dilatant structures, and through the network of structures to ascent conduits (commonly dikes). Dike emplacement occurs along a preferred direction independent of anisotropy, suggesting stress control, which with the macroscopic fracture-like discontinuities characterizes the process of formation as a fracture phenomenon. Petrographic continuity of leucosome with granite in dikes suggests a once continuous melt-bearing network, and indicates that material in leucosomes and dikes underwent final crystallization at the same time. Blunt dike tips and zigzag propagation paths point to ductile fracture as the mode of formation, probably by pore growth and coalescence of melt pockets. The coupling between fracture formation and mass transfer appears to be significant for fracturing in crustal environments above the solidus, where a large amount of plastic strain may accumulate before fracture and where dislocation-mediated ductile fracture may be expected.
DE: 8102 Continental contractional orogenic belts
DE: 8159 Rheology--crust and lithosphere
DE: 5104 Fracture and flow
DE: 3640 Igneous petrology
DE: 3660 Metamorphic petrology
SC: Tectonophysics [T]
MN: 2004 AGU Fall Meeting