1 2 3 Supplementary Information for manuscript: Extreme localized exhumation at syntaxes initiated by subduction geometry Rebecca Bendick1,2, Todd A. Ehlers2 4 5 1 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 2 Department of Geosciences, University of Montana, Missoula, MT 59802, USA. Department of Geosciences, University of Tübingen, 72074 Tübingen, Germany. Extended description of methods The numerical simulations used in this study are built with the COMSOL multiphysics finite element solver including the Arbitrary Lagrangian Eulerian (ALE) formulation (Fullsack, 1995) to accommodate large deformation and erosion on the upper boundary of the model domain. The specifics of the model setup follow: ALE All of the model domain boundaries are fixed in space. Mass may be advected through the top boundary only; once mass passes through the z=zmax surface (the Earth’s surface), it disappears from the computation. Therefore, the top temperature and velocity boundary conditions are always applied on z=zmax (100 km), Total mass is conserved such that the sum of the flux out of the upper surface and the outflow over the rigid indenter equals the inflow at x=0 (Fig. S1). Fluid mechanics The velocity field is calculated for a viscous fluid on the model domain using the NavierStokes equation where the inertial terms are assumed to equal zero (laminar flow), ππ π ππ‘ = ∇ β [−π + π(∇π + (∇π)π» )] + π with π∇ β π = 0. In this expression, ο² is the fluid density, u is the velocity vector, ο is the viscosity, and F is the body force tensor. The upper boundary at z=zmax is free, so that fluid can move through z=zmax (100 km) to form topography (or be “eroded” using the ALE formulation). The inflow condition at x=0 is a constant normal velocity of 20 mm/yr. The outflow condition is a free boundary, where again fluid can move out of the model domain. The y=0 and y=ymax (800 km) boundaries have free boundary-parallel slip, and the basal boundary, including the top surface of the indenter, has a fixed velocity boundary condition set at the inflow velocity to approximate net convergence between the indenter and the overriding fluid (Fig. S1). The viscosity of the fluid in the model domain is either constant (in the constant viscosity simulations, Table 1 in manuscript), or determined by the temperature solution for the domain, 1−π 1 −1 π through π = 2 π΄ ⁄π πΜ ⁄π ππ₯π (ππ π), with varying values for n used in the sensitivity testing (Table 1 in text, and Fig. S2). In this expression, ο is the viscosity, A is a conditioning coefficient, n is a viscosity exponent, πΜ is the strain rate, Q is the activation energy, R is the ideal gas constant, and T is the temperature. Temperature The temperature field is calculated for a viscous fluid including both advection and ππ diffusion with the advection-diffusion equation, ππΆπ ππ‘ + ππΆπ π β ∇π = ∇ β (π∇π) + π. In this expression, ο² is the fluid density, πΆπ is the heat capacity, T is the temperature and t Bendick and Ehlers Supplemental Material 1 47 48 49 50 51 52 53 54 55 56 57 58 59 the time, u is the velocity vector, k is the thermal diffusivity, and Q is internal heat production. The initial temperature condition on the model domain is a 1D steady state continental geotherm for the boundary conditions specified below, no heat advection, −π§ and with internal heat production given by π = π0 ππ₯π ( ). This temperature relation is βπ also imposed on the inflow boundary (x=0) for the duration of the model run (e.g. Fig. S3). The outflow condition is free in the temperature problem. The y=0 and y=ymax boundaries are insulated, so the net lateral heat flux equals zero. The basal boundary condition is set at a fixed temperature of 980ºC; the surface boundary condition at a fixed temperature of 30ºC. The temperature on the top of the rigid indenter (Fig. S1) at different depths is given by the solution (Molnar and England, 1990) for the steady state limit on a slipping π discontinuity, π ′ = . In this expression, T’ is the steady state temperature 1+π√(π§π ππ πππΏ)/π 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 on the “fault”, when T is the standard geothermal temperature for the given depth, b is a dimensionless correction factor, π§π is the depth to the discontinuity, V is the slip rate ο€ is the component of the dip of the fault in the thrust direction, and ο« is the thermal diffusivity. The initial mesh and the model conditions are also summarized in Figure S1. The range of parameter values used in the sensitivity testing are given in Table 1 in the text and shown in manuscript Figures 2 and 3a. Thermochronometer Age calculation: Apparent cooling ages are predicted for rocks exhumed to the surface for comparison to observed zircon fission track (ZFT) ages. The temperature history of exhumed rocks is determined using transient particle tracking. A discrete set of 10,000 particles was released into the model domain on the x = 0 plane (Fig. 2a, Fig. S5). The timedependent coupled velocity-temperature solutions were then recorded as particles move through the domain. The resulting time series for particle temperature, pressure, and position were exported from COMSOL and used in the TERRA cooling age prediction software (Ehlers et al., 2005). Cooling rate dependent ZFT ages were calculated for samples exposed at the surface using the effective closure temperature concept for natural radiation damaged zircon (Brandon et al., 1998; Ehlers et al., 2005). Bendick and Ehlers Supplemental Material 2 82 83 84 85 86 Supplemental Figures Figure S1: Model setup including mesh geometry and boundary conditions for the baseline model. (Top) Top view, (Bottom) View including basal boundary conditions. 87 88 89 Bendick and Ehlers Supplemental Material 3 90 91 92 93 Figure S2: Temperature-dependent viscosity laws. All of the viscosity relations except the constant viscosity case have the Arrhenius form; only the viscosity exponent varies. Larger values of n approach constant viscosity. In all cases, the average effective viscosity, given by the integral of the viscosity curve, is 1021 Pa s. 94 95 96 Bendick and Ehlers Supplemental Material 4 97 98 99 100 101 Figure S3: The temperature distribution in the model domain after 10 Ma of collision. Note the upward advection of geotherms at the indenter front, resulting in locally steep temperature gradients, hence young cooling ages for particles passing near the indenter front. 102 103 104 Bendick and Ehlers Supplemental Material 5 105 106 107 Figure S4: Surface velocity contours for an additional “fast” model, where the convergence rate is doubled from the baseline model to 40 mm/yr. The maximum vertical velocity is 9 mm/yr at the center of the bull’s eye. 108 109 Bendick and Ehlers Supplemental Material 6 110 111 112 113 114 Figure S5: (a) Temperature-time plot for selected particles released in the model domain. Note that it takes approximately 15 Ma for particles to cross the modeled volume from the inflow point; all of the vertical advection occurs within a subsequent 1015 Ma. (b) Depth-temperature plot for selected particles released in the model domain. 115 116 117 118 119 Bendick and Ehlers Supplemental Material 7 120 121 122 123 124 125 126 127 128 129 130 131 132 133 Supplemental Material References Brandon MT, Roden-Tice MK, Garver J, 1998, Late Cenozoic exhumation of the Cascadia accretionary wedge in the Olympic Mountains, northwest Washington State: GSA Bulletin, v. 110, p. 985-1009. Ehlers, T.A., Chaudhri, T., Kumar, S., Fuller, C.W., Willett, S.D., Ketcham, R., Brandon, M.T., Belton, D., Kohn, B., Gleadow, A.J., Dunai, T., Fu, F., 2005, Computational tools for low-temperature thermochronometer interpretation: Reviews in Mineralogy and Geochemistry, v. 58, p. 589-622. Fullsack, P., 1995, An arbitrary Lagrangian-Eulerian formulation for creeping flows and its application in tectonic models: Geophys. J. Int., v. 120,p. 1-23. Molnar, P., and England, P., 1990, Temperatures, heat flux, and frictional stress near major thrust faults: J. Geophys. Res., v. 95, p. 4833-4856. Bendick and Ehlers Supplemental Material 8