The example in this directory simulates quasi-static interseismic and
coseismic deformation for a subduction zone. It is based on the 2011
M9.0 Tohoku earthquake off the east coast of Japan.

The main features of this example are:

  * Generating a finite-element mesh using CUBIT
    + Nonplanar geometry
    + Variable mesh resolution
  * Spatially variable coseismic slip and creep
  * Maxwell viscoelastic relaxation

The example is broken up into three steps of increasing
complexity. Step 1 focuses on the coseismic slip, Step 2 focuses on
interseismic deformation, and Step 3 combines the two into a pseudo
earthquake cycle deformation simulation. In a similar fashion as other
examples, we place all of the general parameters in pylithapp.cfg and
the parameters specific to each simulation in stepXX.cfg.

We model the crust with a linear elastic bulk constitutive model and
the mantle with a linear Maxwell viscoelastic model. Both constitutive
models use plane strain formulations in these 2-D models.

The parameters for the bulk constitutive models are defined in
  mat_concrust.spatialdb
  mat_oceancrust.spatialdb
  mat_conmantle.spatialdb
  mat_oceanmantle.spatialdb

The simulation will output the displacements on the ground surface at
every time step, the displacements over the entire domain every 10
years, the fault slip and tractions every time step, and the stresses
and strains for each material every 10 years.

For each of the simulations, we recommend examining the displacements,
stress field, and fault slip. ParaView parses the 2-D stress tensor as
a vector, so the tensor components sigma_xx, sigma_yy, sigma_xy are
listed as x, y, z.


Mesh generation (optional)

  NOTE: The result of this step will overwrite the included file
        tri3_mesh.exo. You may want to copy/rename this file so that
        you have a backup copy in case you have difficulty running
        CUBIT.

  Start CUBIT and play the journal file "mesh_tri3.jou". We highly
  recommend that you study the contents of the journal files to
  understand the mesh generation process.


Step 1. Coseismic slip simulation

  This simulation involves coseismic slip between the continental
  crust and top of the subducting oceanic crust. The slip also extends
  down into the top of the mantle below the continental crust.

  The parameters for the earthquake slip are defined in
    fault_slip_coesismic.spatialdb

  Run the simulation via the following command:
    pylith step01.cfg.


Step 2. Interseismic deformation simulation

  This simulation involves aseismic creep along the interfaces between
  the subducting oceanic crust and the mantle. The slip rate is a
  constant 8 cm/yr.

  The parameters for the creep along the top of the slab are defined in
    fault_creep_slabtop.spatialdb

  Run the simulation via the following command:
    pylith step02.cfg


Step 3. Pseudo-earthquake cycle model

  This simulation combines the interseismic deformation from Step 2
  with the coseismic slip from Step 1. The earthquake rupture occurs
  midway (150 years) through the 300 year simulation.

  Run the simulation via the following command:
    pylith step03.cfg


Step 4. Friction controlled afterslip

  This simulation uses the stress changes associated with the the
  coseismic deformation in Step 1 in a simulation of afterslip
  governed by static friction. The afterslip occurs at the down-dip
  extent of rupture where the coseismic slip increases the shear
  tractions.

  Run the simulation via the following command:
    pylith step04.cfg

Suggestions variations

  The list below includes some suggested modifications to the problem
  that will allow you to become more familiar with PyLith while
  examining some interesting physics.

  * Change the resolution of the mesh by editing the mesh_tri3.jou
    journal file. Change the resolution and bias factor.

  * Add depth dependent viscosity to the mantle and crust. This
    requires using the linear Maxwell plane strain bulk constitutive
    model in the crust as well and creating spatial databases that
    include viscosity for the crust. Specifying a depth dependent
    variation in the parameters will require adding points, updating
    num-locs accordingly, and changing data-dim to 1.

  * Modify the spatial database files for the material properties to
    use depth-dependent elastic properties based on PREM (Dziewonski
    and Anderson, 1981, 10.1016/0031-9201(81)90046-7). See
    http://geophysics.ou.edu/solid_earth/prem.html for a simple table
    of values.

    Add points, update num-locs accordingly, and change data-dim to be 1.

  * Modify the CUBIT journal files to use quad4 cells rather than tri3
    cells. This requires using the pave mesh scheme.

  * Create a simulation with multiple earthquake cycles by lengthening
    the duration of the simulation and adding additional earthquake
    ruptures. See examples/3d/hex8/step06.cfg for an example with
    multiple earthquake ruptures. Examine spinup towards a
    steady-state solution.

