[CIG-LONG] Faults do not depart away from the axis

[Walter Landry]

Taichi SATO <taichix at aori.u-tokyo.ac.jp> wrote:
> Dear Walter and everyone,
> 
> How is going on about dyke hill problem?
> I'd like to include this numerical modeling as a part of my PhD thesis, 
> so I am waiting for the solution, although it is my personal problem.

Sorry it has taken so long.  This problem is actually important to me,
because it is a realistic model that a number of people are interested
in.  You have been incredibly patient, so I have been able to refine
the model with your help.

In any case, I think I finally figured out how to get it to work.  I
noticed in the Buck et. al. Nature 2005 paper that they have a dike
region that covers the entire height of the model.  That turns out to
be important for reasons that are not entirely clear.

Also, to simulate the material in the dike, I modified the
NonNewtonian rheology such that, if the temperature is high enough,
then the material melts and has a uniformly low viscosity.  I then set
the temperature of the dike to be above the melting point.

The temperature is still fixed throughout the domain, so as the
material moves out of the dike, it has a lower temperature and starts
accumulating strain.  So there is no need for a strain limiting
region.

I am including the input file and a screenshot after 100 steps.  You
will have to update your code from the repository.  I ran it with the
command

  ./build/bin/Gale dike_air.xml -ksp_type preonly -pc_type lu --maxTimeSteps=100 --outputPath=output_dike_air_1480_full | tee gale_out_air_1480_full 

Let me know if that works for you.

> In this time, I have another questions.
> 1. How many time steps are equivalent to 1 kyr or 1Myr, assuming the 
> scaling factor is 10^22 s ?
>     I would line to know the scaling, XX steps = 1kyr, or 1step = XX kyr.

The time stepping is variable.  So steps 1-10 might cover a small
amount of time, while 11-20 might cover a larger amount of time.  For
every timestep, Gale prints out something like 

  TimeStep = 48, Time = 9.76277e-10

So if you are scaling your time by 10^22, then that means that the
data for timestep 48 is at

  9.76277e12 seconds = 309575 years

> 2. Would you tell me the way to set the asthenosphere as visco-elastic 
> material and the lithosphere as elasto-plastic material ?
> For visco-elastic material, I think one is to set large value for 
> initial Softening Strain, is it OK?

Gale does not have elasticity.  Sorry.  I thought it might have it
after the merge, but it was not included in the code I received.

Cheers,
Walter Landry
walter at geodynamics.org
-------------- next part --------------
A non-text attachment was scrubbed...
Name: dike_air.xml
Type: text/xml
Size: 37573 bytes
Desc: not available
Url : http://geodynamics.org/pipermail/cig-long/attachments/20100709/96b8d60d/attachment-0001.bin 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: dike_accumulated_strain.png
Type: image/png
Size: 196743 bytes
Desc: not available
Url : http://geodynamics.org/pipermail/cig-long/attachments/20100709/96b8d60d/attachment-0001.png