[CIG-SHORT] Intermediate Results

[Brad Aagaard]

On 3/11/13 11:17 AM, BOK10 at pitt.edu wrote:
> The differences I'm noticing between the two simulations (one with loose
> tolerances, and the other with tighter tolerances), is when I'm looking at
> how much slip there has been. The time (just based off the timestamp) it
> takes for a fault to slip >1m in the loose tolerance model is 300 years
> (60 timesteps), whereas in the tight tolerance version it's 100 years (20
> timesteps). I'm specifically looking at the length of time it takes for
> slip to exceed a certain amount in my problem.

I understand that you have a certain quantity you are comparing between 
simulations, but focusing on just that number is not helping you 
identify the causes of the differences. You need to dig deeper in the 
simulation output. Look at the time history of the deformation 
(displacements and slip). Find what is different. Isolate the time when 
do the differences appear. Try to determine if the differences are 
localized in space.

Additionally, examine if the behavior is identical before any earthquake 
occurs. Determine if the first earthquake occurs at the same time and 
with the same amount of slip in all of the simulations. If not, isolate 
the differences in the stress on the fault.

Brad