[CIG-SHORT] Problem with fault intersection in 3D simulations

[Brad Aagaard]

On 02/14/2018 07:16 AM, Josimar Alves da Silva wrote:
> Hi Surendra,
> 
> Thank you for your suggestion ! I am able to run my simulation fine now 
> and I don't get the previous error message. In this approach what you 
> suggest is to split the Thrust fault in two different faults that are 
> not connected (with the strike slip fault between them). I assume that 
> the new Thrust fault segments are allowed to slip independently now, 
> correct ?

When you have two faults that intersect in an X-style intersection, you 
need to pick one fault to be the through-going fault and one to be split 
into two segments (which should be treated as completely independent 
faults with their own nodesets, buried edges, spatial databases, etc). 
It sounds like you picked the strike-slip fault to be the through-going 
fault. In that case, the two segments of the thrust fault will be 
separated; if the thrust fault slips without any slip on the strike-slip 
fault, there will be a region of zero slip along the intersection with 
the strike-slip faults.

If you remove the vertices of the strike-slip fault from the thrust 
fault, you need to mark the vertices that are along that edge as buried 
edges. Otherwise, the fault will be extended one cell past the edge (as 
shown in the attached image from the troubleshooting tutorial; you 
should create a nodeset corresponding to the buried edges of the purple 
region).

> Charles, do you have any comments regarding this solution proposed by 
> Surendra ? Have you done something similar in the past ?

The preferred approach for fault intersections is to NOT remove the 
vertices on the strike-slip fault from the thrust fault. Simply include 
those vertices in the buried edges of the thrust fault while treating 
the thrust fault as two completely separate faults. Note that you must 
list the through-going fault FIRST in the list of interfaces, so that it 
is created first and then the two thrust fault segments.

Brad
-------------- next part --------------
A non-text attachment was scrubbed...
Name: faultedges.png
Type: image/png
Size: 6904 bytes
Desc: faultedges.png
URL: <http://lists.geodynamics.org/pipermail/cig-short/attachments/20180214/d236ef72/attachment-0001.png>