[CIG-SHORT] questions about pylith spatial data base

[Brad Aagaard]

On 7/31/17 5:14 AM, tu xiang wrote:
> Hello, everyone
> I have several questions about pylith spatial data base and hope someone
> would like to give me a hand.
>
>
>     Question 1

data-dim refers to the topology of the points in the spatial database. A 
single point is a data-dim of 0. Points along a line are a data-dim of 
1. Points on a plane are data-dim of 2. Points in a volume are a 
data-dim of 3.

space-dim refers to the spatial dimension of the problem. We specify the 
spatial dimension in two locations, once for the data set and once for 
the coordinate system (we use this same coordinate specification in 
other applications so we didn't want to remove it). The spatial 
dimensions must match in both places.

>
> // -*- C++ -*- (tell Emacs to use C++ mode for syntax highlighting)
>
> //
>
> // This spatial database specifies the distribution of slip on the
>
> // fault surface. In this case we prescribe a piecewise linear, depth
>
> // dependent distribution of slip. The slip is 2.0 m left-lateral
>
> // with 0.25 m of reverse slip at the surface with a linear taper from
>
> // 2.0 m to 0.0 m from -2 km to -4 km.
>
> //
>
> #SPATIAL.ascii 1
>
> SimpleDB {
>
>   num-values = 3
>
>   value-names =  left-lateral-slip  reverse-slip  fault-opening
>
>   value-units =  m  m  m
>
>   num-locs = 3
>
>   data-dim = 1 // Locations of data points form a line.  What does
> data-dim mean? It means linear interpolation?
>
>   space-dim = 3  what does this space-dim mean? What is the difference
> between this space-dim and the next space-dim?
>
>   cs-data = cartesian {
>
>     to-meters = 1.0e+3 // Specify coordinates in km for convenience.
>
>     space-dim = 3  what does this space-dim mean? What is the difference
> between this space-dim and the above space-dim?
>
>   } // cs-data
>
> } // SimpleDB
>
> // Columns are
>
> // (1) x coordinate (km)
>
> // (2) y coordinate (km)
>
> // (3) z coordinate (km)
>
> // (4) left-lateral-slip (m) (right-lateral is negative)
>
> // (5) reverse-slip (m)
>
> // (6) fault-opening (m)
>
> 0.0  0.0  0.0     2.00  0.25  0.00
>
> 0.0  0.0 -2.0     2.00  0.00  0.00
>
> 0.0  0.0 -4.0     0.00  0.00  0.00
>
>
>
>
> Figure 1
>
>
>     Question 2

The topology of the spatial database (data-dim) does not have to match 
the object you are specifying the spatial variation for. You can specify 
slip as a function of depth using points along a line. For a nonplanar 
fault to specify a 2D variation in the slip, you can specify points on a 
plane (horizontal or vertical or 3D) or points on a 3D surface. If you 
use a 3D surface, we recommend using a fine grid a nearest interpolation.

>
> cs-data = cartesian {
>
>     to-meters = 1.0e+3 // Specify coordinates in km for convenience.
>
>     space-dim = 3
>
>   } // cs-data
>
> } // SimpleDB
>
> // Columns are
>
> // (1) x coordinate (km)
>
> // (2) y coordinate (km)
>
> // (3) z coordinate (km)
>
> // (4) left-lateral-slip (m) (right-lateral is negative)
>
> // (5) reverse-slip (m)
>
> // (6) fault-opening (m)
>
> 0.0  0.0  0.0     2.00  0.25  0.00
>
>
>
> Is the x, y, z coordinate must on the fault? If I construct a curve
> fault (Figure 1), it will be difficult to ensure the point (x, y, z) is
> on the fault because of the precision of the value.
>
>
>
>
>
>  Figure 2 curve fault
>
>
>     Question 3
>
> If I want to specify a complex slip or traction distribution (Figure 2),
> how to specify?

See the examples/3d/subduction/step06 in the PyLith tutorials from the 
CDM2017 workshop (https://wiki.geodynamics.org/software:pylith:cdm2017).

>
>
> Figure 2 complex slip or traction distribution
>
> Looking forward to your reply.
> Best regards,
> Tu Xiang
>
>
>