[CIG-SHORT] material model for landers/hector mine benchmark

[Brad Aagaard]

Brad-

I think the main obstacle is that PyLith (and likely GeoFEST) can't 
implement a linear variation of material properties within a "layer".
I don't think the "layers" in the material model need to be part of 
the mesh unless there is a significant discontinuity in the elastic 
properties (of which I only see one in your plot of the material 
properties). That is why I thought breaking it up into 10-12 layers 
with only small jumps in elastic properties between layers without 
remeshing would probably work okay. Marking the material model layer 
of the elements would be a little trickier, but it could be done 
simply by marking them according to the vertical coordinate.

Brad


On Thursday 13 July 2006 9:41 am, Bradford H. Hager wrote:
> I'm a little confused about how this would work with Carl's mesh,
> which has very little structure with depth - maybe only 2-3
> boundaries.  Since so most of the elements would straddle the layer
> boundaries in a 20-layer model, wouldn't that be a problem?
>
> Wouldn't a 20-layer model require a new mesh?
>
> If all you really want is a simpler way of interpolating, maybe the
> model could be simplified to about 5 regions with linear variation
> of properties with depth within each region?



>
> Comments?
>
> Brad
>
> On Jul 13, 2006, at 12:18 PM, Charles Williams wrote:
> > I think I would go along with simplifying the model into a
> > reasonable number of layers (probably < 20).  For now, this would
> > make things easier.
> >
> > Charles
> >
> > On Jul 13, 2006, at 12:03 PM, Brad Aagaard wrote:
> >> Hi all:
> >>
> >> Brad Hager has provided a material model for the Landers/Hector
> >> Mine benchmark (see end of email and attachment). The material
> >> model is basically a piecewise linear variation in material
> >> properties with depth with a resolution of 500 m - 1000 m. How
> >> difficult will it be for GeoFEST and PyLith to use this material
> >> model? Would it be more appropriate (i.e., a reasonable
> >> compromise) to simplify the material model into something like
> >> 10-12 layers so that it could be implemented more easily?
> >>
> >> Brad
> >>
> >> On Monday 10 July 2006 12:17 pm, Bradford H. Hager wrote:
> >>> Below in ascii and .pdf (with plot) is
> >>> the SCEC CVM 4.0 Vp, Vs, rho structure at (34.5 N, 116.4 W), a
> >>> point about midway between Landers and Hector Mine.  It is
> >>> given at 1,000 m depth intervals at half km postings, with
> >>> extra sampling at the surface and at the Moho (30 km).  Below
> >>> 49.5 km, set the velocity constant at the value at 49.5 km.
> >>>
> >>> Since we will be heading towards incorporating the CVM, I think
> >>> that interrogating the CVM is a better way to go than setting
> >>> values for a few layers.
> >>>
> >>> Values of the Lame' parameters lambda and mu can be obtained
> >>> from
> >>>
> >>> Vs = sqrt(mu/rho)
> >>> Vp = sqrt[(lambda + 2 mu)/rho]
> >>>
> >>> *******************
> >>>
> >>> Lat	Long	depth	Vp	Vs	rho
> >>> 34.5	-116.4	0	5166.7	2869.2	2680.8
> >>> 34.5	-116.4	500	5216.7	2896.6	2688.7
> >>> 34.5	-116.4	1500	5316.6	2951.3	2704.5
> >>> 34.5	-116.4	2500	5516.2	3060.5	2736
> >>> 34.5	-116.4	3500	5715.8	3169.7	2767.5
> >>> 34.5	-116.4	4500	5915.4	3278.9	2799
> >>> 34.5	-116.4	5500	6103	3440.1	2828.7
> >>> 34.5	-116.4	6500	6213.1	3590.1	2846
> >>> 34.5	-116.4	7500	6257.6	3676.8	2853.1
> >>> 34.5	-116.4	8500	6302.1	3763.6	2860.1
> >>> 34.5	-116.4	9500	6346.6	3850.4	2867.1
> >>> 34.5	-116.4	10500	6371.2	3845.9	2871
> >>> 34.5	-116.4	11500	6375.7	3750.3	2871.7
> >>> 34.5	-116.4	12500	6380.3	3654.6	2872.4
> >>> 34.5	-116.4	13500	6384.8	3559	2873.2
> >>> 34.5	-116.4	14500	6389.4	3463.3	2873.9
> >>> 34.5	-116.4	15500	6394	3367.7	2874.6
> >>> 34.5	-116.4	16500	6369.6	3446.9	2870.8
> >>> 34.5	-116.4	17500	6428.7	3527.2	2880.1
> >>> 34.5	-116.4	18500	6487.7	3607.6	2889.4
> >>> 34.5	-116.4	19500	6546.8	3687.9	2898.7
> >>> 34.5	-116.4	20500	6605.9	3768.3	2908.1
> >>> 34.5	-116.4	21500	6664.9	3848.7	2917.4
> >>> 34.5	-116.4	22500	6700.3	3890.9	2923
> >>> 34.5	-116.4	23500	6712	3895.1	2924.8
> >>> 34.5	-116.4	24500	6723.6	3899.3	2926.7
> >>> 34.5	-116.4	25500	6735.3	3903.5	2928.5
> >>> 34.5	-116.4	26500	6747	3907.7	2930.3
> >>> 34.5	-116.4	27500	6758.6	3911.9	2932.2
> >>> 34.5	-116.4	28500	6770.3	3916.1	2934
> >>> 34.5	-116.4	29500	6782	3920.3	2935.9
> >>> 34.5	-116.4	29999	6787.9	3922.4	2936.9
> >>> 34.5	-116.4	30001	7715.4	4507.7	3083.3
> >>> 34.5	-116.4	30500	7716	4508.3	3083.4
> >>> 34.5	-116.4	31500	7717.2	4509.4	3083.6
> >>> 34.5	-116.4	32500	7718.5	4510.5	3083.7
> >>> 34.5	-116.4	33500	7719.7	4511.6	3083.9
> >>> 34.5	-116.4	34500	7720.9	4512.7	3084.1
> >>> 34.5	-116.4	35500	7722.2	4513.8	3084.3
> >>> 34.5	-116.4	36500	7723.4	4514.9	3084.5
> >>> 34.5	-116.4	37500	7724.6	4516.1	3084.7
> >>> 34.5	-116.4	38500	7725.9	4517.2	3084.9
> >>> 34.5	-116.4	39500	7727.1	4518.3	3085.1
> >>> 34.5	-116.4	40500	7728.6	4519.5	3085.3
> >>> 34.5	-116.4	41500	7730.4	4520.9	3085.6
> >>> 34.5	-116.4	42500	7732.2	4522.2	3085.9
> >>> 34.5	-116.4	43500	7734	4523.6	3086.2
> >>> 34.5	-116.4	44500	7735.8	4524.9	3086.5
> >>> 34.5	-116.4	45500	7737.6	4526.2	3086.8
> >>> 34.5	-116.4	46500	7739.4	4527.6	3087.1
> >>> 34.5	-116.4	47500	7741.2	4528.9	3087.3
> >>> 34.5	-116.4	48500	7743	4530.3	3087.6
> >>> 34.5	-116.4	49500	7744.8	4531.6	3087.9
> >>> <Landers_structure.pdf>
> >>
> >> _______________________________________________
> >> CIG-SHORT mailing list
> >> CIG-SHORT at geodynamics.org
> >> http://geodynamics.org/cgi-bin/mailman/listinfo/cig-short
> >
> > Charles A. Williams
> > Dept. of Earth & Environmental Sciences
> > Science Center, 2C01B
> > Rensselaer Polytechnic Institute
> > Troy, NY  12180
> > Phone:    (518) 276-3369
> > FAX:        (518) 276-2012
> > e-mail:    willic3 at rpi.edu
> >
> >
> > _______________________________________________
> > CIG-SHORT mailing list
> > CIG-SHORT at geodynamics.org
> > http://geodynamics.org/cgi-bin/mailman/listinfo/cig-short