[cig-commits] [commit] devel, master: added a readme.txt to DATA/crustmap/ (by Matthias Meschede) (1db8550)

cig_noreply at geodynamics.org cig_noreply at geodynamics.org
Thu Nov 6 08:04:06 PST 2014


Repository : https://github.com/geodynamics/specfem3d_globe

On branches: devel,master
Link       : https://github.com/geodynamics/specfem3d_globe/compare/bc58e579b3b0838a0968725a076f5904845437ca...be63f20cbb6f462104e949894dbe205d2398cd7f

>---------------------------------------------------------------

commit 1db8550dc4f7561097b906c7085102ff9fc6c306
Author: Daniel Peter <peterda at ethz.ch>
Date:   Wed Mar 3 01:51:05 2010 +0000

    added a readme.txt to DATA/crustmap/ (by Matthias Meschede)


>---------------------------------------------------------------

1db8550dc4f7561097b906c7085102ff9fc6c306
 crustmap/readme.txt | 89 +++++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 89 insertions(+)

diff --git a/crustmap/readme.txt b/crustmap/readme.txt
new file mode 100644
index 0000000..a7064ab
--- /dev/null
+++ b/crustmap/readme.txt
@@ -0,0 +1,89 @@
+CRUSTMAPS Version 0.0.0.1
+
+---------------------------------------------------------------------------------------------------
+HOW TO USE THE INCLUDED CRUSTMAPS:
+
+to use CRUSTMAPS uncomment in constants.h.in (before ./configure):
+
+"
+!!  integer, parameter :: ITYPE_CRUSTAL_MODEL = ICRUST_CRUSTMAPS
+"
+
+SPECFEM3D_GLOBE is set to use the crustmaps provided in the SPECFEM3D_GLOBE/DATA/crustmap/ folder.
+
+-------------------------------------------------------------------------------------------------
+WHAT ARE CRUSTMAPS?
+
+the newly provided crustmaps in SPECFEM3D_GLOBE are a set of maps to provided a high resolution
+crust model in SPECFEM3D_GLOBE on specified gridpoints in the crust.
+Crustmaps can be used to integrate different high resolution crust models in lower resolution
+models like crust2.0.
+
+There are 4 types of crustmaps:
+
+topography crustmaps (e.g. eucrustt3.cmap - eucrustt7.cmap)
+s-velocity crustmaps (e.g. eucrusts3.cmap - eucrusts7.cmap)
+p-velocity crustmaps (e.g. eucrustp3.cmap - eucrustp3.cmap)
+density    crustmaps (e.g. eucrustr3.cmap - eucrustr7.cmap)
+
+these are simple ascii data files and you can change them easily. The current crustmap resolution
+is set to 1/4 degree.
+The Values are ordered on a grid that runs from 90° lat (0° lng to 180° lng ) to -90° lat.
+Every line in a crustmap files sets all values for one specific latitude:
+
+(length of one line = 360 * 4 entries)
+(total number of lines = 180 * 4)
+
+The value for (lat = 89.875, long = 0.125) is stored in entry (line 1,column 1).
+you get the value that is north west of your (lat, long) point by using:
+
+line = int(0.5+((90.0-lat)*CRUSTMAP_RESOLUTION))
+column = int0.5+(xlng*CRUSTMAP_RESOLUTION)
+
+Values in between points are interpolated.
+the topography file sets the thickness of every layer (positive number, depth under next layer)
+the s velocity, p velocity and density maps set the respective values for every (lat,lng) point in every layer
+The current Layer model follows crust2.0 (7 Layers). The ocean and ice layer of crust2.0 are not used,
+since they can't be used by the code. The crystalline crust is (rather arbitrarily) seperated
+into 3 layers, upper, middle and lower crust.
+
+The provided maps combine The moho topography of crust2.0 and Eucrust with the velocity values from Crust2.0
+(Layers not provided by Eucrust are taken from crust2.0 and stretched or squeezed according to the new
+topography)
+
+Specfem will stretch the Moho as regarding to the lowest layer.
+The first two layers are regarded sediment layers (see option INCLUDE SEDIMENTS in constants.h)
+but except from this option they are not treated differently.
+
+----------------------------------------------------------------------------------------------------
+
+HOW TO CREATE YOUR OWN CRUSTMODEL:
+
+To create your own crustmodel you have to provide your own set of crustmaps. (you can either change the
+filenames in model_crustmaps.f90 --> subroutine read_general_crustmap() or exchange the given files)
+
+Make sure you provide the same Resolution as the default Crustmaps (1/4 degree) or change the Variable
+CRUSTMAPS_RESOLUTION in constants.h.in. You can also change the Variable NLAYERS_CRUSTMAPS and try to use
+a different number of crustmap layers (not tested).
+
+If you have a set of new local crust layers, and want to integrate it in e.g. crust 2.0. It is easiest to
+use a matlab/octave function like griddata to interpolate between all your given data points.
+
+If your crust model is not complete (e.g. you have only 2 layers and not 3 or no velocity or density data)
+you should try to combine your data with crust2.0 to provide a complete set of values.
+You can also just change the different topography maps and the old velocity etc.. values are assigned to your
+new points.
+
+IMPORTANT JACOBIAN PROBLEMS:
+
+if you experience Jacobian problems while running the mesher, you have to
+
+I. smooth your model:
+go to line 479 in the model_crustmaps.f90 and try to understand and change the following.
+
+II.
+if that doesn't work, you have to change your mesh resolution (NEX,...)
+provide a less dramatic crustmap or turn off moho stretching.
+
+
+



More information about the CIG-COMMITS mailing list