Wills Map
From SCECpedia
Contents
grd2etree
Found background information from
https://scec.usc.edu/scecpedia/UCVM_Utah https://github.com/SCECcode/UCVMC/wiki/Manual-Pages
and found where Patrick probably worked at in generating the ucvm.e that we are currently using:
/home/scec-00/patrices/opt/vs30
created a configure file sample.conf
# grid2etree UCM Map config file # Domain corners coordinates (degrees): proj=+proj=aeqd +lat_0=36.0 +lon_0=-120.0 +x_0=0.0 +y_0=0.0 lon_0=-129.75 lat_0=40.75 rot=55.0 # Domain dimensions (meters): x-size=1800000.0 y-size=900000.0 # Spacing spacing=250.0 # Etree parameters and info title=UCVM_Elev_Vs30_Map_Wills_Wald author=P_Small date=05/2011 outputfile=ucvm.e # Grid data directories elev_hr_dir=/home/scec-00/patrices/opt/ned elev_lr_dir=/home/scec-00/patrices/opt/bath vs30_hr_dir=/home/scec-00/patrices/opt/vs30/wills_gridfloat vs30_lr_dir=/home/scec-00/patrices/opt/vs30/wald_gridfloat
NOTE: the ucvm.e newly recreated is 1.2G but UCVMC's ucvm.e is 683M
command used to generate ucvm.e
./grd2etree -f sample.conf
new ucvm.e
A new ucvm.e is created by replacing the vs30_hr_dir with the rasterized data file from Kevin and update the hdr file to match with what is expected by grd2etree
#BYTEORDER I LAYOUT BIL nrows 37900 ncols 41104 NBANDS 1 NBITS 32 BANDROWBYTES 164416 TOTALROWBYTES 164416 PIXELTYPE FLOAT #ULXMAP -124.406528862058 #ULYMAP 42.0090923024621 xllcorner -124.406528862058 yllcorner 42.0090923024621 cellsize 0.00025 NODATA_value -9999 byteorder LSBFIRST XDIM 0.00025 YDIM 0.00025
Verify using CCA+GTL
| With fixed scalebar, | ||
|---|---|---|
| With variable scalebar, | ||
command used to generate the plots,
./plot_horizontal_slice.py -b 33,-123 -u 39.5,-115 -d vs -c cca -a sd -s 0.05 -e 0 -o cca.png
Comparison of Wills Map 2006 versus 2015
These two wills maps were made from gis files released by the state.
These two wills maps are extracted by ucvm. The etree vs30 values are based on wills 2006, so we expect to see data from above in these maps.
Description of Wills Digital Map
2015 is the latest version that I am aware of. The supplementary data of the paper provides the map as a GIS shapfile (http://www.seismosoc.org/Publications/BSSA_html/bssa_105-6/2015105-esupp/index.html). I then rasterized it to a binary floating point file (which similar to the 2006 file, except with floats instead of short ints) using the gdal_rasterize command (http://www.gdal.org/gdal_rasterize.html).
The original and rasterized data is stored here on opensha.usc.edu:
/export/opensha-00/data/siteData/wills_2015
I suggest you use raster_0.00025.flt, which is rasterized with 0.00025 degree spacing (~25 meters). The raster_0.00025.hdr file contains the necessary coordinates for georeferencing it, but I'll summarize here:
The first location in the file is at:
x/longitude/ULXMAP = -124.406528862058
y/latitude/ULYMAP = 42.0090923024621
This is the upper left (northwest) point in the map.
numX/numLon/NCOLS = 41104
numY/numLat/NROWS = 37900
Grid spacing (both X and Y): 0.00025 degrees
Data type: 4-byte/32-bit little-endian floating point numbers
Mesh ordering: fast XY, so the first few points are:
Index 0 (at byte 0): -124.406528862058, 42.0090923024621
Index 1 (at byte 4): -124.406528862058+0.00025, 42.0090923024621 = -124.406278862058, 42.0090923024621
Index 2 (at byte 8): -124.406528862058+0.0005, 42.0090923024621 = -124.406028862058, 42.0090923024621
...
