Sections

Introduction
Quick start
Regions

Contiguous US

Sample output
Distribution

Alaska

Sample output
Distribution

Distribution Discussion

Organization of Directories
Format of Inversion Output
Software

Reference

Herrmann, R. B., C. J. Ammon, H. M. Benz, A. Aziz-Zanjani and J. Boschelli (2021). 
Short-Period Surface-Wave Tomography in the Continental United States—A Resource for Research 
Seismol. Res. Lett. 92, 3642–3656
doi: 10.1785/0220200462.

Introduction

This distribution presents surface-wave tomography results for the contiguous United States and Alaska. This effort started many years ago with tomography code written by Charles J. Ammon. Initially only group velocity measurements were obtained from earthquake data, but later these were augmented through an analysis of ambient noise cross-correlations to obtain phase and group velocities for Love and Rayleigh waves. The distinguishing aspect of this effort is the focus on dispersion at periods less than 5 seconds. This was possible though the use of BH and HH data streams instead of the 1Hz LH streams used in other studies. Thus the computer storage and computation times were substantial compared to other using the LH channels.

We have found the results to be a useful tool for regional earthquake modeling by permitting us to answer the question "What seismic model should be used for this earthquake?"

This distribution provides the following:

SLU tomography results with codes to access those results and to make plots
SLU tomography data sets and inversion codes so that others can include our measurements in their inversions using our codes or some other tomography code
Access to Love and Rayleigh phase velocity data from Ekström. http://www.ldeo.columbia.edu/~ekstrom/Projects/ANT/USANT15.html Ekström, G. (2017). Short-period surface-wave phase velocities across the conterminous United States, Phys. Earth Planet Int. , http://dx.doi.org/10.1016/j.pepi.2017.07.010
Access to global dispersion model GDM52 http://www.ldeo.columbia.edu/~ekstrom/Projects/SWP/GDM52.html. Ekström, G. (2011). A global model of Love and Rayleigh surface wave dispersion and anisotropy, 25-250 s, Geophys. J. Int., doi:10.1111/j.1365-246X.2011.05225.x group velocity tomography
Access to representative reference velocity models to predict dispersion, to include the CUS and WUS models from SLU and CRUST1.0 http://igppweb.ucsd.edu/~gabi/crust1.html

The distribution consists of gzip'd tar files which will create the directory structure. For example,

gunzip -c SLU.CUS.OBSERVATIONS.tgz | tar xf -

will create the directory SLU.CUS and then the subdirectory SLU.CUS/OBSERVATIONS. The tomography code just requires a C compile, e.g., gcc. Codes to search though the tomography results require a FORTRAN compiler, e.g., gfortran. Some shell scripts which create graphics require GMT4 or the Computer Programs in Seismology codes.

Quick start

If you just require the dispersion as a function of latitude and longitude without any graphics, just perform the following steps (the CUS results will be used as an example):

Download and install SLU.CUS.display.tgz, SLU.CUS.other.tgz, SLU.CUS.result.tgz and SLU.CUS.srcbin.tgz
Compile the software (SoftwareScripts.). The important code that will be used is the script FMKDSP which calls the executable doquad.

The commands to unpack the tgz files, compiled the codes and then access the dispersion results are as follow:

gunzip -c SLU.CUS.display.tgz | tar xf -
gunzip -c SLU.CUS.other.tgz | tar xf -
gunzip -c SLU.CUS.result.tgz | tar xf -
gunzip -c SLU.CUS.srcbin.tgz | tar xf -
cd SLU.CUS
cd bin/General
make all
cd ..
make all
cd ../nGRIDREGION
../bin/FMKDSP 39 -91 .. > disp.txt

The dispersion at 39N 91W will be in the file  disp.txt   in the surf96 format of Computer Programs in Seismology. The format is very simple as shown in this example:

SURF96 R U X 0 2.000000 2.8629 0.05
SURF96 L U X 0 2.000000 3.3987 0.05
SURF96 R C X 0 2.000000 3.0831 0.05
SURF96 L C X 0 2.000000 3.4015 0.05

Column 1: SURF96 defines the file type
Column 2: R or L to indicate Rayleigh or Love wave, respectively
Column 3: C U or G to indicate phase velocity (km/s), group velocity (km/s) or anelastic attenuation (1/km), respectively
Column 4: X or T for experimental or theoretical
Column 5: an integer indicating the mode. 0 for fundamental, 1 for 1st, etc
Column 6: period in seconds
Column 7: dispersion in km/s or anelastic attenuation coefficient in units of (1/km)
Column 8: error in column 7. Not used here. The 0.05 is just a place holder

For a more interesting display results from the following sequence:

cd SLU.CUS
cd nGRIDREGION
FDODISP 39 -91

This creates the dispersion files tomocus.disp, tomoek.disp and tomogdm52.dist for the SLU.CUS, Ekström (2017) and Ekström (2011), respectively. These are also combined in the file all.disp. In addition the CALPLOT file ALL.PLT is created. This figure is displayed in 39.-91.png. If the command FDODISP 60 -140 is executed in SLU.AK/nGRIDREGION, the file tomoek.disp will have zero characters since the Ekström (2017) results are just for the contiguous U.S. Also, there will be a tomoak.disp in lieu of tomocus.disp.

Contiguous United States

Sample Output

Some examples of what can be done with the tomography results are CUSExamples.

Distribution

The .tgz files for this region are as follow:

SLU.CUS.data.tgz	  SLU.CUS.processing.tgz
SLU.CUS.display.tgz	  SLU.CUS.result.tgz
SLU.CUS.movie.tgz	  SLU.CUS.spk.processing.tgz
SLU.CUS.OBSERVATIONS.tgz  SLU.CUS.spk.tgz
SLU.CUS.other.tgz	  SLU.CUS.srcbin.tgz

The file size and creation data of each are as follow:

 214881440 Dec 11 13:02 SLU.CUS.data.tgz
    132879 Dec 11 13:01 SLU.CUS.display.tgz
  11620190 Apr  3 15:05 SLU.CUS.movie.tgz
 520782962 Dec 11 13:03 SLU.CUS.OBSERVATIONS.tgz
   4756727 Dec 11 13:01 SLU.CUS.other.tgz
     32857 Dec 11 13:01 SLU.CUS.processing.tgz
 206119704 Dec 11 13:01 SLU.CUS.result.tgz
     28886 Dec 11 13:02 SLU.CUS.spk.processing.tgz
 174201792 Dec 11 13:02 SLU.CUS.spk.tgz
     66474 Dec 11 13:01 SLU.CUS.srcbin.tgz

The entries in the table detail the individual .tgz files and provides links for download. They are unpacked using the command gunzip -c file.tgz | tar xf - . Depending on the use, not all files must be downloaded and installed. Obviously the most important are the original measurements in OBSERVATIONS.

File	View results	Redo Inversion	Description	Discussion
SLU.CUS.data.tgz	NO	YES	Complete data set for the SLU inversion.
SLU.CUS.display.tgz	YES	YES	The directory nGRIDREGION for plotting dispersion result for a given latitude/longitude
SLU.CUS.movie.tgz	YES	NO	MP4 movies showing dispersion maps as a function of period.
SLU.CUS.OBSERVATIONS.tgz	NO	NO	Original observations by source that were use to create the large MALL.DSP.gz file used	ObservationFormat/
SLU.CUS.other.tgz	YES	YES	Ekstrom17 and GDM52 dispersion. Also the CRUST1.0 and SLU velocity models
SLU.CUS.processing.tgz	NO	YES	Processing scripts for performing the SLU tomography
SLU.CUS.result.tgz	YES	NO	Tomography results
SLU.CUS.spk.processing.tgz	NO	YES	Scripts for spike test resolution analysis
SLU.CUS.spk.tgz	YES	NO	Results of the spike test
SLU.CUS.srcbin.tgz	YES	YES	Codes for inversion and display

Alaska

Sample Output

Some examples of what can be done with the tomography results are given in AlaskaExamples.

Distribution

The .tgz files for this study are are as follow:

SLU.AK.data.tgz		 SLU.AK.processing.tgz
SLU.AK.display.tgz	 SLU.AK.result.tgz
SLU.AK.movie.tgz	 SLU.AK.spk.processing.tgz
SLU.AK.OBSERVATIONS.tgz  SLU.AK.spk.tgz
SLU.AK.other.tgz	 SLU.AK.srcbin.tgz

The file size and creation data of each are as follow:

  21577393 Dec 11 13:01 SLU.AK.data.tgz
     87827 Dec 11 13:01 SLU.AK.display.tgz
   5055211 Apr  3 15:05 SLU.AK.movie.tgz
  39445686 Dec 11 13:01 SLU.AK.OBSERVATIONS.tgz
   4757288 Dec 11 13:01 SLU.AK.other.tgz
     32986 Dec 11 13:00 SLU.AK.processing.tgz
  82739738 Dec 11 13:00 SLU.AK.result.tgz
     30536 Dec 11 13:01 SLU.AK.spk.processing.tgz
  96691858 Dec 11 13:01 SLU.AK.spk.tgz
     65425 Dec 11 13:01 SLU.AK.srcbin.tgz

The entries in the table below detail the individual .tgz files and provides links for download. They are unpacked using the command gunzip -c file.tgz | tar xf - . Depending on the use, not all files must be downloaded and installed. Obviously the most important are the original measurements in OBSERVATIONS.

File	View results	Redo Inversion	Description	Discussion
SLU.AK.data.tgz	NO	YES	Complete data set for the SLU inversion.
SLU.AK.display.tgz	YES	YES	The directory nGRIDREGION for plotting dispersion result for a given latitude/longitude
SLU.AK.movie.tgz	YES	NO	MP4 movies showing dispersion maps as a function of period.
SLU.AK.OBSERVATIONS.tgz	NO	NO	Original observations by source that were use to create the large MALL.DSP.gz file used	ObservationFormat/
SLU.AK.other.tgz	YES	YES	Ekstrom17 and GDM52 dispersion. Also the CRUST1.0 and SLU velocity models
SLU.AK.processing.tgz	NO	YES	Processing scripts for performing the SLU tomography
SLU.AK.result.tgz	YES	NO	Tomography results
SLU.AK.spk.processing.tgz	NO	YES	Scripts for spike test resolution analysis
SLU.AK.spk.tgz	YES	NO	Results of the spike test
SLU.AK.srcbin.tgz	YES	YES	Codes for inversion and display

Distribution Discussion

Organization of Directories

will be seen. If everything is unpacked, the following directory structure discussed in DirectoryStructure/ will be seen.

What can be done with the inversion results?

The examples linked above ( CUSExamples and AlaskaExamples ) present various plots that highlight how the inversion results can be used to characterize the dispersion as a function of latitude and longitude.

Format of Inversion Output

In each region, there are subdirectories for each dispersion data set: LOVEc for Love wave phase velocity; LOVEU for Love wave group velocity; RAYLc for Rayleigh wave phase velocity and RAYLU for Rayleigh wave grout velocity. Example directories would be SLU.AK/LOVEc or SLU.CUS/RAYLU. The organization and description of vc_XXX_001.xyz and per.uniq.NUM files are described in TomographyFormat/

Software

After downloading and unpacking the srcbin/t.tgz file, instructions on how to compile the codes and their use are given in SoftwareScripts.

To use the tomography results that are distributed, it is only necessary to compile the codes in the src/General and src directories. One will then be able to get the dispersion from this study as well as from the Ekström (2011, 2017) studies in separate files.

To use other scripts, the Computer Programs in Seismology codes must be installed. In addition the conversion of Encapsulated Postscript to PNG files used the ImageMagick software which is available for MacOS, LINUX and Windows.