Gsnoop Recipes
K. Baver
Abstract:
This document presents steps for producing some commonly used output from
Gsnoop, the program that extracts global parameter information from Solve
spoolfiles. Some examples include advice on setting up the Solve solutions
that create the spoolfiles. Explanations of the Gsnoop commands are minimized
for the benefit of users who wish to produce output without spending a lot of
time learning Gsnoop. For better explanations of the commands and general
information about Gsnoop, please refer to the Gsnoop User's Guide.
Table of contents:
- 1 Use of this Document
- 2 Preliminary steps
- 3 Production of topocentric velocity (tvel) files
- 4 Production of solution files from a global solution
- 5 Global TRF IERS SINEX output
-
- 5.1 Solve steps
-
-
- 5.2 Steps for Gsnoop run 1 (site and covariance run)
-
-
- 5.3 Steps for Gsnoop run 2 (apriori run)
-
-
- 5.4 Editing of Gsnoop files
-
-
- 5.5 Concatenation of section files into SINEX output file
-
1 Use of this Document
This document is organized as follows:
Preliminary Steps section
Sections listing steps to produce specific types of Gsnoop output
global TRF IERS SINEX section
Most Gsnoop output is simple to produce. Users can follow a few preliminary
steps, then follow a few more steps for a specific output type (e.g.,
topocentric velocity files). The exception is global TRF IERS SINEX output,
which requires many complicated steps.
To produce a global SINEX file, follow the steps given in the final section
of this document. To produce any other Gsnoop output, follow the preliminary
steps in the next section, then follow the specific steps listed in that
output's section.
2 Preliminary steps
The steps in this section should be followed for all output in this document,
except for global TRF IERS SINEX generation, which has its own set of steps
described in its own section.
1. Type c or n to identify the Solve spoolfile from which Gsnoop should
extract information.
N solicits the full path to the spoolfile.
C only works in Gsnoop versions that are connected to a solution
archiving system (a system that tracks spoolfiles and other Solve
output files). C solicits the spoolfile's three solution
archiving identifiers: the solution owner (e.g., CM for Chopo Ma),
the solution tag (e.g., 1122a) and the version number (e.g., 3).
The identifiers should be entered under the following headers (with
the values left justified):
xx ######## nnn
cm 1122a 3
2. Type 3 to enter a 6 character identifier, gsnoop_file_id. This will
be appended to all output files from this Gsnoop session.
3. Optionally type 2 to enter an output directory other than /tmp.
3 Production of topocentric velocity (tvel) files
The following steps produce tvel files which may be input into Snoop along
with a site evolution Solve spoolfile to produce two types of site evolution
plots (total site motion and site motion with respect to a plate model).
Gsnoop steps:
Perform preliminary steps from the "Preliminary steps" section of this
document.
Type p to go to the Gsnoop "positions" menu.
Type 2 to select the XYZ and UEN position and velocity files.
Type m to start processing.
Output will be placed in:
tvel_<gsnoop_file_id>
on /tmp or on the output directory chosen in the preliminary steps.
4 Production of solution files from a global solution
A complete geodetic VLBI global solution package includes a variety of
files. (For an example, see the GSFC VLBI web site's global solutions link
at http://lupus.gsfc.nasa.gov/.) A subset of the files are generated through
Solve and Gsnoop as follows:
Solve steps:
Run a back solution of the basic global solution. The back solution
control file must:
1. use the full arc list from the basic solution
2. set NO_NET_ROTATION_POSITION and NO_NET_TRANSLATION_POSITION
constraints
3. turn off the nutation modfile (use $MAPPING NUTATION_SERIES NONE)
Gsnoop steps:
Perform preliminary steps from the "Preliminary steps" section of this
document.
Type p to go to Gsnoop's "positions" menu.
Type 2 to select the XYZ and UEN position and velocity files.
Type the letter o to go to Gsnoop's "other output" menu.
Type 1 to select site latitudes and longitudes.
Type 4 to select IERS site files.
Type 5 to select IERS source files.
Type m to start processing.
Generating the source file will cause Gsnoop to ask:
Include sessions without good source data (Y)(default)/(N)?
The proper response is n (no).
Output will be placed in the following files on /tmp or on the output
directory chosen in the preliminary steps. The equivalent web files are
listed to briefly indicate the files' general contents. For a description of
the contents, please refer to the Gsnoop user's guide.
Gsnoop output file name GSFC VLBI web file name
----------------------- ------------------------
cvel_<gsnoop_file_id> geocentric_positions-velocities
iers_sit_<gsnoop_file_id> geodetic_positions
iers_vel_<gsnoop_file_id> geodetic_velocities
sitell_<gsnoop_file_id> site_locations
iers_src_<gsnoop_file_id> sources
tvel_<gsnoop_file_id> topocentric_velocities
5 Global TRF IERS SINEX output
The following steps generate a global TRF IERS site submission in the
SINEX 1.00 format. There are several stages, covered in subsequent
sub-sections. All steps are always required, except for the steps concerning
continuous piecewise linear (cpl) output, which are only required for solutions
that treated sites as cpl.
Overview of global TRF IERS SINEX generation
--------------------------------------------------------------------
Run Solve to create a set of input solutions.
Run Gsnoop to extract the solutions' data and generate two sets of
output files containing sections of the SINEX report. The runs extract:
site and covariance information (including any cpl site information)
apriori (constraint only) information
Edit the Gsnoop output files to handle special cases.
Form the final SINEX file by concatenating the Gsnoop files with other
files containing SINEX sections that apply to all SINEX submissions.
5.1 Solve steps
1. Run a full, forward TRF solution. Then run a back solution of the TRF
solution, using all arcs used in the TRF solution. Save the spoolfile.
(Or run a full, complete TRF solution and save its spoolfile.)
The purpose of this spoolfile is to generate information for all sites in
the TRF solution. This document will call the spoolfile and its solution
"back-s" for reference.
2. Run a single arc, back solution of the TRF solution set up as follows:
Set the following control file lines:
in the $OUTPUT section:
COVARIANCES YES CGM STA
in the $CONSTRAINTS section:
STATIONS YES
(In F_solve, the $CONSTRAINTS STATIONS line should be:
STATIONS YES SIGMA 1.D1)
To generate the covariance output in F_solve, the user must also set
in the $SETUP section:
FAST_MODE NONE
SINGULARITY_CHECK NO
Save the control and covariance files from this solution. (The covariance
file is Solve scratch file CVRFxx, where xx represents the user's Solve
run initials.)
This solution ("back-c" for reference) is used to generate covariance
information.
3. Run a single arc, back solution of the TRF solution set up
as in step 2, but with one additional change. Set the following
control file line:
in the $OUTPUT section:
NORMAL_MATRIX ZERO
Save the control and covariance files from this solution.
This solution ("back-a" for reference) is used to generate apriori
information.
5.2 Steps for Gsnoop run 1 (site and covariance run)
Perform the following Gsnoop steps to generate the site and covariance
sections of the SINEX file. Steps 6, 7 and 15 are required if the input
solution treated one or more sites as continuous-piecewise linear (cpl).
Otherwise, these three steps are skipped.
1. Type 3 to enter a 6 character identifier, gsnoop_file_id (e.g., sitcov).
This will be appended to all output files from this Gsnoop session.
2. Optionally type 2 to enter an output directory other than /tmp.
3. Type v to enter the covariance file from solution back-c.
This will solicit a c for a solution catalogued under
a solution archiving system, or u to type a full path to an
uncatalogued file. Then type the solicited information.
4. Type k to enter the control file from back-c. Again Gsnoop will
solicit a (C)atalogued or (U)ncatalogued file, and so on.
5. Type c or n to enter the name of the spoolfile from solution
back-s. Use c if the spoolfile was catalogued using a solution
archiving system. Use n to enter a full path to the spoolfile.
(6. If the solution had cpl sites, type p to go to the "positions" menu.)
(7. If the solution had cpl sites, type s to enter a list of cpl sites.
Enter all cpl sites, one by one.
Enter DONE at the end to quit the S option.)
8. Type the letter o to go to the "other output" menu.
9. Type 8 to turn on Sinex non-covariance output.
10. Type 9 to turn on Sinex covariance (non-apriori) output.
11. Type m to start processing.
12. Gsnoop will ask how the sites in the SINEX SITE/ID and SOLUTION/EPOCHS
sections should be sorted.
Enter the letter l to sort the sites by longitude.
Enter a to sort them alphabetically. (A is recommended.)
13. Gsnoop will solicit an overall constraint level, if the individual
parameters have different constraint levels. Possible levels are
"fixed/tight", "significant" or "free/very loose/unconstrained".
Gsnoop will indicate numbers that represent each category. (GSFC uses
"significant".)
14. Gsnoop will solicit a comment to be inserted into the SINEX file.
(15. If the user selected the "positions" menu's s command in steps 6 and 7,
Gsnoop will display a menu for UEN plotting. Type t to terminate,
then y to confirm the termination.)
These steps will place the following SINEX sections in the following files:
section file
------- --------------------------
sinex_header_line sinex_header_<gsnoop_file_id>
INPUT/HISTORY sinex_inhist_<gsnoop_file_id>
INPUT/FILES sinex_input_files_<gsnoop_file_id>
SITE/ANTENNA sinex_site_ant_<gsnoop_file_id>
SITE/ECCENTRICITY sinex_site_ecc_<gsnoop_file_id>
SITE/RECEIVER sinex_site_recvr_<gsnoop_file_id>
SITE/ID iers_siteid_<gsnoop_file_id>
SOLUTION/EPOCHS iers_solep_<gsnoop_file_id>
SOLUTION/STATISTICS iers_solstat_<gsnoop_file_id>
SOLUTION/ESTIMATE iers_solest_<gsnoop_file_id>
SOLUTION/MATRIX_ESTIMATE L COVA iers_solmat_<gsnoop_file_id>
5.3 Steps for Gsnoop run 2 (apriori run)
Perform the following Gsnoop steps to generate the apriori sections of the
SINEX file. The steps are the same as for the first Gsnoop run, except that
the back-a control and covariance files are used instead of the back-c files,
and command 0 (SINEX covariance apriori output) is used instead of command 9,
in step 10.
1. Type 3 to enter a 6 character identifier, gsnoop_file_id (e.g., aprior)
to be appended to all output files from this Gsnoop session.
2. Optionally type 2 to enter an output directory other than /tmp.
3. Type v to enter the covariance file from back-a.
Type c to enter a solution archiving system file or u to enter a
full path to an uncatalogued file.
Type the solicited information.
4. Type k to enter the control file from back-a. Again Gsnoop will
solicit a (c)atalogued or (u)ncatalogued control file, and so on.
5. Type c or n to enter the name of the spoolfile from solution "back-s".
(Use c for a solution archiving system spoolfile or n to enter a
full path.)
(6. If the solution had cpl sites, type p to go to the "positions" menu.)
(7. If the solution had cpl sites, type s to enter a list of cpl sites.
Enter all cpl sites, one by one.
Enter DONE at the end to quit the s option.)
8. Type the letter o to go to the "other output" menu.
9. Type 8 to turn on Sinex non-covariance output.
10. Type the number 0 to turn on Sinex covariance (apriori) output.
11. Type m to start processing.
12. When asked, type l or a to sort the SITE/ID and SOLUTION/EPOCHS sites
by longitude or alphabetically (recommended).
13. When asked, enter an overall constraint level (e.g., signficant).
14. When asked, enter a comment.
(15. If the user selected the "positions" menu's s command in steps 6 and 7,
Gsnoop will display a menu for UEN plotting. Type t to terminate,
then y to confirm the termination.)
These steps will place the following SINEX sections in the following files:
section file
------- --------------------------
SOLUTION/APRIORI iers_solestap_<gsnoop_file_id>
SOLUTION/MATRIX_APRIORI L COVA iers_solmatap_<gsnoop_file_id>
5.4 Editing of Gsnoop files
Some editing of the Gsnoop files is necessary.
All files must be edited to remove the comment lines at the top (the two
lines that start with C).
Each site is allowed only one entry in iers_siteid_<gsnoop_file_id>. So the
additional lines for episodic sites (e.g., post-quake lines) must be manually
deleted. (Note that some of these lines may have ???'s in the station
description field. This simply indicates an episodic estimation of a site
which has only one antenna.dat entry because the IERS no longer assigns a new
domes number for episodic motion. These lines should be removed like any other
episodic line.)
The user may want to edit the sinex_input_files file created in the first
Gsnoop run. Currently this file just lists the input spoolfile, but the user
may want to add lines for the two input covariance files as well.
Installations other than GSFC will want to edit the following files from
the first Gsnoop run to properly identify the final SINEX file as their own or
as a joint effort between installations. The following example shows a case in
which one installation (GSFC) has created a sinex file for another installation
(USNO) using the second installation's solutions.
file suggested changes
---------------- ---------------------------------------------------
sinex_header Set CRE field (1st agency) to GSF
Set OWN field (2nd agency) to USN
sinex_inhist Same as sinex_header
Also add comments to explain:
USN = USNO, the United States Naval Observatory
GSF = GSFC, the Goddard Space Flight Center
Also add contact notes:
(Files were created for USNO from USNO data by GSFC.
For questions about the data and solution, contact:
Marshall Eubanks at tme@CygX3.usno.navy.mil
For SINEX formatting questions or problems, contact:
Karen Baver at kdb@leo.gsfc.nasa.gov)
sinex_input_files
Set Own field (field 1) to USN
Set the comment to:
USN = USNO, United States Naval Observatory
5.5 Concatenation of section files into SINEX output file
To form the final SINEX output file, the SINEX sections contained in the
Gsnoop output files must be concatenated with other files containing SINEX
sections that apply to all solutions (e.g., the SINEX ending marker).
These section files and a script to concatenate them (concat_1.0) are located
under a sub-directory of the Gsnoop source directory, iers_forms/sinex_1.0.
To produce the final SINEX output file, copy the Gsnoop files, the
general SINEX section files and concat_1.0 into a working directory.
Next edit the general section files as follows:
1. Update the Solve version number or release date in file_ref_all.
2. Installations other than GSFC will want to edit the following files
to identify themselves as the owners of their SINEX files:
a) file: file_ref_all
Change information in the following lines as appropriate:
DESCRIPTION: USNO TRF solution
CONTACT tme@CygX3.usno.navy.mil
SOFTWARE USNO Solve version 6.0004
HARDWARE HP9000S700
b) file: input_ack_all
Change and add credit lines as appropriate:
USN United States Naval Observatory VLBI Analysis Group
Finally, from the working directory, concatentate all SINEX section files
into the final SINEX file by running:
concat_1.0 <gsnoop_file_id_1> <gsnoop_file_id_2> <final_sinex_file>
where gsnoop_file_id_1 is the 6 character identifier appended to
files in the first Gsnoop run (e.g., sitcov)
gsnoop_file_id_2 is the identifier appended to files from the
second Gsnoop run (e.g., aprior) and
final_sinex_file is the name of the final SINEX output file
Questions and comments about this document should be sent to:
Karen Baver (kdb@leo.gsfc.nasa.gov)
Last update: 2000.09.06