TOC | Last Section | Next Section | INDEX |
This will be the longest tutorial of the group, not because every step must be done every time run_survey is used, but because every step should be done at least once in processing a project or block of data. Recall that run_survey was developed for batch processing of large or long term projects. Once the processing is setup and has run successfully once, many tasks are guaranteed to be correct thereafter. Broadly speaking, task breakdown into:
This is, in essence a review of the pages.sum file with a tutorial on using the xplot program. An editor or some program for viewing an ASCII file is required.
All information for this initial review is given in the pages.sum file. Most items in this file are "tagged" with a unique label to facilitate using a search command in an editor or the grep command. In this tutorial, it is assumed the file has be opened with an editor or viewer capable of moving within the file. In this tutorial segments of a sample pages.sum are imbedded and links to the appropriate sections of the sample pages.sum are provided.
At the end of the last tutorial, the current directory was ~/tutorial
The first section of this file is effectively a summary of the summary.
In particular, note the "OVERALL RMS". This is the RMS of the post-fit residuals so the smaller the better. This number is dependent upon the quality and quantity of the data generated by the sites, and the length of the baselines. Typically good values are 0.010 to 0.015. The overall RMS of this run is at the high end of this range but no cause for alarm. Larger values indicate manual editting may be required. This should be done for every solution.
If any singularities, free parameters which could not be estimated from the data, occurred during the solution, these would be listed immediately below the "OVERALL RMS". None were found in this solution. This should be done for every solution.
Before leaving the top of the pages.sum file, one should scan the other information provided verifying that the observable and modelling are as expected. For baseline solutions, the only entries of concern are the
Immediate below the initial summary, the satellites, ephemeris and stations used in the solution are listed.
page5 VERSION: 9702.08
DATE AND TIME OF THIS SOLUTION: 97/02/09 11:54:55
ANALYST'S NAME: Steve Hilla
START TIME: 1996/12/25 00:00:00 (DAY-OF-YEAR: 360.000000)
STOP TIME: 1996/12/25 23:59:30 (DAY-OF-YEAR: 360.999653)
PROGRAM OPERATION: FULL RUN
FREQUENCY: ION-FREE (L3)
OBSERVATION INTERVAL: 30 (SECONDS)
ELEVATION CUTOFF: 15 (DEGREES)
TROPO INTERVAL: 7200 (SECONDS) [PIECE-WISE LINEAR PARAMETERIZATION]
X AND Y POLE INTERVAL: NONE
UT1 INTERVAL: NONE
DD CORRELATIONS: OFF
OVERALL RMS = 0.0151 (METERS)
------------------------------------------------------------------------------
NUMBER OF SATELLITES: 25 (PRN NUMBERS ARE LISTED BELOW)
Now jump to / search for "INPUT FILES". In particular, note the files listed as "STATION INFO" and "ANTENNA PATTERN" files. Verify that the list is complete. In this tutorial, we retrieved the IGS and CORS site information files and the antenna pattern correction files as shown.
DELETION VALUE = 0.100 M RHC ROTATION CORRECTION= ON RAD PRESS SCALE FACTORS= 1 SOLID EARTH TIDE MODEL = IMPROVED TROPO MODEL = NMF TROPO PARAMETERS TSF = YES , RATE = YES MAX TROPO MODEL TIME = 7200.000 SEC PM = NO , PM PIECE-WISE = NO LOD= NO , LOD PIECE-WISE= NO MIN EOP MODEL TIME = 86400.000 SEC INPUT FILES ----------- STATION INFO = "../files/site_info.cors" STATION INFO = "../files/site_info.igs" SATELLITE INFO = " " DESCRIPTIVE TEXT = " " POLE POSITIONS = " " UT1 - UTC VALUES = " " BROADCAST ORBIT = " " ANTENNA PATTERN = "../files/ant_info.002" (ant_info.002)
Next we jump to / search for "POST-FIT RMS". This section gives the pre-fit (a priori) and post-fit RMS's, the total number of observations used in the solution and the total number omitted. The overall, post-fit RMS was discussed above. Here note the number of number of used and omitted observations. Typically, for baselines shorter than a 1,000 km, one should have more than 10,000 observation per baseline per day. With this data set, two baselines were defined and the total number of observations is well above this estimate. The number of omitted data should always be some small percentage of the total. These should be checked every solution.
Below these overall run statistics, the RMS, observation count and omitted percentage are broken out by baseline verses satellite. These matrices of statistics often prove valuable for quickly localizing problems. For the sake of brevity, only part of one table is shown.
07 1996/12/25 01:10:30 - 1996/12/25 04:37:00 24 1996/12/25 04:37:00 - 1996/12/25 10:01:00 17 1996/12/25 10:01:00 - 1996/12/25 13:12:30 01 1996/12/25 13:12:30 - 1996/12/25 17:46:00 29 1996/12/25 17:46:00 - 1996/12/25 21:02:30 31 1996/12/25 21:02:30 - 1996/12/25 23:59:30 ============================================================================== OVERALL STATISTICS: A PRIORI RMS: 0.1270 M FROM 27294 OBSERVATIONS; 699 OMITTED POST-FIT RMS: 0.0151 M FROM 27294 OBSERVATIONS; 699 OMITTED STD. ERROR: 0.1510 M FROM 27294 OBSERVATIONS; 699 OMITTED WEIGHTED MEAN TIME = 96/12/25 11:48:18.19 ( 50442.49187723) FREE PARAMETERS = 258; PORTION OF MATRIX USED = 65 ------------------------------------------------------------------------------ POST-FIT RMS BY SATELLITE VS. BASELINE BLOCK= 1 OVERALL 01 02 03 04 05 06 07 09 ------------------------------------------------------------------------------ chl1-gode| 0.017 0.016 0.016 0.016 0.015 0.017 0.016 0.018 0.018 sol1-gode| 0.012 0.016 0.011 0.011 0.013 0.016 0.011 0.014 0.011 RMS BY PRN 0.016 0.014 0.014 0.014 0.017 0.014 0.016 0.015
Now jump to / search for "STATION INFORMATION". Each station used in the processing has a block in this section. Each station subsection is separated by a line containing on pluses. Stations whose information is taken from site information files should be detailed. In particular, note the second line of this subsection, labelled "INFO FROM". This is the source file for the a priori information presented and used in the processing. If a file does not appear in a site information file, this line states "INFO FROM DATABASE". In this case, the site information ultimately comes from the POM file and will be quite limited. This is the case for SOL1. The site information is broken in pieces:
If this processing is part of a larger or long term project, the estimated coordinate adjustments themselves will serve as a valuable quality test because the coordinates of all sites will be well determined.
These coordinate estimates are the purpose of this processing and so should be check every time.
STATION INFORMATION (INPUT AND OUTPUT VALUES):
STATION NAME: gode a (Greenbelt, MD, USA (Godd)
INFO FROM "../files/site_info.igs"
EXCEPT PHASE CENTER OFFSETS
WHICH ARE FROM THE ANTENNA PATTERN FILE
RECEIVER ROGUE SNR-8000 F/W=UNKNOWN S/N=129
ANTENNA (JPL D/M+CRT) DORNE MARGOLIN T S/N=129
MONUMENT: 40451M123
PLATE: noam
MET DATA WAS MODELLED USING STATION HEIGHT, STATION LATITUDE, DATE, AND TIME
INPUT VALUES VALID AT 1996/360 11:48 = 1996/360.49188 = 1996.9822
REFERENCE FRAME: ITRF94
1130773.9006 -4831253.5661 3994200.3883 gode MON @ 1996.0000 (M)
( 0.0087) ( 0.0075) ( 0.0087) gode MON SIGMAS (M)
-0.0148 -0.0017 0.0001 gode VEL (M/YR)
( 0.0017) ( 0.0015) ( 0.0017) gode VEL SIGMAS (M/YR)
N= 0.0000 E= 0.0000 U= 0.0614 40451M123 TO ARP (M)
N= 0.0000 E= 0.0000 U= 0.1100 ARP TO L1 PHASE CENTER (M)
N= 0.0000 E= 0.0000 U= 0.1280 ARP TO L2 PHASE CENTER (M)
A PRIORI COORDINATES AT 1996/360 11:48 = 1996/360.49188 = 1996.9822
1130773.9006 -4831253.5661 3994200.3883 gode MON @ 1996.0000
-0.0146 -0.0017 0.0001 VEL TIMES 0.9843 YRS
0.0109 -0.0464 0.0387 40451M123 TO ARP
---------------------------------------------
1130773.8969 -4831253.6142 3994200.4271 gode ARP @ 1996.9822
0.0195 -0.0832 0.0693 ARP TO L1 PHASE CENTER
---------------------------------------------
1130773.9164 -4831253.6974 3994200.4963 gode L1 PHS CEN @ 1996.9822
ADJUSTED COORDINATES AT 1996/360 11:48 = 1996/360.49188 = 1996.9822
(USING DATA FROM 96/360 00:00:00.00 TO 96/360 23:59:30.00)
1130773.9164 -4831253.6974 3994200.4963 gode L1 PHS CEN @ 1996.9822
0.0000 0.0000 -0.0001 + XYZ ADJUSTMENTS
--------------------------------------------
1130773.9164 -4831253.6975 3994200.4962 NEW gode L1 PHS CEN @ 1996.9822
-0.0195 0.0832 -0.0693 - L1 PHASE CENTER TO ARP
--------------------------------------------
1130773.8969 -4831253.6142 3994200.4270 NEW gode ARP @ 1996.9822
-0.0109 0.0464 -0.0387 - ARP TO 40451M123
--------------------------------------------
1130773.8860 -4831253.5678 3994200.3883 NEW gode MON @ 1996.9822
( 0.0016) ( 0.0014) ( 0.0016) NEW XYZ SIGMAS
1130773.8858 -4831253.5678 3994200.3883 NEW gode MON @ 1997.0000
STATION NAME: gode a page5 VERSION: 9702.08 RUN DATE: 97/02/09 11:55
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
X Y Z DESCRIPTION
------------------------------------------------------------------------------
1130773.9164 -4831253.6975 3994200.4962 NEW gode L1 PHS CEN @ 1996.9822
1130773.8969 -4831253.6142 3994200.4270 NEW gode ARP @ 1996.9822
1130773.8860 -4831253.5678 3994200.3883 NEW gode MON @ 1996.9822
1130773.8858 -4831253.5678 3994200.3883 NEW gode MON @ 1997.0000
0.0000 0.0000 -0.0001 XYZ ADJ (M) FOR gode
( 0.0016) ( 0.0014) ( 0.0016) XYZ ADJ SIGMAS (M) FOR gode
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
LAT LON HGT COMPUTED USING GRS-80
ELLIPSOID'S SEMI-MAJOR AXIS
AND FLATTENING
------------------------------------------------------------------------------
39 1 18.21789 283 10 23.41716 14.6745 NEW gode L1 PHS CEN @ 1996.9822
76 49 36.58284 WEST LONGITUDE
39 1 18.21789 283 10 23.41716 14.5645 NEW gode ARP @ 1996.9822
76 49 36.58284 WEST LONGITUDE
39 1 18.21789 283 10 23.41716 14.5031 NEW gode MON @ 1996.9822
76 49 36.58284 WEST LONGITUDE
39 1 18.21789 283 10 23.41715 14.5031 NEW gode MON @ 1997.0000
76 49 36.58285 WEST LONGITUDE
-0.0001 0.0000 0.0000 NEU ADJ (M) FOR gode
( 0.0002) ( 0.0012) ( 0.0023) NEU ADJ SIGMAS (M) FOR gode
Two windows should appear. One simply allows the user to exit the xplot program. The other enables selection of the specific plot. A plot file can actually contain several related plots. One or several of the could be selected for display. In this case, the only available selection is "ION FREE chl1-gode (aaab01)". Highlight this selection and "Plot". The plot window should now appear. The first thing to note is the scale listed in the upper, left of the window. This is a multiplier for the units on the vertical axis to give the units listed in the axis label. In this example, the scale implies values of hundredths of a meter, centimeters, which is acceptable. The plotted data should look like random noise centered around zero.
The output from the diff should be a few lines only showing differences in the user and machine, and different run times/dates, e.g.
The list program can strip the estimated coordinate adjustments or modified monument coordinates from the pages.sum file and presents this information, with some statistics, in a form suitable for plotting using any XY plotting program.
The xplot program is a critical tool for manual processing. xplot allows one to zoom into the area under the cursor using the center an left mouse buttons respectively, or to zoom in more generally by boxing an area while holding the left mouse button. In either case, once can zoom out using the right mouse button. In addition, there are buttons in the window for selecting specific satellites to plot or displaying an ASCII list of plotted values. From this list edit instructions can be generated using the mouse.
TOC | Last Section | Next Section | INDEX |
bl_7.html
March 4, 1999
Steve Hilla