[IGSREPORT-14] EMR analysis report

Jan Jan
Thu Nov 19 17:32:17 PST 1992


***********************************************************************
IGS Electronic Report     19-NOV-1992 17:32:17       Message Number 14
***********************************************************************
 
>From:    Jan Kouba 
Subject: EMR analysis report
         -------------------

-- 
Jan Kouba
Geodetic Survey of Canada
Ottawa, Ontario, Canada
Internet: kouba at geod.emr.ca


    from:    J. Kouba
    Subject: Geodetic Survey Division (GSD), SMRSS, EMR analysis results
             for the GPS weeks # 665,666 (Oct 04-Nov 14, 1992)

    submitted by:
    J. Kouba * , P. Tetreault, R. Ferland, F. Lahaye, M. Caissy, P. Heroux

 
    Geodetic Survey Division (GSD), Surveys Mapping & Remote Sensing Sector
    (SMRS), Energy Mines and Resources (EMR), 615 Booth Str., Ottawa, K1A 0E3
*   on secondment to the Geodetic Survey from the Geophysics Div., Geological 
    Survey of Canada (GSC), EMR, 1 Observatory cresc., Ottawa, K1A 0Y3


       
      Our analysis utilizes the JPL GIPSY/OASIS II (UNIX) software system 
and data from all (up to 7) Canadian stations augmented by up to 12, globally 
distributed, core stations (see Table 1 for the list of stations).  
      Submitted results start with GPS Week # 665 and include only days when 
AS was not invoked. The EOP for the weeks #665, #666 are resubmitted here 
with corrected DUT1 results.
      This effort is a part of a joint pilot project to test the
so called "Active Control System" in Canada and involves the  Geodetic
Survey (GSD), SMRSS, the Geophysics Division and Pacific Geoscience Centre of
the Geological Survey of Canada, all three institutions are part of the
Department of Energy, Mines & Resources (EMR).


        Table 1 : Station Summary (also indicated are day no. for which a 
particular station is not included in the daily solution, either due to being 
down or data being late) 

 station               IGS code        missing day no. of 1992

Albert Head, CAN       ALBH
Algonquin, CAN         ALGO*        
Fairbanks              FAIR*        308 
Goldstone              GOLD*        318,319 
Hartebeesthoek         HART*        
Holberg,CAN            HOLB
Kokee Park             KOKB         278 -> 319 (see remarks on EMR prod. #6)
Madrid                 MADR*        280
McMurdo                MCMU         308
Pamatai                PAMA         317
Penticton, CAN         DRAO 
Prince Albert, CAN     PRAL         315->319 (discontinued on Nov 10/92)
Santiago               SANT*        280 ,308,319
St. Johns, CAN         STJO
Canbera                TIDB*        294->297,318,319
Tromsoe                TROM*        311  (see remarks on EMR products #5)
Wettzell               WETT*        311  (     "          "     "       )
Yarragadee             YAR1*
Yellowknife, CAN       YELL*   
* indicates constrained stations, remaining are solved for (50m apriori sig.)

    SOLUTION CHARACTERISTICS

Reference frame and orientation
        Nominally ITRF91 at epoch 1992.5, realized by constraining up to 11 
above stations at the IGS coordinates (see IGS message #90): 
Orientation is realized through the above coordinates and the fixed, once a 
week, DUT1 values from the current NEOS/IERS (Bull. A).

Constants and models 
- no change, see the IGS mess. #101 for details-
Estimation strategies
- no change, see the IGS mess. #101 for details - , except for the estimation
of DUT1, stochastic orbit modeling, new weighting of L3,P3, see IGS mess. #130.
  ALGO was used as time reference.                           
  

  Table 3:  Summary of estimated parameters and apriori weighting
  (L3 and P3 undifferenced phase and smoothed pseudoranges; sigmas of 4/15mm 
100/520mm were used with an exponential el. weighting;  > 15 deg @ 5min)

 parameter         model                  apriori weighting (sigmas)
station X,Y,Z      unknown parameters     50m (fixed stations .002-.02m)
pole  x,y            "       "            3m 
Dut1                 "       "            3m ; fixed/reset once a week
trop. zen. delay     "       "            0.2m
trop. bias         random walk            .01m/sqrt(1hour)
satellite states   unknown parameters     1km; .005m/s or 1m; .0005m/s
rad. press Gx,Gz     "       "            10%
rad. press Gy        "       "            1.0 10e-9 m/s^2
initial real cy amb. "       "            300000km
station clocks     white noise            1s
satellite clocks     "     "              1ms and .1ms for Block I and II,resp.

       Aposterioi variance factors typically are between 0.5 and 1.0 (noise 
about 3mm and 200mm for phase and smoothed pseudorange observations, resp.). 
Due to data quality for the AS days the results for the corresponding 
days (with AS turned on) are not included here. 

    
   SUMMARY OF CANADIAN EMR SOLUTION 

1. Analysis centre    : Geodetic Survey Division (GSD), SMRSS , EMR
2. Solution identifier:  Submission 921118 
3. Software used      :  GIPSY/OASIS II (UNIX )
4. Definition of Terrestrial frame:
   origin             : Nominally ITRF91
   orientation        :    "        "
   reference epoch    : 1992.5
   velocity of light  : 299792458 m/s
   GM                 : 389600.4414 km^3/s^5
   gravity model      : GEMT3( 8,8) +C21+S21
   permanent tide corr: None
   ocean loading      : Pagiatakis, global model
5. Adjusted parameters:
   undifferenced phase and smoothed pseudorange data  > 15 degrees @ 5 min.
   single day (24h) arc with 6 IC and 3 rad. parameters per satellite (19 
   satellites)
   tropospheric zenith delay corr. parameter augmented with random walk stoch. 
   process
   initial phase ambiguity parameters (one for each satellite station pass or 
   initial phase)
   station positions X,Y,Z, up to 19 stations, ALGO, FAIR, GOLD, HART, SANT, 
   TROM, WETT, YELL, MADR, TIDB and YAR1 constrained.
   x, y pole position once a day (DUT1 fixed/solved; DUT1,x,y apriori sigma=3m)
   station clock biases once per each epoch/station (except for ALGO h. maser
   which provides the time reference) with sigma 1 s
   satellite clock biases once per each epoch/satellite with 1ms or .1ms sigma


Pole solution for non AS days

MJD      date          x(")      y(")   DUT1(s)   xsig     ysig    utsig   xyc

48899.5 10/04/92                 AS on
48900.5 10/05/92                 AS on
48901.5 10/06/92     .07167    .50294   .27509   .00026   .00022    0      0
48902.5 10/07/92     .07473    .50256   .27270   .00022   .00020   .00014  0
48903.5 10/08/92     .07819    .50175   .27027   .00023   .00020   .00020  0
48904.5 10/09/92     .08108    .50194   .26763   .00026   .00023   .00024  0
48905.5 10/10/92                 AS on
48906.5 10/11/92                 AS on
48907.5 10/12/92                 AS on
48908.5 10/13/92     .09046    .49696   .25832   .00024   .00020    0      0
48909.5 10/14/92     .09286    .49683   .25606   .00022   .00019   .00014  0
48900.5 10/15/92     .09555    .49630   .25408   .00022   .00019   .00020  0
48911.5 10/16/92     .09713    .49540   .25207   .00022   .00020   .00024  0
48912.5 10/17/92                 AS on
48913.5 10/18/92                 AS on
48914.5 10/19/92                 AS on
48915.5 10/20/92     .10662    .49358   .24355   .00024   .00020    0      0
48916.5 10/21/92     .10730    .49159   .24042   .00028   .00023   .00014  0
48917.5 10/22/92     .11000    .48936   .23702   .00027   .00022   .00020  0
48918.5 10/23/92     .11328    .48819   .23345   .00024   .00020   .00024  0
48919.5 10/24/92                 AS on
48920.5 10/25/92                 AS on
48921.5 10/26/92                 AS on
48922.5 10/27/92     .12075    .48671   .22139   .00021   .00017    0      0
48923.5 10/28/92     .12255    .48555   .21904   .00021   .00017   .00014  0
48924.5 10/29/92     .12370    .48429   .21700   .00021   .00017   .00020  0
48925.5 10/30/92     .12495    .48288   .21485   .00021   .00017   .00024  0
48926.5 10/31/92                 AS on
48927.5 11/01/92                 AS on
48928.5 11/02/92                 AS on
48929.5 11/03/92     .13410    .47630   .20678   .00024   .00020    0      0
48930.5 11/04/92     .13657    .47631   .20433   .00022   .00019   .00014  0
48931.5 11/05/92     .13767    .47511   .20164   .00022   .00019   .00020  0
48932.5 11/06/92     .13937    .47317   .19904   .00022   .00020   .00024  0
48933.5 11/07/92     .14077    .47220   .19637   .00022   .00018   .00028  0
48934.5 11/08/92     .14196    .47006   .19386   .00022   .00019   .00031  0
48935.5 11/09/92     .14403    .46865   .19140   .00022   .00018   .00034  0
48936.5 11/10/92     .14540    .46694   .18850   .00021   .00017    0      0
48937.5 11/11/92     .14694    .46560   .18624   .00021   .00017   .00014  0
48938.5 11/12/92     .14962    .46350   .18414   .00022   .00018   .00020  0
48939.5 11/13/92     .15121    .46181   .18200   .00025   .00020   .00024  0
48940.5 11/14/92     .15248    .46089   .17971   .00026   .00021   .00028  0

Products generated:

The following products are generated and uploaded into CDDIS @ NASA/GSFC

EMR06677.EPH        GPS ephemeris for week # 667 in sp3 format
EMR06687.EPH        GPS ephemeris for week # 668 in sp3 format
EMR06697.EPH        GPS ephemeris for week # 669 in sp3 format
EMR06707.EPH        GPS ephemeris for week # 670 in sp3 format
EMR06707.ERP        pole x,y and DUT1 solutions for weeks# 665 - 670
 
EMR06707.SUM        name of this text file



remarks on EMR's products:
1. only days with no AS included
2. ALGO hydrogen maser used as a time reference. In the sp3 EPH file sat.
   epochs with no reference (e.g. ALGO not observing a satellite) have a value 
   of 99999. microseconds. 
   ALGO clock error & drift were about -100ns(48900) &-19ns/day w.r.t. GPS time
   during this period.
4. DUT1 var-covariance matrix is fully populated, consequence of a random walk
   orbit model. i.e. cov(i,j)= var(i,i). Differences or LODs are uncorrelated
   with var= var(j,j)-var(i,i); j=i+1. DUT1 solutions may or may not be real,
   they may reflect orbit (longitude) mismodeling common to all satellites.
5. no prn25 data on Oct 23/92 ; TROM, WETT data files on CDDIS for Nov 6/92
   contains data for Nov 8!
6. KOKB starts on Nov 9/92;not used due to a poor (pseudorange) data quality.



More information about the IGSREPORT mailing list