ILRS Implementation of SINEX (1999)

The SINEX (Solution INdependent EXchange) format is used in the geodetic community to store and transfer solutions of various parameters derived in various types of analysis. The official SINEX format can be obtained from:

https://www.iers.org/IERS/EN/Organization/AnalysisCoordinator/SinexFormat/sinex.html.

This document gives some more guidelines for the use of the SINEX format, in particular related to the ILRS pilot project "positioning and earth orientation". Most of the elements are as defined in the standard format (suggestions are given only), but the SINEX file as it is used by ILRS also contains some new elements, (i) which are of importance if you solve for biases, and (ii) which specify the analysis in more detail.

In general, one can say that the parameter solution is given by the equation

dx = [ AT W A + V0(inv) ] (inv) times [ AT W R ]

In summary, the block SOLUTION/ESTIMATE contains the solution dx, the block SOLUTION/MATRIX_ESTIMATE contains the a posteriori covariance matrix (i.e. vf [ AT W A + V0(inv) ] (inv) ), and the block SOLUTION/MATRIX_APRIORI contains the a priori covariance matrix [ vf V0 ]. Further details are given below.

HEADER
- mandatory
- constraint code "2", must be determined from data (not hard-coded)
- entries for Solution Contents separated by exactly 1 blank
- all numbers preceded by leading zeroes (if relevant)
- data: epoch of first and last observation, respectively
- for ILRS only: if bias parameters (range, time, troposphere, scale)
  are solved for and included in this SINEX file, specify so by giving
  a "B" in the Solution Contents (in theory, the format has to be
  extended to 6(1X,A1) ).

FILE/REFERENCE
- mandatory

FILE/COMMENT
- mandatory
- every line must start with a "*" symbol
- it must include a general description of the analysis, plus a table
  which specifies elements of this analysis (including the headers):
* general:		    (options)	      (example)
* - analysis center		  specify	  DEOS
* - technique			SLR/LLR		SLR
* - software		       specify	       GEODYN II, SOLVE
* dynamic model:
* - gravity field
*   + static component		     specify	     GRIM5-S1
*   + maximum degree and order		 specify	 20 x 20
*   + time dependent component		 specify	 C(2,0)-C(4,0),C(2,1),S(2,1)
* - GM [m**3/s**2]		 specify	 398600.4415D9
* - solid earth tides		    specify	    Wahr model
* - ocean tides			  specify	  Ray model
* - atmospheric tides		    specify	    applied
* - direct solar radiation	     specify	     applied; CR fixed at 1.14
* - albedo		     specify	     not applied
* - earth shine			  specify	  not applied
* - thermal forces		 specify	 not applied
* - dynamic polar motion	   yes/no	  applied
* - empirical accelerations	      specify	      AT constant; R 1cpr
* - third body attraction	    specify	    Sun, Moon, Venus, Mars,
*			      Jupiter, Saturn
* - relativistic effects	   specify	   not applied
* reference frame:
* - station coordinates
*   + a priori values		    specify	    ITRF2000
*   + estimated			  specify	  all
*   + reference point		    specify	    o.c. for stations with zero
*			      ecc; SLR monument for others
*   + permanent tide correction		  yes/no	 yes
* - station velocities
*   + a priori values		    specify	    ITRF2000
*   + estimated			  specify	  no
* - earth orientation parameters
*   + a priori values		    specify	    IERS C04
*   + frequency [days]		     specify	     1
*   + estimated			  specify	  yes
* - precession
*   + model		      specify	      IAU 1976
*   + corrections		specify		not applied
* - nutation
*   + model		      specify	      IAU 1980
*   + corrections		specify		not applied
* - station deformations:
*   + tidal uplift		 specify	 Love model
*			      (h2=0.609, l2=0.0852)
*   + ocean loading		  specify	  IERS 1996 Conv + extensions
*   + atmospheric pressure loading     specify	       linear regression;
*			      Van Dam coefficients
* - geocenter
*   + motion		       specify	       not explicitly modeled
*			      or corrected
*   + tidal frequencies		      specify	      no
* - pole tide			specify		applied
* - definition of origin	   specify	   defined by station
*			      coordinates solution
* - definition of orientation		specify		ITRF200 a priori
*			      (constrained at 1 m)
* measurement model:
* - satellite used		 specify	  LAGEOS-1 + LAGEOS-2
* - wavelength of observations [nm]    specify		532, except 7810 (423)
*			       and 7939 (694)
* - biases:
*   + range biases		 specify	  estimated
*   + time bias			  specify	   no
*   + tropospheric bias		      specify	       no
* - data weighting		 specify	  r.s.s. of single-shot
*			       precision and overall
*			       model accuracy (2 cm)
* - center of mass [m]		     specify	      0.251
* - model for tropospheric delay       specify		Marini-Murray
*   (including mapping function)
* - minimum elevation [deg]	      specify	       20
* - relativity scale		   SSB/LET/LE	  LET
* - speed of light [m/s]	   specify	    299792458
* - relativistic effects	   specify	    not applied
* integration:
* - arc length			 specify	  7 days
* - integrator			 specify	  Cowell 11th order
* - step-size [s]		specify		 100
* - estimated parameters	   specify	    station coordinates at epoch;
*				EOPs at 1-day intervals;
*				satellite state-vector;
*				satellite accelerations;
*				1 range bias per station
* - constraints			  specify	    1 m on station coordinates
*				and EOPs;
*				5x single shot precision
*				on range bias

INPUT/HISTORY
- recommended

INPUT/FILES
- optional

INPUT/ACKNOWLEDGEMENT
- optional

SITE/ID
- mandatory
- point code for SINEX internal referencing only -> use "A"
  (unless more than 1 solutions for a station)
- (if relevant): minus sign of lat/lon in "degrees" component,
  not repeated in "minutes" and "seconds" components

SITE/DATA
- optional
- times must refer to observation epochs

SITE/RECEIVER
- not for ILRS

SITE/ANTENNA
- not for ILRS

SITE/GPS_PHASE_CENTER
- not for ILRS

SITE/ECCENTRICITY
- mandatory
- eccentricities must be subtracted if different from zero
- 7824 and 7810: zero (i.e. refer to o.c. of "new" stations)
- 7838: zero (i.e. no eccentricity to be subtracted)
- to be generated by software/data used; not copy of standard
  input file
- the new official ILRS eccentricity files in SINEX are:

     ftp://cddisa.gsfc.nasa.gov/pub/slrocc/ecc_une.snx
     ftp://cddisa.gsfc.nasa.gov/pub/slrocc/ecc_xyz.snx


SOLUTION/EPOCHS
- mandatory
- 1st epoch = epoch 1st observation
- 2nd epoch = epoch last observation
- 3rd epoch = centroid (weighted mean)
- SOLUTION/EPOCHS, not SOLUTION/EPOCH

BIAS/EPOCHS
- for ILRS only
- mandatory when biases are solved for
- format identical to format of SOLUTION/EPOCHS block, but:
- the Site Code gives the station id
- the Point Code gives the satellite id (e.g. L1, L2, LC, E1, E2, EC
  for LAGEOS-1, 2, LAGEOS combined, ETALON-1, 2, and ETALON combined,
  respectively)
- the Solution ID gives the sequential number of the bias for this
  particular station; if just 1 bias is solved for a particular
  station, this parameter remains "1".
- the Observation Code specifies the type of bias (R, T, S or Z for
  range bias, timing bias, scale bias and troposphere bias at zenith,
  respectively)
- 1st epoch = epoch 1st observation
- 2nd epoch = epoch last observation
- 3rd epoch = centroid (weighted mean)

SOLUTION/STATISTICS
- mandatory
- include #observations, #unknowns (all parameters, including satellite
  and other parameters not in this SINEX file), sampling interval (120 s
  for LAGEOS), squared sum of residuals, degrees of freedom and variance factor vf.

SOLUTION/ESTIMATE
- mandatory
- standard deviations, for ILRS pilot project:
  must reflect the 1 meter a priori uncertainty on station coordinates
  and (the equivalent for) earth orientation parameters.
  Apply a scaling by the square root of the variance factor vf (which should be
  around 1 already if you used realistic (i.e. around 1 cm) values for the
  weights of the SLR observations and 1 meter a priori sigma values for the
  station coordinates; if not, this scaling factor may be in the order
  of 0.01).
- ILRS/AWG pilot project:
   + refer all positions to 1 single epoch
   + loosely constrained (i.e. indexed "2", determined by
     software, not hard-coded)
- new for ILRS:
  the Parameter Type may specify a bias parameter: RBIAS, TBIAS, SBIAS
  or ZBIAS (flush left), for range bias, time bias, scale bias and
  troposphere bias at zenith, respectively. In this case, the Site
  Code must give the station id, the Point Code must give the
  satellite id (e.g. L1, L2, LC, E1, E2, EC for LAGEOS-1, 2, LAGEOS combined,
  ETALON-1, 2, and ETALON combined, respectively), and
  the Solution ID must give the sequential number of the bias for
  this particular station. The Epoch must refer to the beginning of
  the pass (i.e. matching what is given in the BIAS/EPOCHS block).
  The units for the bias parameters are [m], [millisec],
  [ppb] and [m], respectively. A range bias of +1 meter means that
  the observation(s) is(are) 1 meter long; a time bias of +1 sec
  means that the observation(s) is(are) 1 second late.

SOLUTION/APRIORI
- mandatory
- standard deviations: see the comments in SOLUTION/ESTIMATE (and
  apply the same scaling factor sqr(vf)). The result will be that the
  sigma included in this part of the SINEX file will have a value of about
  1 meter, as requested by the project description.
- ILRS/AWG pilot project:
   + refer all positions to 1 single epoch
   + loosely constrained (i.e. indexed "2", determined by
     software, not hard-coded)
- see the entry "new for ILRS" in SOLUTION/ESTIMATE

SOLUTION/MATRIX_ESTIMATE
- mandatory
- no preference for COVA, CORR or INFO
- must be consistent with SOLUTION/ESTIMATE
- standard deviations: see the comments in SOLUTION/ESTIMATE (and
  apply the same scaling factor vf, but now squared)
- ILRS/AWG pilot project:
   + refer all positions to 1 single epoch
   + loosely constrained (i.e. indexed "2", determined by
     software, not hard-coded)
- see the entry "new for ILRS" in SOLUTION/ESTIMATE

SOLUTION/MATRIX_APRIORI
- mandatory
- no preference for COVA, CORR or INFO
- must be consistent with SOLUTION/ESTIMATE
- standard deviations: see the comments in SOLUTION/ESTIMATE (and
  apply the same scaling factor vf, but now squared)
- ILRS/AWG pilot project:
   + refer all positions to 1 single epoch
   + loosely constrained (i.e. indexed "2", determined by
     software, not hard-coded)
- see the entry "new for ILRS" in SOLUTION/ESTIMATE

FOOTER
- mandatory

in general: all epoch values to be rounded of to nearest integer