From: Henno.Boomkamp@esa.int To: igsleo@igscb.jpl.nasa.gov Message-ID: <41256A50.005D1381.00@esoc.esa.de> Date: Fri, 18 May 2001 19:02:30 +0200 Subject: [IGSLEO-8] Analysis Centre coordination for LEO PP Sender: owner-igsleo Precedence: bulk ****************************************************************************** IGS LEO Mail 18 May 10:16:34 PDT 2001 Message Number 8 ****************************************************************************** Author: Henno Boomkamp Dear Colleagues, This mail aims to answer some items listed in IGSLEO mail No. 4 (actions raised during the Potsdam meeting) and No. 5 on data formats. The comments below show the present status of the Coordination and Analysis tasks assigned to the ESOC Associate Analysis Centre Coordinator. A few concrete suggestions are included in these comments. These are merely intended to provide a starting point, under the assumption that an incomplete standard is better than no standard at all. It is expected that many of the points that are now left open can be made more concrete after the first CHAMP data processing results become available. ----------------------------------------------------------------------- (1) Identify responsible technical point of contact ----------------------------------------------------------------------- As from May 2001, LEO Pilot Project AAC and AACC activities at ESOC are handled by Henno Boomkamp, with support from Carlos Garcia and John Dow. --------------------------------------------------- (2) Develop processing standards --------------------------------------------------- The natural way for developing processing standards is to study existing practice at the various Analysis Centres, and only agree upon a common standard in cases of identified incompatibilities. For this purpose it would be useful if the ESOC AACC could obtain some documentation from those Analysis Centres that are already prepared for routine data processing, describing their present solution set-up and processing standards. Of interest are in particular details like: - Summary of dynamic models (gravity field, drag modelling) - Use of Earth rotation parameters, reference frame realisation - Which observables are used (differenced or undifferenced GPS, phase data, other tracking data like SLR, DORIS, ...) - Handling of LEO satellite clock, - Handling of LEO accelerometer data, if available - LEO orbits estimated with respect to fixed IGS GPS orbits/clocks, or simultaneous solution for LEO and GPS The following requests are therefore made at this point : (a) from GFZ Potsdam (C. Reigber) : to forward documentation on CHAMP processing standards to ESOC AACC (b) from JPL and/or CNES : to forward documentation on planned JASON data processing standards to ESOC AACC (c) from any Analysis Centre that plans on processing the CHAMP data that will be arriving shortly : a summary on their standards and/or solution set-up by e-mail, and if possible some results. -------------------------------------------------------------- (3) Exchange format for orbit comparisons -------------------------------------------------------------- The SP3 format seems to be adequate. Differences in vector intervals, arc lengths and starting points can be overcome by careful N-point interpolation, at least during the initial stages of the PP when precision will not yet be at cm level. It is suggested to maintain substantial overlaps between successive solution arcs to enable elimination of inaccuracies around the start and end of the arcs. A matter to be settled as soon as possible would be the assignment of ID numbers to the various LEO satellites. It has been suggested to modify the SP3 format to include the ID code Lxx (c.f. Gxx for GPS, Rxx for GLONASS). In that case, it will be helpful to have agreement on the numbers 'xx' to be used for each LEO satellite. At ESOC, we are not aware of any existing ID lists of this type. Lacking further information, the simple proposal is made here to assign the numbers in order of launch date, starting with L01 for TOPEX/Poseidon, etc. With that system, the only remaining discussion is to decide which LEO satellites are to be included in the numbering system. At least all LEO satellites with GPS data should be covered, so at this point the following table is suggested : ID Satellite Launch ------------------------------------------------------- L01 TOPEX/Poseidon 08/1992 L02 GPS-MET 12/1994 L03 GFO 02/1998 L04 Orsted 01/1999 L05 SUNSAT 02/1999 L06 CHAMP 07/2000 L07 SAC-C 11/2000 L08 JASON-1 2001 L09 GRACE (*) 2001 L10 GRACE (*) 2001 L11 GOCE 2004 (*) order of the GRACE satellites to be determined in the future Any further comments on this point are welcome. In particular, other satellites of interest that may have been overlooked in the list above should be reported as soon as possible, so that a more definitive proposal can be prepared as to which satellite corresponds to which Lxx number. -------------------------------------- (4) LEO AAC data sheets -------------------------------------- The proposal is to start with the standard IGS data sheet, and see in practice which additional information would be useful. This point is closely related to the action (2) on analysing the various processing standards and modelling details, and can be coordinated by the ESOC AACC during the course of the Pilot Project. -------------------------------------------------- (5) Arc length and parametrisation -------------------------------------------------- In this early stage the natural approach is to let every Analysis Centre determine a solution set-up that seems to provide the best results in their processing environment, independent of other AC's. After analysis of the different solution set-ups and processing standards by the ESOC AACC (see point 2) it may be useful to standardise certain aspects of the solutions, in particular modelling aspects that affect interchange of data products such as reference frame realisations. ------------------------------------------------------------ (6) Orbit comparisons and test definitions ------------------------------------------------------------ For orbit comparisons, the following methods are proposed: (a) In case that very few solutions are available, pair-wise comparisons of SP3 files can be performed to provide an idea of their relative accuracy (b) Once that sufficient (stable) orbit solutions are available, a combination solution can be produced in the style of the IGS GPS orbits. Typically this would be a staged process, with a first unweighted solution to estimate standard deviations of each input orbit, followed by a second, weighted solution. To avoid reference system problems as much as possible, the following two recommendations are made here : (a) to apply a 7-parameter transformation to the LEO orbit in order to bring it in line with the IGS / GPS reference frame (this should be relevant only in the case of combined solutions GPS/LEO) (b) to clearly indicate if the GPS orbits were solved together with the LEO orbit, or not (this may be one of the issues to be added to the LEO AAC data sheets) -------------------------------------------------------- (7) Specify data period for processing -------------------------------------------------------- For the time being, little data is available for routine processing but this will change in the short term. At present no specific period is indicated and all available data is analysed to the best possible extent. ------------------------------------------------------------------------ (8) LEO receiver clock solution and comparisons ------------------------------------------------------------------------ This matter forms part of the analysis described under (2). For the time being no standard will be set, but the LEO Analysis Centres are requested to inform the ESOC AACC about their models for the LEO clock, in order to assess the need for standardisation. ------------------------------------------------ (9) SLR for external validation ------------------------------------------------ See point (7) - no specific processing has been identified yet. Centres with the relevant capabilities are encouraged to process CHAMP laser data, as an independent source of validation. As will be clear, these comments leave almost total freedom to the Analysis Centres to set up their preferred processing environment. The short-term intention of the ESOC AACC is to collect as much concrete information as possible regarding the various solution standards and regarding the first CHAMP processing results. For any feedback or additional inquiries, you can now contact me directly, or send a mail to the IGS-LEO mailing list. Best regards, Henno Boomkamp