-->

Call for Participation "Positioning & Earth Orientation"

Version 2.2
R. Noomen, G.M. Appleby, P.J. Shelus
May 21, 2003

Summary

The ILRS officially solicits the operational production of daily Earth orientation parameters (x/y-pole and LOD) and 28-day global station coordinates, based on SLR measurements on the LAGEOS and Etalon satellites. Two different types of contributions are requested: (i) the direct estimation of the parameters mentioned above, and/or (ii) the quality control and merger of such individual solutions. Irrespective of the actual form of their participation, contributors will have to satisfy a number of requirements as specified in this document. In principle, the ILRS will accept any contribution that satisfies the technical and operational requirements. As for the quality control and combination element, the ILRS will designate one prime combination center and other(s) to act as back-up(s).

Version 2.2 of this Call for Participation differs from the previous version (2.1) in the sense that the frequency and timing of the various analysis activities in each week has been changed. Contributors who already responded to the version 2.1 of this Call for Participation, in December 2002, do not need to repeat their notification of interest.

1. Introduction

The International Laser Ranging Service (ILRS) coordinates the development of the Satellite Laser Ranging (SLR) and Lunar Laser Ranging (LLR) technique, the acquisition, distribution and analysis of the resulting measurements on target satellites, as well as that of the analysis products, amongst others. To this aim, the ILRS embodies a number of elements, such as tracking stations, analysis groups, working groups dealing with specific aspects of the measurements, archiving facilities, etc. Being one of the space geodetic observation techniques, the laser ranging technique is capable of observing and monitoring a wide range of aspects of "system Earth" and beyond, some of them with a unique reliability. Traditional products relate to station coordinates and velocities, and the rotation of the Earth, but other valuable products (geocenter, absolute scale, orbit (calibration), etc.) have also been developed and are being used by external "clients".

The ILRS is one of the nine services of the International Association of Geodesy (IAG), which also includes the International Earth Rotation and Reference Systems Service (IERS; cf. http://www.iers.org). As such, the ILRS is responsible for the generation of operational and scientific, quality-controlled products, for inclusion in various official IERS products. The ILRS, its Analysis Working Group (AWG) in particular, is in the process of developing such a unique and quality controlled official ILRS product (for various parameters). This Call for Participation deals with two products, namely (aspects of) the rotation of the Earth and the coordinates of the global network of SLR stations, both to be determined in near-real time. Contributions are solicited for (i) the direct computation of the Earth rotation and station coordinates, and/or (ii) the quality control and combination of various solutions into one official ILRS product. In the case that a particular analysis group is interested in both tasks, it should submit two independent proposals. Both topics are covered by the AWG pilot project "positioning and earth orientation", which has been in steady development for a number of years now.

This document will describe the two possible contributions. For each one, the input, product, formats and requirements are presented. Some general information on the overall procedure is also given.

The IERS Bulletin A is one of the customers for ILRS products. These bulletins are released once per week (on Thursday), which then drives the operational scheme of activities described here.

2. Data analysis

This chapter describes the various aspects of contributions for the data analysis itself, i.e. the determination of EOPs and station coordinates from SLR observations on satellite targets as specified hereafter.

2.1. Input

SLR observations on geodetic satellites are required to generate the individual EOP and coordinates solutions. As a minimum, this will include the range measurements taken on the cannonball satellites LAGEOS-1 and LAGEOS-2. Contributing analysis groups are strongly encouraged to also include data taken on the satellites Etalon-1 and Etalon-2, although this is not strictly required.

The measurements are available in either daily or hourly files at the CDDIS and EDC archiving facilities. The files and their naming convention will be explained below. The data from most sites are typically available within a few hours after the actual data taking. Considering the time frame of the analysis steps, it is probably best to work with the hourly files, although they contain measurements on all possible satellites and the data on satellites other than the LAGEOSs and Etalons would have to be filtered out. All files include unsorted SLR observations received by the archiving facilities on a particular day or hour, and they are not necessarily taken on that specific day. Occasionally, they may consist of re-releases issued after the detection of a problem in a previous first release.

Hourly CDDIS files are available from ftp://cddis.gsfc.nasa.gov/slr/data/npt/allsat, with subdirectory YYYY (representing the 4-digit year). The files are named new_qlYYMMDDA.allsat where YY is the 2-digit year, MM is the 2-digit month, DD is the 2-digit day of the month, and A is the letter for the hour of that day (e.g. "a" is hour 00, "b" is hour 01, etc.). Daily CDDIS files are available from ftp://cddis.gsfc.nasa.gov/slr/data/npt, in the directory SATNAME/YYYY, where SATNAME is the ILRS satellite name (e.g. "lageos1"). The files are named new_qlYYMMDD.SAT, where SAT is a 3-letter CDDIS abbreviation for the satellite (e.g. "lag" for LAGEOS-1). These daily files contain single satellite data, are unsorted and are usually available at 07:30 EST. These daily files contain any normal points received within the past 24 hours.

Hourly EDC files are available from ftp://ftp.dgfi.badw-muenchen.de/pub/laser/qldata/hourly_global. The files are named global_qlYYMMDD##, where ## is the hour of the day (i.e. "05" is the fifth hour of that day). Daily EDC files are available from ftp://ftp.dgfi.badw-muenchen.de/pub/laser/qldata, and then in the subdirectory SATNAME (e.g. "lageos1"). The files are named SAT_YYMM.DD, where SAT is a 3-letter EDC abbreviation for the satellite (e.g. "la1" for LAGEOS-1). These daily files contain single satellite data only taken on that particular day. These EDC daily files are dynamic and are updated automatically as new data becomes available.

2.2. Products

The analysis groups are required to process the observations in intervals of 28 days. Since products are to be delivered on a weekly basis, successive 28-day data arcs will shift by 7 days. In each analysis, the contributors are to generate a loosely-constrained solution for EOPs and station coordinates. The EOPs must include x-pole, y-pole and excess Length-Of-Day (LOD), all computed on a daily basis, where the intervals encompass one complete day (the reference epoch is the midpoint of each day). Daily UT parameters may also be solved for, but they are of course considered as weakly-determined parameters by any satellite technique (because of the strong correlation with the value for the right ascension of the ascending node) and are not to be included in the analysis product that is submitted to the combination center. Similarly, satellite (orbital) parameters are not required in the submission. Irrespective of the actual timing of the data period, the analysis center is to provide (loosely constrained) daily EOP solutions for all 28 days.

The solution must also include loosely constrained station coordinates, with the midpoint of each 28-day interval as reference epoch. The coordinates solutions are to refer to the official station markers. The latter are system dependent and are either the intersection of the optical axes or an actual ground monument with an associated set of eccentricities. Eccentricities are measured from the ground monument to the intersection of the axes. Eccentricities for all sites are available from the CDDIS in SINEX format in both Up, North and East (ftp://cddis.gsfc.nasa.gov/slrocc/ecc_une.snx) and X,Y,Z (ftp://cddis.gsfc.nasa.gov/slrocc/ecc_xyz.snx). Sites for which the marker is the optical axes intersection, have a "zero" eccentricity vector and have an "S" in their IERS DOMES number. This definition follows the agreements which were made earlier for the AWG pilot project "positioning and earth orientation". Station velocities are to be kept fixed at the rates according to the ITRF2000 solution (http://lareg.ensg.ign.fr/ITRF/ITRF2000); if not available they may for example be modeled according to the NUVEL-1A NNR model (DeMets et al, 1994).

Based on the outcome of initial AWG analyses, contributors are required to use the AWG station categorization in their analysis; this differentiates between Core, Contributing and Associate Stations (for the current analysis purposes). Although the discussion on this categorization has not been finalized at the time of writing of this document, specific analyst actions are required for the various stations in the different classes; they are specified below. In principle, this categorization of the tracking stations follows the rules that have been established during the International Workshop on Laser Ranging in Shanghai (1996). This qualification is/will be performed by the AWG on a semi-annual basis, and a preliminary classification based on these rules is available by e-mail request to the ILRS Analysis Coordinator.

With the purpose of avoiding numerical singularities or otherwise weak (coordinates) solutions, it is recommended that a minimum of 30 acceptable normal points for the sum of observations on LAGEOS-1 and LAGEOS-2 is required for stations to be included in a particular data reduction; if this criterion can not be met, the station is to be deleted from the analysis for this specific 28-day solution period.

Each analysis contributor is free to construct their own computation model and/or analysis strategy, but a number of constraints do exist:

  • The computation model must follow the IERS 2000 Conventions (http://maia.usno.navy.mil/conv2000.html) as closely as possible.
  • The observations of AWG "Core" and "Contributing" stations must be included in the analysis, unless the data yield is too small to satisfy the "=30 NP" criterion. Data weighting can be done according to the analyst's preference. The analyst is free to choose whether to include observations of other stations in their analysis (again, recognizing the "=30 NP" rule). If included, the weights of the latter category of stations must be 1/1000 times the (average) weight of the AWG "Core" stations (i.e. the measurements must be "downweighted").
  • Range biases should not be modelled or estimated for the AWG "Core" stations, unless there is a clear evidence for the existence of a physical systematic error. As for the other stations, the analyst is free to introduce range biases where deemed necessary.
  • The solutions must be loosely constrained with an a priori standard deviation sapriori on station coordinates and EOPs. The value of sapriori must be (approximately) equal to 100 times the (average) standard deviation of the SLR observations of the particular station as applied by the analyst. Minimum constraints are not to be applied in this stage of the analysis.

It is strongly recommended that analysis centers intending to respond to this Call for Participation draw upon the experience of processing SLR data gained by the ILRS Analysis Working Group during the Pilot Project ("pos+eop") phase of this work. The minutes of all AWG meetings are available via the ILRS website.

The procedure is aimed at the generation of an official ILRS product for EOPs, for inclusion in the IERS Bulletin A, and a time-series of global station coordinates. The procedures for generating individual solutions, quality control and combination should be developed in such a way that alternative products (e.g. time-series of geocenter solutions) are easily obtainable.

2.3. Formats

The SLR observations, both on the LAGEOS and on the Etalon satellites, are available in the ILRS NP format.

The analysis results (EOPs, station coordinates) are to be provided in the SINEX V2.0 format (ftp://alpha.fesg.tu-muenchen.de/iers/sinex/format or via the ILRS website at http://ilrs.gsfc.nasa.gov/products_formats_procedures/site_positions_velocities/sinex_file_description.html), in such a way that the original unconstrained normal equation can be reconstructed by the combination centers. The format also prescribes the units of the various parameters. In other words, the analysis contributor either has to provide the results in the form of a full normal equation (where the existence of numerical singularities induced by stations with insufficient numbers of observations is checked and eliminated beforehand; the singularities between the EOPs and the global network exist by definition) or in the form of a full covariance matrix and solution, which is checked to be invertable (i.e. the solution has to be loosely constrained, the constraint has to be fully removable) by the analysis contributor before submission. Contributors are urged to access the ILRS web pages mentioned here to get clarification on the use of the format and to avoid obvious errors and misinterpretations.

The solutions are to be submitted in a compressed (".Z") form, both to the CDDIS and EDC. They have to be uploaded to the subdirectory /incoming/ilrs/products/ABC/ (identical on both computer systems), with the filename ABC.yymmdd.snx.Z. Here "ABC" is to be replaced by a code for the particular analysis institute, and "yymmdd" is the date (including a possible leading zero) of the last day of the 28-day time span. The institute codes will be made available at request. CDDIS and EDC will automatically transfer the results to the directories /pub/slr/ilrsac/products/ops/ABC/.

2.4. Operational timeline

The timeline of the operational activities is shown in Figure 1. The schedule is dictated by the time of release of the IERS Bulletins A, which is in the course of the Thursday of each week. Counting back, this provides weekly deadlines for the other activities within this operational scheme. The schedule for the various actions is illustrated in Figure 1, which also shows the successive data periods to be analysed and the status of the various EOP products.

Because of the weekly frequency, successive analyses will overlap for 21 of the 28 days. The 28-day interval was chosen to guarantee a good and robust dataset for the generation of specific products (in particular relating to station coordinates and, as a possible future product, geocenter). As shown here, the EOP products that will be delivered to IERS will only relate to the last 7 days of each analysis, so a direct overlap of EOP solutions is avoided. For the station coordinates, the solutions are derived from largely overlapping datasets, which means that the successive network solutions are not fully independent. When a time-series of independent network solutions is required, users will have to sample one out of 4 successive solutions.

As can be seen in Figure 1, the contributing analysis groups will have to collect all SLR data taken on the target satellites until 24:00 hrs GMT of each Monday. The reduction to parameter solutions (including a local quality control check) will have to be done on the subsequent Tuesday, and the analysis results will have to be submitted to the CDDIS and EDC data centers on 24:00 hrs GMT of that day at the latest. This will necessitate the development of a fully autonomous analysis system.

3. Quality control and combination

This chapter describes the various aspects of the merger of the individual solutions for the EOPs and station coordinates into official ILRS products.

3.1. Input

Unlike the institutes contributing to the data reduction and generating "direct" analysis products (EOPs, station coordinates), the groups involved in the quality control and combination (named "combination centers" hereafter) will not deal with the original SLR observations. Instead, solutions for EOPs (x/y-pole and LOD) and station coordinates are the input for the quality control and combination part of this Call for Participation. These solutions are provided by the individual analysis groups, and are available (in compressed form) in the SINEX V2.0 format in the CDDIS and EDC computers on the locations /pub/slr/ilrsac/products/ops/ABC/, where "ABC" is a code for the various contributing analysis groups. The results are available on each Tuesday of the week, at 24:00 hrs GMT at the latest.

3.2. Products

The combination center will generate a single solution, including both EOPs and station coordinates. A subset (covering the EOPs for the 7 "new" days only, with station coordinates now mapped onto ITRF2000; cf. Figure 1) is to be extracted from this, for submission to the IERS customer. The EOPs must include x-pole, y-pole and LOD, where each individual EOP interval encompasses one complete day (00:00 - 24:00 hrs; the epochs hold for the midpoint of each day). The station coordinates will hold for the reference points as described in the previous chapter, and will have to refer to the midpoint of the data period. Station velocities clearly are of no concern here (cf. Section 2.2).

The combination center will have to do the following tasks:

  • Check the technical quality of each individual contribution, i.e. check the possibility to remove the (loose) constraints, to invert the covariance matrix and turn the solution into a normal equation configuration. Also, the adherence of the input solutions to the format and contents definition (such as reference ids, eccentricities etc.) should be verified. Failure to adhere to the AWG agreements on this may result in the rejection of individual solutions altogether.
  • The individual EOP solutions are to be combined into a single solution with best possible quality and with realistic uncertainty estimates.
  • The full solution (i.e. averaged station coordinates and averaged EOP solutions for 28 1-day intervals) has to be in the relevant ILRS archives (at CDDIS and EDC) before 24:00 hrs GMT of the processing day, i.e. Wednesday (cf. Figure 1). The location is (on both computers) /incoming/ilrs/products/ABC/, and the results will be moved to the final directory /pub/slr/products/ops/pos+eop/ automatically. This solution also has to be provided in the SINEX V2.0 format, in a compressed (i.e. ".Z") form, with the name ABC.pos+eop.yymmdd.snx.Z (where "ABC" represents the code for the combination center, and "yymmdd" represents the date of the final day in the 28-day interval).
  • This full solution has to be mapped onto ITRF2000 to bring the results into a consistent reference frame, and the resulting EOP solutions for the "new" days have to be submitted to IERS and other interested parties on a timely basis, i.e. before 24:00 hrs GMT of each Wednesday. A copy will have to be put in the appropriate ILRS archives at the same time. The location is (on both computers) /incoming/ilrs/products/ABC/, and the results will be moved to the final directory /pub/slr/products/ops/eop/ automatically. These submissions have to be in the SINEX V2.0 format, in a compressed (i.e. ".Z") form, with the name ABC.eop.yymmdd.snx.Z (where "ABC" represents the code for the combination center, and "yymmdd" represents the date of the final day in the 28-day interval).
  • The results of the quality control and combination activities will have to be summarized in a (weekly) report, which will have to be distributed among the contributing analysis centers, the AWG (co-)coordinator, the customer(s) and other interested people and/or organizations. This report will include an evaluation of the input solutions (values, claimed and actual accuracy), statistics on the quality control, and combination results (values, statistics). The location is (on both computers) /incoming/ilrs/products/ABC/, and the results will be moved to the final directory /pub/slr/products/ops/summary/ automatically. These submissions have to be with the name ABC.pos+eop.yymmdd.sum.Z (where "ABC" represents the code for the combination center, and "yymmdd" represents the date of the final day in the 28-day interval).
  • The procedure is aimed at the generation of an official ILRS product for EOPs, for inclusion in the IERS Bulletin A, and at a time-series of coordinates for the global SLR network (including the corresponding covariance matrix). The procedures for generating individual solutions, quality control and combination should be developed in such a way that alternative products (e.g. time-series of geocenter observations) are easily obtainable.

    3.3. Formats

    The combination center will receive the information on EOPs and station coordinates in the SINEX V2.0 format (ftp://alpha.fesg.tu-muenchen.de/iers/sinex/format or via http://ilrs.gsfc.nasa.gov/products_formats_procedures/site_positions_velocities/sinex_file_description.html). The combination product on EOPs as well as the full solution for internal ILRS use also will have to be represented in the aforementioned SINEX V2.0 format. Again, check the relevant ILRS web pages to avoid errors and misinterpretation.

    3.4. Operational timeline

    As is the case for the analysis contributors, the combination centers will have to satisfy a strict timeline, which is driven by the acquisition of the original SLR observations and the daily product delivery. Again, the overall flow is summarized in Figure 1.

    The analysis products (i.e. the parameter solutions, either in the form of a normal equation or in the form of a solution accompanied by a full covariance matrix and a priori information) will be available at 24:00 hrs GMT on each Tuesday. According to the current scheme, the combination center will do its work on the next day, and deliver the official product at 24:00 hrs GMT of that Wednesday at the latest. These products will be sent to the customers and the ILRS archiving facilities.

    4. Selection criteria

    Individual analysis groups are invited to respond to this invitation, for either or both of the functions specified here. The acceptation and/or selection of the candidate participants in the two activities will be based on a number of criteria:

    Analysis contributions:

    A1. The contributor must be an ILRS (Associate) Analysis Center.

    A2. The contributing analysis groups will have to pass benchmark tests as being developed by the AWG; this stage is aimed at preventing gross blunders in the analysis scheme or analysis software. In particular, the analysis results for the 3-dimensional satellite orbit, 3-dimensional station coordinates, EOPs and measurement corrections will have to match a reference product within certain limits (the values of which are TBD at the time of writing; cf. Section 5). The evaluation will be done under responsibility of the AWG. If an analysis institute fails to comply, it will not be able to contribute to the official ILRS product on operational EOPs and station coordinates (and future derivatives) until any software problem(s) have been fixed.
    Technical details on the benchmark requirements and data can be found on http://ilrs.gsfc.nasa.gov/working_groups/awg/awg_pilot_projects/awg_software_benchmarking.html.

    A3. The candidate contributor must show its ability to do a self-test of the quality (format, contents-wise) of its solution, e.g. by using the tools offered by the University of Newcastle (http://ucscgi2.ncl.ac.uk/~nkn3), or similar testing procedures curently under development at other institutes.

    A4. The candidate contributor must express its willingness to perform the role of official contributor to this operational analysis for (at least) 3 years.

    A5. The contributor can not expect any financial support of the ILRS.

    A6. The contributor must demonstrate that it can perform its function within the time constraints mentioned earlier and with a technical quality that is compatible with that of the other analysis groups.

    Combination contributions:

    C1. The contributor must be an ILRS (Associate) Analysis Center.

    C2. The candidate contributor must express its willingness to perform the role of combination center for (at least) 3 years.

    C3. The combination center (prime or back-up) can not expect any financial support of the ILRS.

    C4. The contributor must demonstrate, in the (internal) development phase as well as during a full test phase including the data reduction groups, that it can perform its function within the time constraints mentioned earlier and with good technical quality.

    In principle, the prime and back-up combination centers will perform the same task (albeit that they will probably do so with different procedures and software). The selection of the prime combination center will depend on the results of the full-scale test phase running through the period June 1 - October 1, 2003; participants in this operational product development are obliged to contribute their products from the initial date onwards.

    5. Milestones

    In the course of June 2003, the AWG will establish the parameters for meeting the "benchmark" criteria (cf. Section 4). Candidate data reduction (combination) contributors will have to notify their intentions with argumentation to the ILRS Analysis Coordinator by e-mail (address: see below) before May 31, 2003. This will have to address issues A1-A5 (C1-C4) mentioned in Section 4. Contributors who already responded to the version 2.1 of this Call for Participation, in December 2002, do not need to repeat their notification of interest. The results will be reviewed in October 2003 (the AWG meeting in Wettzell, end of October, will probably finish this selection process).

    The selection of the data reduction centers will be finalized there, in particular considering item A6. The evaluation of the candidate combination centers and the assessment of prime and backup centers (multiple centers are possible for the latter function) will also be done during that particular meeting, using the criteria C1-C4. Both for the candidate data reduction centers as well as for the candidate combination centers, outright rejections are possible. At the Wettzell 2003 meeting, the AWG is expected to reach a final recommendation, which will go to the ILRS Governing Board for consent. Everything running as planned, the system will be fully operational as of November 1, 2003.

    6. Further information

    For further information, one may contact the ILRS analysis coordinator Ron Noomen (e-mail ron.noomen@deos.tudelft.nl, tel. +31-15-2785377), the ILRS analysis co-coordinator Pete Shelus (e-mail pjs@astro.as.utexas.edu, tel. +1-512-471-3339) or the ILRS Governing Board analysis representative Graham Appleby (e-mail gapp@nerc.ac.uk, tel. +44-0-1487-772477). Information (background, developments, format, etc.) is also available at the ILRS website http://ilrs.gsfc.nasa.gov.

    7. References

    DeMets, C., R. Gordon, D. Argus and S. Stein, Effect of recent revisions to the geomagnetic reversal time scale on estimates of current plate motions, Geophys. Res. Lett., 21, 2191-2194, 1994.

    Figure 1: Time schedule of operational analysis.