Minutes of the EUREF AC Workshop (Paris, Sep. 09-10, 1999)
by Carine Bruyninx, Royal Observatory of Belgium
The first official solutions for the EUREF permanent network became available in the beginning of 1996. These solutions were based on individual
subnetwork solutions from 4 EUREF Local Analysis Centres.
The number of stations in the network at the time was less then 40. In April 1997, the first EUREF Local Analysis Workshop was organised in Brussels. The main topics of discussion at the time included the establishment of guidelines for data analysis and redistribution of the EUREF subnetwork data. Representatives from eight of the nine EUREF Local Analysis Centres participated in the workshop.
Presently close to 90 stations (more than twice the number in 1996) make up the EUREF permanent network with twelve analysis centres delivering weekly solutions. This growth of the network provides us with an opportunity to take the EUREF network further than its initial purpose, the maintenance of the European Reference Frame.
This opportunity was formally recognised at the ninth EUREF symposium, held from June 2 to June 4 1999, via Resolution No 4.
In order to discuss how the EUREF permanent GPS network can evolve towards other disciplines, an EUREF Analysis Centre workshop has been organised in Paris, France on September 9-10, 1999. It was an opportunity for the AC's to interact with scientists who may be interested in working with the EUREF dense regional network toward the establishment of multi-disciplinary EUREF products.
Topics for the workshop included :
Viewgraphs of the presentations can be viewed by clicking on the title of the presentation.
Summary of Discussions/Action Items
CONNECTION WITH OTHER SCIENCES
1. Sea level monitoring
Q: Does EUREF accept to integrate the permanent stations installed following the recommendations of COST Action 40 (EOSS) to monitor tide gauges, into its permanent network?
A: All new permanent stations that comply with the EUREF standards can be included into the EUREF network. Of course, new stations should be located
far enough from already existing stations.
If the tide gauge is monitored using one permanent station and this station fulfils EUREF standards, then this station can be integrated in the EUREF network.
If two permanent stations are installed for the monitoring of one tide gauge, then EUREF proposes to include the most stable station into its network. EUREF will monitor the long-term stability of the station and maintain station history information. The tie to the non-EUREF permanent station, in the direct vicinity of the tide gauges, can be considered as an eccentricity problem. A dedicated analysis centre that could process the data of all the permanent stations used to monitor the tide gauges can maintain this tie. If this analysis centre processes this network following the EUREF analysis guidelines, then, through the common sites, this network can easily be linked to the EUREF network.
Q: One of the strengths of the permanent EUREF network lies in its overlapping data processing (3 AC's for each station). However, presently already 21 stations are not processed by 3 analysis centres and the network is still growing with about 13-14 stations each year. Are the EUREF AC's willing to enlarge their subnetwork to guarantee that all EUREF stations can be processed 3 times?
Q: Can the EUREF AC also make available velocities for the non-EUREF stations included in their subnetwork? This would be very helpful for to study deformations and compute strain rates. The EUREF network is presently not dense enough in a lot of regions.
A: Neither the weekly solution submitted by the EUREF AC's, nor the combined EUREF solution, does include any velocity estimation. These solutions,
available in SINEX format, do only include the coordinate estimation and their covariance info. The combination centre at BKG only does velocity
estimation in order to generate the multi-year EUREF solution that is submitted each year to the IERS. The IERS combines this multi-year solution with
other solutions from various space techniques to generate the different realisations of the ITRS. It is important to note that velocity estimations
are closely linked to the underlying reference frame. It is consequently necessary to have the information about the reference frame before make
any interpretations of the velocities.
From the coordinate time series that are made available at the EUREF web-pages is can be seen that the velocity obtained for some of the EUREF stations is biased by antenna changes or environmental changes (installations/removals of radomes, snow on the antenna). For stations belonging to the EUREF network, most of this information is available and maintained at the EUREF web-site. However, if velocities should be made available for non-EUREF stations, then no guarantee whatsoever can be made about the quality of these velocities. Scientists interested in site velocities of non-EUREF stations are welcome to contact any of the individual AC's to make available their results and work together with other groups in the field of geology.
Not all EUREF stations have a monumentation and geology suited for deformation studies. Specific guidelines about this have been drawn by H. Drewes (CSTG document). It is recommended to distribute this document among the station managers and ask them to make the necessary information about the site stability available to the EUREF Central Bureau (ACTION: C. Bruyninx).
Q: Can EUREF make available ionosphere maps over Europe?
A: In principle all analysis centres should be able to generate these maps. However, the individual maps need to be combined and there is no combination centre yet. If these maps need to be available in real-time, then this would require a non-negligible amount of supplementary work from the analysis centres. Before the analysis centres start this work, it is necessary to know if there is a need for this type of product. The ASI/CGS analysis centre will look into this problem (ACTION: L. Ferraro/F. Vespe).
Q: Would it be possible to make available daily subnetwork solutions in addition to the weekly solutions? This would allow to investigate if the signals visible in the coordinates are real signals (geophysical, monumentation) or by-products of the GPS computation method.
A: The computation of daily combined solutions for the EUREF network would increase the workload on the combination centre with a considerable amount. If the AC's would make available their daily solutions (without them being combined) then it is necessary to know that there is a real interest to have these daily solutions available. The general feeling is that there are still so many things unexplored in the weekly time series and that it is too soon to start investigating the daily time series yet.
Q: Would it be possible to encourage the installation of permanent EUREF stations with Water Vapour Radiometer instruments, meteo sensors, radiosonde launches, SLR and VLBI?
A: Yes. In the past, collocation with SLR and VLBI was already encouraged for reference frame maintenance, but now this collocation could also be useful to compare different troposphere estimates. Within this frame the collocation with meteo sensors, WVR and radiosonde launches should be encouraged. The information about the different collocations should be made available at the EUREF Central Bureau (ACTION: C. Bruyninx).
6. Near real-time EUREF products
Q: For several near real-time applications is it necessary to have good quality GPS orbits available in real-time. Will EUREF make available these orbits?
A: The EUREF AC's recognise the need for real-time orbits. Presently the only orbits available in real-time are the predicted IGS orbits. However, the quality of these orbits is not always sufficient for high accuracy applications. It is important to now if IGS plans to make available near real-time orbits in the near future. It is recommended to write a letter to the IGS to emphasise the EUREF's need for near real-time orbits and to inquire what the IGS policy is for making available these orbits (ACTION EUREF TWG). If IGS does not have any plans to make available near real-time orbits is a close future, then one of the EUREF analysis centres agrees to compute these orbits (ACTION: L. Mervart) and make them available to the other EUREF AC's.
Q: Is there a need for a liaison person between the EUREF AC's and the COST Action 716? This person could then also act as a "troposphere coordinator" for the EUREF AC's.
A: Yes! A "troposphere coordinator" should play the motor driving the EUREF AC's towards near real-time tropospheric total zenith delay (TDZ) estimates and he should be the link to the users community, through its engagement in COST Action 40. The troposphere coordinator will have to work closely together with the network coordinator. Task of the troposphere coordinator could be to :
Optimal would be to appoint a troposphere coordinator within those people that are already now very active within the EUREF data analysis. The network coordinator will contact the AC's and ask for volunteers for this job.
COMBINED EUREF SOLUTION / OPTIMISATION OF DATA ANALYSIS
Q: Presently the EUREF combined solution is available with a delay of 4 weeks after observation. Taking into account that the precise IGS orbits are available after 2-3 weeks, is it possible to have the combined solution faster available?
A: Yes, all EUREF AC's should be able to make their subnetwork solution available within 5 days after the availability of the IGS orbits. A circular letter will be distributed among the AC's to inform them about this new deadline (ACTION/NEW GUIDELINE AC's).
Q: For some stations the solutions between the different AC do have inconsistencies at the cm-level. This can not be seen from the summary files of weekly combined EUREF solutions. Can this problem be made more visible?
A: Yes, the residuals of the weekly combination solution can be included in the summary file (ACTION: Combination Centre/M. Becker).
Q: Since 11 of the 12 EUREF AC's are using the same software to produce their weekly solutions submitted to EUREF, wouldn't it be interesting to make available the parameters used by each analysis centre to the other analysis centres?
A: Yes, this would certainly contribute to the standardisation of the different AC solutions. It would give more visibility to the processing methods of the individual analysis centres and it would speed up the implementation of new developments. This is not only interesting for those analysis centres working with the same software (Bernese), but also for the analysis centre (ASI/CGS) working with a different software. Even when different software's are used, some of the input parameters are still comparable. At the EUREF CB a special account, with limited access, will be created (ACTION: C. Bruyninx) where all analysis centres can put the files that control their processing (for the Bernese users: the OPT, PCF and SCRIPT directories) (ACTION: EUREF AC's).
Q: Should the EUREF AC's include data from combined GPS/GLONASS receivers in their data analysis?
A: This will be discussed at the next EUREF Technical Working Group Meeting (October 28-29). Analysis centres that have experience with the processing of GLONASS data will be invited to present their experiences and results at this meeting. The outcome of the IGEX analysis workshop (end of September) will also give valuable information. Since the EUREF solution is submitted to the IGS, it is important to know the IGS point of view about this subject (ACTION: EUREF TWG).
Q: When the new version of the Bernese software becomes available, then it will be possible to apply ocean loading corrections. In order to do this some station dependent parameters need to be available. Can the EUREF Central Bureau create an interactive web page where these ocean loading correction parameters can be obtained?
A: Presently, H.-G. Sherneck provides ocean loading parameters after an e-mail request. A fully automated interactive procedure would of course be faster. The EUREF CB will have a look into this problem (ACTION: C. Bruyninx).