EUVN - European Vertical Reference Network
The initial practical objective of the EUVN project is to unify different European height datums within few centimeters. In addition this project is to prepare a geokinematic height reference system for Europe and to connect levelling heights with GPS heights.
At all EUVN points P three-dimensional coordinates in the ETRS89 (XP, YP, ZP)ETRS and geopotential numbers cP = W0 UELN - WP will be derived. Finally the EUVN is representing a geometrical-physical reference frame. In addition to the geopotential numbers cP normal heights Hn = cP / γ will be provided (γ is the mean normal gravity value between the ellipsoid and the telluroid).
The application of the GPS technique for practical levelling would dramatically extend if the geoid would be known precisely enough in relation to the concerned GPS reference system and the levelling reference system. To derive such a geoid, an European reference geoid is required in the reference system ETRS89 and the reference system of UELN. Up to now there is no precise geoid available for Europe with an accuracy of a few centimeters which fulfils the requirement for the practical applications. This proposal points out a possibility to derive a geoid tailored for the GPS-levelling methods by combining the existing reference network EUREF / ETRS89 with the UELN95.
The EUVN project contributes to the realization of an European vertical datum and to connect different sea levels of European oceans with respect to the work PSMSL (Permanent Service of Mean Sea Level) and of anticipated accelerated sea level rise due to global warming. The project provides a contribution to the determination of an absolute world height system.
Three kinds of observation groups are necessary:
- GPS measurements for the determination of the ellipsoidal heights of all defined EUVN points
- levellings between the EUVN sites and the UELN nodal points for the determination of the physical height of all defined EUVN points
observations of sea level at tide gauge stations.
In total the EUVN consists of about 196 sites: 66 EUREF and 13 national permanent sites, 54 UELN and UPLN of Central and Eastern Europe (United Precise Levelling Network of Central and Eastern Europe) stations and 63 tide gauges (see Figure 1).
The GPS observations for the EUVN were carried out in the period from May 21 to May 29, 1997. Three types of GPS receiver were used: 35 Turbo Rogue Receivers, 134 Trimble SSI or SSE and 51 Ashtech Z12. The time interval was set to 30 s, the elevation mask was 5 degree. The campaign was running very smoothly and everybody who participated in the campaign supported the action successfully.
The data preprocessing after the EUVN campaign performed by 9 EUVN Preprocessing Centers (PPC) was mainly a check concerning completeness and consistency of the data and the auxiliary information. The Preprocessing Centers were requested to prepare complete access information and/or data flow guidelines for the observing agencies before the start of the campaign.
The task of the EUVN GPS Analysis Center (AC) was to process the data of a special subnetwork. A subdivision of the whole EUVN Network was done under the aspect of receiver type and regions.
10 European institutions were ready to contribute as Analysis Centers. On the Analysis Center Workshop in September 1997 in Leipzig the subdivision of EUVN was discussed and decided (see Figure 2) (Ineichen et al., 1998). The AC of Croatia was responsible for the analysis of the collocation points and the investigation of the biases introduced by using different antenna types within one GPS network. Simultaneously with the EUVN97 Campaign the Baltic Sea Level (BSL) GPS campaign was performed. The BSL97 GPS campaign was processed by the Finnish Geodetic Institute.
The Astronomical Institute of the University of Bern (AIUB) and the BKG were responsible for the computation of the final GPS solution of EUVN (Ineichen et al, 1998).
The analysis centers produced different solution types in order to investigate the influence of the processing strategy on the results. Mainly the following three types of solutions were looked at:
- 15 degrees without weighting: The standard solution with the highest priority. Data down to an elevation cut-off angle of 15 degrees were used for generating this solution type. All observations were introduced with the same weight. This solution type corresponds to the processing strategy used for the permanent EUREF network at the time of the EUVN campaign.
- 5 degrees with elevation-dependent weighting: Measurements down to an elevation cut-off angle of 5 degrees were used for this solution type. In addition, the observations were weighted with w = cos2 (z), where z is the zenith angle of the observed satellite.
Satellite-specific weighting: The IGS precise orbit files (in SP3 format) contain accuracy codes for each satellite. These accuracy codes can be used by the Bernese GPS Software to weight the corresponding observations. Not all analysis centers delivered solutions of this type, and therefore no combined solution was generated.
The question which solution to choose as the official EUVN97 solution (the unweighted 15-degree solution or the weighted 5-degree solution), was discussed during the Analysis Center Workshop at Wettzell (April 2-3, 1998): The unweighted 15-degree solution was selected as the official one. The following aspects had to be taken into account:
- The comparison of the height component of redundant points in both solution types showed a slightly better repeatability for the unweighted 15-degree solution.
- Not all sites within EUVN97 were tracking satellites below 15 degrees with the same quality and quantity. For some sites the number of observations is hardly increasing when changing to the lower cut-off angle, whereas for others the number of observations increased by up to 20 %. Therefore the site coordinates within the EUVN97 GPS network could be more inhomogeneous in the 5-degree solution.
- The elevation-dependent antenna phase center variations are not yet well known below 10 degrees. Introduction of poorly defined corrections could lead to additional systematic errors.
- We do not have enough experience yet with the performance of the tropospheric mapping functions at very low elevations.
The final solution was constrained to ITRF96 coordinates (epoch 1997.4) of 37 stations with an a-priori standard deviation of 0.01 mm for each coordinate component. As a consequence of these tight constraints the resulting coordinates of the reference points are virtually identical with the ITRF96 values.
A comparison for the combined solutions of BKG and AIUB showed that these two solutions were identical.
For many practical purposes it is useful to have the ETRS89 coordinates available. To get conformity with other projects, the general relations between ITRS and ETRS were used. The coordinate transform formula from ITRF (epoch1997.4) to ETRF96 and the final coordinates (ETRF96, epoch 1997.4) are given in Ineichen et al 1998.
In order to reach the goal it was necessary to connect the EUVN stations by levellings with nodal points of relevant levelling networks. So it is possible to use levelling observations to update the gravity related EUVN heights in context with the new adjustment of UELN.
The heights are related to the solution of the UELN 95/98 which was distributed in January 1999 to the participating countries. For islands and countries which are not connected with the UELN, the heights are given to local tide gauges/vertical datums. For all stations GPS / levelling height anomalies by using ETRS89/97.4 heights are given and compared with the European quasigeoid EGG97.
As the EUVN is a static height network it was necessary to know the value of the mean sea level in relation to the tide gauge bench mark at the epoch of EUVN GPS campaign 1997.5. However for future tasks it is useful to have available the monthly mean values over a period of some years.
The PSMSL, as member of the Federation of the Astronomical and Geophysical Data Analysis Service (FAGS), is in principle in charge of the data collection. The information which are sent to the (PSMSL databank in general should also be made available for the EUVN project. The complete information about the connection between tide gauge and GPS site could supplied for only 54 of the 79 EUVN tide gauge stations (see Figure 3). Figure 4 shows the heights of the mean sea level 1997 above the GPS / levelling quasigeoid of EUVN (Wöppelmann et. al., 2002).
EUVN is a step to establish a fundamental network for a further geokinematic height reference system such as European Vertical System under the special consideration of the Fennoscandian uplift and the uplift in the Carpathian-Balkan region.
EUVN - GPS Solution
Results of GPS measurements for the determination of the ellipsoidal heights of all defined EUVN points
The following final products were generated: coordinate files containing geocentric coordinates and ellipsoidal coordiantes in system ITRF96 epoch 1997.4 and ETRS89 (ETRF96,epoch 1997.4) and a file in the Solution INdependent EXchange (SINEX) format, containg in addition to the coordinates the variance-covarance information of the EUVN97 GPS network.
For detailed information see paper "EUVN97 Combined GPS Solution"
|Coordinate Reference System
|EUVN_ETRF.CRD (txt, 16KB)
|EUVN_ETRF.BLH (txt, 16KB)
|ITRF96 epoch 1997.4
|EUVN_ITRF.CRD (txt, 16KB)
|ITRF96 epoch 1997.4
|EUVN_ITRF.BLH (txt, 16KB)
|ITRF96 epoch 1997.4
|Solution in SINEX format
|EUVN_ITRF.SNX.Z (x-compress, 2MB) (Unix-Compressed)
EUVN - Height Solution
For detailed information see paper "The EUVN Height Solution - Report of the EUVN Working Group"
EUVN Tide Gauge Data
For detailled information see paper "Status Report on sea-level data collection and analysis within the EUVN Project"
Compilation of the Numerical Results
Find a table with the
- Ellipsoidal coordinates in ITRF96 Epoch 1997.4
- Normal heights H in UELN95/98
- Difference ellipsoidal height h - normal height H
- EGG97 Geoid
- Difference EGG97 - (h-H)
- Ellipsoidal height of the mean see level 1997 above the EUVN tide gauge stations
- Normal height of the mean see level 1997 above the EUVN tide gauge stations
Download of table with numerical results:
- Table (PDF, 97KB) - EUVN Numerical Results