Advances in Positioning and Reference Frames

The World Geodetic System (WGS) was conceived as a practical geodetic reference system that maintains consistency with the best scientific terrestrial reference system at the time but also retains some stability. It has evolved from its beginnings as WGS 60 to its present manifestation as WGS 84. A significantly improved Earth Gravitational Model and global geoid were released in 1996, and the realization of the WGS 84 reference frame is consistent with the International Terrestrial Reference Frame (ITRF) 1994 at the 5-cm level. Evaluations of WGS 84 relative to ITRF94 have been made by comparing International GPS Service (IGS) GPS orbits with the National Imagery and Mapping Agency’s precise orbits and by computing WGS 84 positions for a globally distributed set of IGS stations with known ITRF94 coordinates. The results indicate that there is no practical difference between ITRF94 and WGS 84 for mapping, charting, navigation and many survey applications.

The tools and infrastructure are now in place to realize a global, terrestrial, kinematic reference frame with few millimeter precision, which is both spatially dense, and can be reliably updated on a frequent basis (e.g., monthly). A procedure is outlined by which IERS can continuously realize the ITRF using methodology developed by the International GPS Service (IGS). However, spatial densification not only requires such a processing scheme, but also requires the recruitment of interested groups to participate in this venture.

Taking into account the growing number of permanent GPS stations in Europe, the EUREF subcommission decided in 1995 to coordinate the activities related to this network for the maintenance of the European Reference Frame.The EUREF permanent GPS network, which is the European densification of the IGS network, presently consists of more than 60 GPS stations, covering 23 countries all over Europe. The processing scheme allows for distributed processing : 10 Local Analysis Centers each analyze a part of the EUREF network. One analysis center is responsible for merging the individual subnetwork solutions into one European solution. This solution in submitted to the International GPS Service for Geodynamics (IGS) within the frame of the IGS Densification Project.Weekly free-network solutions for Europe are available since April 1996. The mean RMS values of the coordinate residuals of the combined solution with respect to the individual solutions is about 2 mm for the north and east components and 6 mm for the height component. The agreement between the solutions of the different analysis centers is of the same quality as the week-to-week repeatabilities of each analysis center.In may 1997, EUREF submitted its solution to the International Earth Rotation Service. As a result, all EUREF stations will show up in the next realization of the International Terrestrial Reference System.

The European Vertical GPS Reference Network (EUVN) is designed to contribute to the unification of different height systems in Europe. The most important practical and scientific aspects are contribution to a unique European height datumconnection of European tide gauge benchmarks as a contribution to monitoring absolute sea level variationsestablishment of fiducial points for the European geoid determination preparation of a European Vertical Kinematic Network.A network of about 195 points distributed over Europe and consisting of 79 EUREF-points, of 53 nodal points of the Levelling Networks of Eastern and Western Europe and 63 tide gauges has been observed in the period of May 21 to 29, 1997 with GPS in order to derive uniform ellipsoidal heights in the frame of ETRS89.

In Commission X (continental networks) a lot of new scientific activities have started. For Europe this task is overtaken since 1989 by the EUREF-subcommission in strong cooperation with the European Survey Agencies represented by the „Comité Européen des Responsables de la Cartographie Officielle (CERCO)”. Until now only the European Terrestrial Reference System 1989 (ETRS 89) has already been realized which is identical to the global realization of the International Terrestrial Reference System (ITRS) in Europe at the epoch 1989.0.The activities for a similar solution for the height component are much more time-consuming. That is why the EUREF-Technical Working Group in view of the different time frames of the main users decided in 1993 to divide the realization into two steps: Step 1: United European Levelling Net 1995 (UELN-95)A „quick“ solution on the 0,1 m-level as height reference for Europeanwide geoinformation using systems, using existing levelling data sets and a simple static adjustment model.Step 2: European Vertical System 2000 (EVS 2000)A realization on the highest possible level (1 cm and better) for National Survey Agencies and scientific requirements. The activities for UELN-95 have to be carried out as usable basis for this step 2. The paper describes the problems of a change from national to continental datums, the status of the first step UELN-95 and the concept of the realization of EVS 2000.

Recent developments of the in-house GPS processing software at the Institute of Engineering Surveying and Space Geodesy (IESSG), University of Nottingham, has led to the capability of processing very large or even global GPS networks. This has allowed the investigation of several methods of constraining a network of stations and comparative results of these tests are presented in this paper.A precise knowledge of the Earth’s position and orientation in space is required for a number of astronomical and geophysical purposes, ranging from the tracking of interplanetary spacecraft, to the study of the Earth’s interior structure. The new software package enables the determination of Earth Rotation Parameters (ERPs) at regular intervals from the global network of GPS data, thus defining an Earth Rotation Parameter series and improving the final coordinate solution accuracies. Tests performed at the IESSG are discussed and preliminary results are shown.

In 1994 and 1996, two major GPS campaigns for geodynamics, named GEODYSSEA, were conducted in South East Asia. The paper gives an introduction to the GEODYSSEA project and its results. In cooperation with the national survey authorities, this occasion was used to realize the ITRF in several countries of South East Asia. This paper describes the background and results of GEODYSSEA and the additional measurements as well as transformations of the national datums of Thailand and Malaysia to the ITRF-94. By use of the GEODYSSEA data and also by repeated occupations of the first order stations in the respective countries an accuracy of 2 cm in position and about 3 cm in height was obtained for the ITRF-94 coordinates. Depending on the accuracy of the existing local coordinates the transformation parameters between the national datum and ITRF-94 were determined with and accuracy of 5 to 10 cm.

As part of the realizations of the International Terrestrial Reference System (ITRS), a new complete solution (called ITRF-C1) has been proposed to be the ITRF conventional frame for several years, and superseding ITRF94.

The three Maritime Provinces have been managing a geographical reference framework for the past 40 years. During that period a large number of spatial data sets have been created related to this reference framework. There is no doubt that the existing reference framework “Average Terrestrial System of 1977 (ATS77)” served its purpose well. However, without a framework in place which is consistent with the Global Positioning System (GPS), the technology of the next decade, it is likely that many future positioning applications will never be tied to a consistent reference framework. The result will be incompatibility between data sets collected by different individuals and agencies, and will result in additional costs to the users of the data.The three Maritime Provinces co-sponsored a study of the best way to prepare for the reference framework technology of the future. This study’s recommendations imply profound changes for the way the provincial reference framework could be handled. This presentation looks at the history of the provincial framework, the establishment of a GPS based reference framework tied to the Canadian Space Reference System, the migration from the existing framework to the new framework, the human resources and cost implications associated with these actions and the impact on the user community.

In 1995, a proposal was developed within WEGENER to install and operate a network of Permanent GPS Stations around the Mediterranean area and adjacent areas. Established with the intention of determining the velocity of tectonic plate motions, the network has to cover the region between the Azores in the west and the Caspian Sea in the east. At the BKG, a part of this network is processed on a daily basis. Loosely constrained solutions are generated and analyzed to investigate geodynamic effects and their influence on geodetic reference systems in the Mediterranean area.The time series of coordinates available today cover a period of more than two years. Regional velocities are compared with NUVEL model velocities. The time series are analyzed for short and medium scale variations. Details of the time series analysis are given. Model calculations are presented and their significance for GPS data processing is discussed. The relationship between the obtained results and tidal deformation, and other possible effects, is being studied.

The Brazilian Network for Continuous Monitoring of GPS (RBMC) is an active geodetic reference network. In this paper its automatic operation and products are described during the first nine months of functioning. It consists of nine permanent GPS stations established in Brazil, in cooperation with many groups. Two of the nine stations (Brasilia and Fortaleza) also belong to the International GPS Service for Geodynamics (GS), while the remaining 7 are integrated into this service by contributing the corresponding data to the SIR IGS Regional Network Associate Analysis Center (RNAAC) located at DGFI (Deutsches Geodaetisches Forschungsinstitut), Germany. The future activities include the densification of the network in the Amazon region (six stations) and in Northeastern and Southern Brazil (one station in each area).

During the past years several activities were launched in Austria to establish a state-of-the- art GPS-network which could serve as a basis for geodynamic applications. Recently, in June 1996, an association of engineering consultants initiated a GPS-campaign which covered the Austrian territory with 330 sites with a mean point-to-point distance of about 25 km. The establishment of such an accurate network (+/- 0.4 cm horizontal, +/- 0.8cm height) will be advantageous to all kinds of surveying activities in the near future and provides a considerable contribution to the densification of the ITRF94 reference frame.The Institute of Theoretical Geodesy and Geophysics (ITGG) in Vienna is one of the two analyses centers, which were appointed to process the GPS-network data. This paper summarizes the computational efforts at ITGG and puts special emphasis on the algorithms used for reference frame fixing in the analyses. Additionally, problems of modelling phase center variations (seven different antenna types were used in the network) are discussed.

Observations of several GPS reference receivers in a regional network enable two kinds of error modelling and reduction. The spatially correlated errors (orbit, ionosphere, troposphere) are modelled epoch-by-epoch and satellite-by-satellite by 2-D linear interpolation. Furthermore, multipath effects are mitigated by averaging correspondent observations of several receivers. As a result, observations of a virtual reference station are created. It is assumed to be located at the rover’s approximate position and its observations are used in the precise baseline positioning of the rover. This approach has been tested with several months of observations including data disturbed by ionospheric irregularities, large tropospheric gradients and large orbit errors. It considerably improved ambiguity resolution and reduced coordinate errors in the positioning of rover receivers.

At the Center for Orbit Determination in Europe (CODE), one of the IGS Analysis Centers, we are routinely processing the GPS data of a dense European network. In order to improve the height estimates and, very strongly coupled therewith, the modelling of the troposphere for this network, we decided to include low-elevation data into the processing, something that has been done in VLBI analyses for a long time. An appropriate mapping function, the option to weight the observations according to the elevation angle, and the estimation of troposphere gradients were implemented to make best use of the data at low elevations. Problems may arise from phase center variations, multipath, and the troposphere.143 days have been processed so far using the new processing options and data down to 5 degrees elevation. Comparing the results to a standard 15-degree elevation cut-off solution shows, that the use of low-elevation data together with an elevation-dependent weighting of the observations reduces the scatter in the height estimates by about 20–30 percent. The estimation of tropospheric gradient parameters improves the repeatability in the horizontal site coordinates by a factor of two for sites successfully tracking down to 10 degrees and below.

The paper introduces a new Wide Area Differential (WADGPS) strategy, based on ‘Orbit Relaxation’. The technique is simpler to implement and less costly to maintain than the conventional WADGPS approach, based on a full dynamical orbit integration. The scheme is best suited for a single large country, such as Brazil, or a continental region, such as Western Europe. It can also be used as an extension of the American WAAS or the European EGNOS. The orbit improvement is achieved by computing corrections to the elements of the mean Keplerian orbit, thus generating a refined broadcast ephemeris, which is transmitted to the users. The paper describes this new approach, in conjunction with recent developments in modelling error components in WADGPS. Results of tests performed at the Institute of Engineering Surveying and Space Geodesy (IESSG), at the University of Nottingham, with both simulated and real data, are presented and discussed.

New Zealand, lying across the Pacific/Australian plate boundary, is subject to ground movements within the country of approximately 5 cm/yr. The existing geodetic control network and datum, New Zealand Geodetic Datum (NZGD) 1949, has been distorted by the effect of this ground movement. Because of these distortions, cadastral boundary coordinates, if held fixed in a coordinate cadastre, would eventually conflict with the position of slowly moving ground marks. However, the use of new technology such as GPS is ideally suited to the development of a coordinate cadastre. A dynamic coordinate cadastre is a possible outcome of recently started projects. A programme is underway to develop a modern control network and new datum (referred to as NZGD 2000) to replace the existing network. This could include up to 30 GPS permanent tracking stations which would monitor the integrity of the new datum and determine rates of ground deformation. A four dimensional (dynamic) datum with coordinates assigned velocities and changing to reflect ground movements is an option. As a separate project, the Department is considering the automation of the survey and land title systems. The automation of the survey system would see the digital capture of land parcel dimensions. Through connection to the geodetic network, geodetic coordinates would be assigned to parcel boundary points resulting in a coordinate cadastre. As the geodetic network moves to reflect ground movements, adjustment for this could be applied to boundary points forming a dynamic coordinate cadastre. In this manner, coordinates of boundary points would encapsulate available evidence of their true ground positions. This would not be a legal coordinate cadastre, but in the absence of other evidence the coordinate could be accepted as evidence to define parcels. To take the model fürther, it is conceivable that the cadastral surveyor in performing surveys would contribute to the geodetic network and dynamic datum. This possible model is debated.

A regional GPS tracking network consisting of eight permanently operating GPS stations in Taiwan and its offshore islands has worked since 1995, in order to establish a high accuracy geodetic network purely based on GPS observations. These GPS stations aim to act as the core stations for establishing a high accuracy three-dimensional reference frame and motivating some other geodetic and geodynamic applications in Taiwan. The GPS measurements of this network are to be expected to an accuracy of a few millimeters. This is to be achieved, through the use of high accuracy GPS technique, by linking this network of stations to several primary stations of the ITRF (IERS Terrestrial Reference Frame). Two one-week GPS data from the preliminary stage of measurements made at these regional GPS tracking stations were collected to compute the network adjusted coordinates and test for their stabilities in three-dimensional components. A test of tropospheric modelling was also carried out to investigate the effectiveness of using associated surface measured meteorological data in high accuracy GPS network adjustment.

One aspect of GPS network design concerns the logistics of the survey. Others aspects include location of stations, length of time for session observations. Efficient logistics enables a survey to be carried out with a lower cost and/or shorter time. A significant influence on the efficiency of the survey comes from the order in which the sessions are observed (the schedule). Previous work has shown that, given a list of sessions, the optimal schedule can be determined given the cost of moving receivers between points and the list of sessions. The solution was obtained by transforming the problem into a Multiple Travelling Salesman Problem enabling more than one working period to be included in the solution. However, the method was only suitable for relatively small networks.To enable larger networks to be solved, the use of heuristic techniques within the field of Operational Research have been investigated. Heuristics enable optimal or near-optimal solutions to be found for very large problems with a reasonable computation time — optimality, however, is not guaranteed. This paper will describe one particular heuristic (Simulated Annealing) and show how it can be applied to the logistics design of GPS networks.

GPS/levelling undulations were computed at 20 accurately-levelled stations of a nationwide GPS network. The prediction accuracy of a multiple regression polynomial fitted through these undulations,as well as, long wavelength errors of EGM96, OSU91A and CARIB97 over Trinidad are evaluated. Data to be used in the computation of a gravimetric geoid is also summarized.The 1903 triangulation of Trinidad is readjusted and compared to the 1963 triangulateration in order to decide which control points are reliable for use in horizontal transformation computations and in order to get deformation-free transformation parameters. In addition a multiple regression transformation and the results of a variance-covariance simulation of similarity transformations is presented.

The Finnish Geodetic Institute has carried out geodetic operations at the Metsähovi research station since the 1970s. During the last ten years a number of new observation sites have been founded for GPS, SLR, VLBI etc. Some of these are located at Sjökulla, 3 km NNW from the original Metsähovi. Questions concerning the positions of the points relative to each other are frequently asked. Both terrestrial and GPS measurements have been used to determine the centering elements and connecting vectors, and measurements have been repeated to detect possible movements of the benchmarks. A brief summary of the observation sites and the local measurements carried out between them is now given.

The Finnish Geodetic Institute has established a network of 12 permanent GPS-stations in Finland (FinnNet). The planning of the network started during winter 92/93. The first stations were installed in 1994 and the last one was built in the autumn of 1996. The stations are equipped with geodetic Ashtech Z-12 receivers and Dorne Margolin type antennas. All the stations are working fully automatically. The network is used as a 1st order reference frame in Finland for GPS users, for crustal dynamics studies and for providing RTCM corrections in real time applications.FinnNet is used as a reference frame or as a source of data for numerous national and international GPS campaigns and projects like DOSE (Dynamics of the Solid Earth), BSL (Baltic Sea Level), BIFROST (Baseline Inferences for Fennoscandian Rebound Observations, Sea Level and Tectonics), EUVN (European Vertical GPS Reference Network), and EUREF permanent GPS Network.We will introduce FinnNet and its function in Finland, present the first results of the EUREF-FIN campaign and give an overview of the projects FinnNet has taken part in. We will discuss in general how permanent networks can be used in regional GPS measurements and call into question the function and need of large episodic GPS field campaigns in the future.

The existing old height system of Switzerland (LN02) consists of purely levelled heights without considering gravity measurements. This causes difficulties in the application of modern techniques such as GPS for height determination and in the data exchange with neighboring countries. Therefore, a new height system LHN95, based on geopotential numbers, is being determined at the Federal Office of Topography. LHN95 is treated as a kinematic network with consideration of the Alpine uplift. The corresponding relative vertical movements are determined by repeated observations of the first and second order levelling lines. For LHN95, it was decided to use orthometric heights. This has some advantages when combining levelling with GPS measurements in mountainous regions.

The basic fundaments of the fiducial and the non-fiducial approaches are presented. The main advantage of the non-fiducial approach over the fiducial one is the simplicity of computation when the choice of the known stations change. In the traditional fiducial technique all main steps of the processing must be repeated. An example involving the São Paulo State Network, which is composed of 24 stations (plus 3 fiducial stations) is presented. Results obtained from both approaches are quite similar.

The STATEFIX project, commissioned and managed by the West Australian Department of Land Administration, represents the densification of the Australian National Network in West Australia to an average spacing of approximately 200km. Some 230 dual frequency baselines (mean baseline length 202km) were observed between March 1996 and November 1996. Data processing and network adjustment were undertaken at Curtin University of Technology. Final results indicate the network is accurate (relative to the ANN) to better than 2cm in the horizontal component (95% confidence), and 6cm in the vertical component (95% confidence). The STATEFIX project has been particularly challenging in terms of the vast area surveyed (over 2,500,000km2) and the isolated nature of many control points, located in uninhabited desert areas. This paper will outline STATEFIX observation, processing and network adjustment strategies, and describe the precision and accuracy analysis performed for the network.

The principles of the new National System of Geodetic Coordinates are the result of analyses carried out by the study group of the Section of Geodetic Networks and the Section of Cartography Geodetic Committee of the Polish Academy of Sciences. As the basis of the new National System of Geodetic Coordinates the European Terrestrial Reference System (ETRF89) and ellipsoid GRS-80 were adopted. Practical realisation of the new reference system was carried out in the Department of Planetary Geodesy, Space Research Centre using GPS technique. In 1992 precise zero-order network EUREF-POL’92 containing eleven stations was established. In 1994 and 1995 its fürther densification (356 points) provided POLREF network which is practical realisation of ETRF’89 Reference System in Poland. The paper gives a survey of all works connected with the establishment of the new National Geodetic System in Poland.

A general description of the South American Geocentric Reference System (SIRGAS) Project is presented. Since its establishment, in October 1993, the two working groups (“Reference System” and “Geocentric Datum”) of the project have carried out the tasks towards defining a new geocentric reference system and an associated Datum for South America. The achievement of the project objectives were scheduled to be reached on 1997 and the corresponding final reports presented at this LAG Scientific Assembly. This was only possible due to the very high cooperation of many groups, as: the South American countries; IAG, IPGH and NMA, the sponsors; and many others from North America and Europe. The development of the project was based on a continental GPS network, established during a very successful campaign in 1995. With its conclusion, South America has now an unified reference system, with an accuracy compatible with the up to date positioning techniques, what is of a great benefit to all countries in the continent.

This paper describes the different stages executed to establish the Reference System to be used in the definition of the geocentric datum in South America, in the frame of the SIRGAS project.The Working Group I developed the activities to plane, to prepare and to coordinate the GPS measuring campaign, besides the processing organization to be executed by the calculating centers. The analysis and decision about the final solution, represented by the geocentric coordinates of the GPS network, are the last phase of the work.The GPS observations were carried out from May 26 to June 04, 1995. 58 principal stations were observed in 11 countries. The campaign was possible due to the collaboration of around 30 participant institutions from South, North America and Europe. DGFI (Germany) and NIMA (USA) were the processing centers.

The geodetic reference system for South America established by the SIRGAS project and adopted by the South American countries provides a set of some 60 stations coordinates over all the continent for the epoch of its observation in May/June 1995 (1995.4). To preserve the consistency with the ITRF, station coordinates have to change in time according to the ITRF kinematic reference frame. The motion of the South American continent cannot completely be modelled by a rigid plate rotation after current models but has to account for extended continuous deformations, in particular along the Andes mountain range. It is proposed to determine individual station velocities of all SIRGAS sites and to derive a continuous velocity field from a combined rigid plate and regional deformation model for the transformation of new stations into the SIRGAS reference frame.

For establishing an accurate and geocentric reference system for South America, Sistema de Referencia Geocéntrico para América del Sur (SIRGAS), a continental GPS campaign including more than 60 stations has been carried out in May/June 1995. DGFI/I was acting as one of the analysis centers processing the entire data set. The paper describes the applied processing strategy and discusses the achieved results.

Cooperating as part of South American Geocentric Reference System (SIRGAS), whose purpose is to establish a geocentric reference system for the whole of South America, Brazil will take the next step to continue this change.The integration of Brazilian Geodetic Network to SIRGAS will be carried out through a simultaneous adjustment of the network using the original data and the software Geodetic Adjustment using Helmert Blocking of Space and Terrestrial data (GHOST). In the adjustment will be adopted the GRS80 elipsoid and the Horizontal Network will be constrained to fit the coordinate values of the eleven SIRGAS stations in the national area.The installation and operation of Brazilian Network for Continuous Monitoring of GPS (RBMC), concluded in 1997, will provide a continuous link to SIRGAS for future users.

The new Venezuelan national geodetic network (Red Geodésica Venezolana, REGVEN) was observed nearly simultaneously with the GPS campaign for the establishment of a geocentric reference system for South America, Sistema de Referencia Geocéntrico para América del Sur (SIRGAS). This paper describes the REGVEN data, its processing and the connection to the SIRGAS reference frame. The accuracy of the geocentric station coordinates is in the order of ±2 cm.

Between 1993 and 1994 the POSiciones Geodésicas ARgentinas (POSGAR) GPS network was observed, and early in 1995 the POSGAR’94 reference frame was established. It consists of 127 evenly distributed stations providing a relative precision better than 1 ppm in the WGS84 reference system.Since 1995, the DGFI/1 and the FCAG are working together to obtain a Geocentric Reference Frame for Argentina, consistent with the one realized by the SIstema de Referenda Geocéntrico para América del Sur (SIRGAS) network.This work, which involved a cooperation in the SIRGAS reference network computation and then the re-processing of Argentina’s reference network is scheduled to be finished in early 1998.

IBGE — the institution responsible for the Brazilian Geodetic System (BGS) worked for more than ten years in order to readjust the observables belonging to this system. This work was finished at the end of 1996. The main reason for this paper is that there is the need for complementary analysis of the results. A few aspects related to the new realization of the SAD-69 in Brazil will be shown, taking into account its accuracy and aspects related to the conversion between the new and the old realizations. For the last case, a studied was carried out by using either the seven or three parameters transformation model. A new proposal for the transformation of coordinates is presented. Further investigations should still be carried out. The analysis of the accuracy was based on 34 GPS stations, which are coincident with the conventional network, being their coordinates considered as true values. A total of 34 adjustments were carried out, in such way that the GPS baseline vectors of each one of the stations were removed in the process. The estimated coordinates were compared with the true ones. The results shown an improvement of the new realization. For a more conclusive analysis, fürther GPS stations would be required.

In response to IGS’s call for participation in the pilot project for the densification of the ITRF, DGFI/I is acting on behalf of the SIRGAS project as a RNAAC for South America. The analysis results of the first year of activity are presented here in terms of the internal stability of the station coordinates and the discrepancies with respect to global solutions. The r.m.s. deviations for both internal and external comparisons are in the centimeter level.

In 1908, a systematic geodetic surveys started with cadastral and mapping objetives. Classical operations were carried out along main paralel and meridian chains till 1960’s when the Fundamental Geodetic Network was finished.A common adjustment using the variation coordinates method was performed in 1963, by the US Army Map Service. The whole 1st. Order data available was evaluated giving a mean direction error of 0″.71 and 0″.89 misclosure triangle. The data included 15 bases, 12 astronomical stations and 248 geodetic stations. Hayford Ellipsoid was used and the Yacare Datum was established.After that, normal geodetic densification of the 1st. Order was performed. The introduction of the EDM in the 1960’s, allowed for electronic polygons, getting field operations faster than classical method.

In this contribution an overview is given of the capabilities of methods for ambiguity resolution in use nowadays. This is done in the context of SSG 1.157, “Ambiguity Resolution and Validation”, although the overview is not restricted to the algorithms used within this group.Methods from the fields of navigation, rapid static surveying, and regional networks are compared, and similarities and differences are indicated. It is shown that the subdivision of the algorithms by their applications is a rather arbitrary one, and that some of the ambiguity resolution algorithms can be applied in all three fields.The capabilities of state of the art algorithms for ambiguity resolution are indicated in terms of maximum distances over which the ambiguities can be resolved and validated. In particular the role of the observational time span and the set of observables used in the estimation (single versus dual frequency, phase-only versus phase and pseudo range) is addressed. As the ionospheric delay is an important reason for being unable to resolve the ambiguities, the consequences of the next solar activity maximum at the end of the century are briefly discussed.

The improvement in precision obtained from GPS observations over recent years has revealed problems related to the local conditions at the GPS sites. In order to fürther improve high precision GPS positioning, orbit determination, and estimation of atmospheric parameters, investigations of site dependent effects are required. Concerns have been raised regarding the antennas and the monuments used and the long- and short-term mechanical and electromagnetic stability of the sites. Here, we review the problems associated with site-specific errors and present recommendations on how to eliminate or minimize these effects.

In order to estimate the characteristics of GPS antennas this paper describes a new approach for the estimation of absolute phase center variations (PCV) in a field calibration. The main objective of this approach is the elimination of multipath from the GPS observable and the elimination of any influence of the reference antenna. This is achieved by forming the so called mean sidereal day time difference between observations of successive days. Since the satellite geometry of GPS satellites repeats every mean sidereal day multipath does as well. Therefore, by differentiating observations of two different days multipath can be eliminated. To gain information about the PCV special rotations have to be applied to the antenna of interest. The paper shows the results for three different antenna types.A second application of the mean sidereal day time difference shows the estimation of small position changes between consecutive days. The mean sidereal day time difference is formed in the same manner as for antenna calibration, however, without any rotation. Therefore multipath and PCV are eliminated from the observable. We achieved with observation periods of 10 minutes a horizontal position resolution of only 1.2 mm which is better than results from standard data processing.

Since January 1994 when the International GPS Service for Geodynamics (IGS) became an IAG service, several classes of IGS products have been generated on daily and weekly basis. IGS orbits, clocks and EOPs, based on weighted averages of the seven IGS Analysis Center (AC) daily solutions, are among these products. Throughout 1996 and 1997, a number of improvements, both in quality and product delivery times were realized resulting in enhanced quality of IGS combined products. The Final and Rapid orbit, clock and EOP delays were reduced significantly on June 30, 1996. Delays were reduced from one month to eleven days for the Final products and from eleven days to 24 hours for the Rapid products. New improved modeling and conventions, such as subdaily EOP modeling, ITRF94 and EOP rate solutions, were introduced on June 30, 1996 by all seven IGS ACs. These enhancements, along with the improved geometry of the IGS tracking network, resulted in steady orbit, clock and EOP solution improvements. The IGS combined orbit, clock and EOP products are currently approaching the 5 cm, 0.5 ns and 0.1 mas precision level respectively. Since March 1997, a new IGS orbit prediction product, also based on a weighted average, has been introduced. The new IGS orbit prediction is available in real-time and is significantly better than the broadcast GPS orbits. Typically, the IGS orbit prediction, when compared to the IGS Rapid orbits, is at or below the 1 metre precision level, while broadcast orbits compare at the 3–5m RMS level. Also, in March 1997, a new IGS LOD/UT1 combination based on a weighted average of AC LOD solutions was officially introduced. Encouraging results were obtained and are presented hereafter.

Since January 1995 GPS observations performed by Polish IGS network stations have been used for monitoring the ionosphere. During January-May 1995 we revealed a number of violations of the ionospheric structure from regularity. The disturbances resulted from magnetic storms with sudden onsets. The storms caused an increase of ionospheric delay by a factor of 1.5–3.0. During disturbances gradients and TID’s effects are also increased. We found that such severe ionospheric conditions have an effect on the determination of GPS equipment biases. The biases determined from code data for disturbed days are different from those for quiet days. The differences can reach up to 0.3–0.5 m in differential delay. It can be expected that the effects will be even more appreciable during the next solar maximum.

An integrated GPS monitoring system has been in operation in the about 10-km2 study area at Olkiluoto, Finland since October, 1994. The system includes a permanent GPS station and a local GPS monitoring network. The permanent station is used to investigate regional crustal deformations in Finnish territory, while the local monitoring network is used to monitor relative deformations within the study area. The analysis of 2.8 years of continuous GPS tracking data shows that the horizontal components of the regional crustal deformations are about 2 mm/yr, while the vertical component is about 8 mm/yr. The local GPS monitoring network has been measured six times at half-yearly intervals. The horizontal components of the local movements obtained from the repeated measurements are less than 1.0 mm/yr. The GPS result indicates that the investigation area at Olkiluoto is rather stable.

It is well known that the variance covariance matrices of GPS baselines (vectors) as obtained from the processing of GPS baseline observations are in general too optimistic. Therefore, it is a common practice, in the adjustment of GPS networks, to scale up the variance covariance matrices by for example multiplying the matrices with a constant factor. Though the method can always manage to get an adjustment to pass the necessary statistical tests, the method is not rigorous and often arbitrary.This paper uses the newly completed Western Australian state wide GPS network STATEFIX as an example to look at the effects of the different ways of determining the scale factors for the variance covariance matrices of GPS baselines. The method of variance component estimation is used to assist the analysis. It is found that the adjustment results can vary significantly when different methods are used to determine the scale factors. For most GPS networks, it is recommended that a scale factor be estimated for each independent GPS observation session using the technique of variance component estimation.

The surface movements of a landslide involving an urban area of the Assisi town, in central Italy, have been measured since 1995 using the GPS technique. This paper presents some design aspects of the GPS network, and a statistical analysis of three annual survey campaigns.

A regional ionospheric model was developed using data from dual-frequency GPS receivers aiming to correct single frequency observations. The data analyzed correspond to an intermediate southern latitude band in a period of stable ionospheric conditions. The efficiency of the ionospheric model was tested by comparing L3, L1 plus the ionospheric model and L1 solutions. This was done using different numbers of GPS stations, and different session lengths were used. These comparisons were performed by similarity transformations, paying particular attention to the scale factor. Advantages and limitations of the model are discussed.

A linear combination functional model, formed from the single-differenced functional equation for baselines from a mobile receiver to three or more reference receivers, is proposed, in which the orbit bias and ionospheric delay can be eliminated, and, in addition, the tropospheric delay, multipath and observation noise can be significantly reduced. As a result, the ambiguity resolution technique that can be employed for medium-range GPS kinematic positioning is similar to that used for the short-range case. Kinematic tests have been carried out in Sydney, Australia, with separations from the nearest reference stations greater than 30km. The carrier phase ambiguities can be resolved for every epoch and the success rate (correct identification of the integer ambiguities) was 100%. This technique is well suited to real-time precise GPS kinematic positioning.

During the SIRGAS 95 GPS campaign several sites were continuously occupied during ten days by two different receiver systems. From the data acquired at all those colocation sites providing sufficiently accurate ground truth, either from terrestrial geodetic observations or from SIRGAS independent GPS measurements, the elevation dependent phase center differences between the involved antennas have been determined. The modelling approach and the achieved results are presented.

The ionospheric delay is one of the main error sources for precise GPS positioning and navigation. Dual-frequency GPS receivers can be used to eliminate the ionospheric delay (to the first order) through a linear combination of L1 and L2 observations (Hofmann-Wellenhof et al., 1994). Taking advantage of this physical dispersive principle, one or more dual-frequency GPS receivers can be used to determine a model of the ionospheric delays over a region of interest and, if implemented in real-time, can support single-frequency GPS positioning and navigation applications.There are, however, several factors which may affect the accuracy of real-time ionospheric delay estimation. One of the dominant factors, the failure in the carrier phase observations, such as the occurrence of cycle slips (for example, Engler et al., 1995), is the topic of this paper. In order to improve the real-time ionospheric delay estimation performance a novel approach to addressing this problem is presented here.The concept and methodology of the proposed algorithm is introduced. Preliminary test results are presented to indicate the performance of this approach for real-time ionospheric delay estimation.

GPS levelling was used in two test areas in Finland in order to investigate the capability of GPS measurements for orthometric height determinations. Both test areas are surrounded by the spirit levelling loops of the Precise Levelling network. Besides, the test areas are covered by a dense network of lower order levelling lines that are connected to the Precise Levelling network. The GPS networks measured in both areas consisted of 51 and 45 GPS sites that were located at spirit levelling benchmarks. The average side length was 15 km.Four different geoid models, two global models (0SU91A, EGM96) and two local models (NKG89, FIN95), were used to convert the GPS ellipsoidal heights to orthometric heights, after which the results were compared to the spirit levelled heights. In all cases the height differences showed a clear tilt which was removed by fitting first and second order surfaces to the differences. When the global geoid models were used for the conversion, the RMS of the height differences varied from ±40 mm to ±80 mm. The RMS of the differences was reduced to appr. ±15 mm when the local geoid models were used. When the errors caused by the spirit levelling in the computed height differences were removed, the estimation of the average accuracy of the GPS levelling on the test areas gave the result ±12 mm.

There has been a burgeoning of activity involving the establishment and operation of permanent GPS stations. The variety of applications is broad, from geodetic reference sites in support of IGS initiatives, to national base stations for real-time differential GPS, “integrity monitoring” and datum maintenance, and even local GPS base stations providing data to surveyors and other users. The level of sophistication of such permanent stations also varies, from systems capable of automatic data collection, transmission and sometimes processing, to relatively simple configurations. All applications, however, share a concern for GPS data “quality”. The detection of cycle slips, data “spikes”, multipath disturbances, and bad data sequences of GPS measurements is a major problem for many applications, even more so in the case of real-time applications.Some quality control (QC) strategies may be implemented either on a single receiver basis or using GPS network data. This paper discusses the issue of QC for permanent GPS receivers, and comments on aspects of QC which require investigation.

This contribution presents and discusses practical results of network-based instantaneous ambiguity resolution. It is shown that high-dimensional integer solutions can be computed efficiently with current ambiguity resolution methods. For that purpose the least-squares ambiguity decorrelation method has been integrated in the network analysis. We also introduce a new ambiguity dilution of precision measure to express the intrinsic precision of the ambiguities and show why it is preferred over the more traditional measures.

High precision GPS kinematic positioning in the post-processed or in the real-time mode is now increasingly used for many surveying and navigation applications on land, at sea and in the air. The distance from the mobile receiver to the nearest reference receiver may range from a few kilometres to hundreds of kilometres. As the receiver separation increases, the problems of accounting for distance-dependent biases grow and, as a consequence, reliable ambiguity resolution becomes an even greater challenge. In this paper, the challenges, progress and outlook for high precision GPS kinematic positioning for the short-range, medium-range and long-range cases, in both the post-processing and real-time modes, will be presented.

The accuracy of trajectory determination by differential GPS (DGPS) is limited mainly by receiver noise, atmospheric errors, orbital errors and multipath. Independent estimates of a trajectory can be obtained if several reference stations are used to generate differential corrections. Using these estimates and an adaptive filtering system it is possible to estimate, characterize and remove some of the errors mentioned above.The objective of this paper is to introduce the adaptive filter as a means of estimating and reducing the errors that influence the accuracy of a trajectory determined by DGPS. An analysis of trajectory estimation for an aircraft using a multiple DGPS receiver configuration and an adaptive filtering system is described. Data collected during an airborne survey carried out over the Canadian Rocky Mountains is used in the analysis. Four ground-based reference stations were spread out over approximately 100 km, providing independent estimates of the trajectory of the aircraft.It is briefly reviewed how an adaptive filter can be used to estimate the correlated part of two discrete-time signals. Since the true trajectory of the aircraft is the underlying signal in each of the four estimates, the adaptive filtering system proves to be a useful tool for the determination of that trajectory and the effect of the errors that influence it. Examples from the analysis are used to demonstrate the effectiveness of the technique.

When post-processing dual frequency carrier phase data, the residual tropospheric delay can easily be the largest remaining error source. This error can contribute a bias in height of several centimetres even if simultaneously recorded meteorological data are used. This is primarily due to the poor representation of the water vapour profile in the tropospheric delay models. In addition, a lack of real-time meteorological data would force the scaling of either surface values or standard atmosphere values; these are also unlikely to accurately represent the ambient atmosphere.To obtain the highest precision in kinematic GPS some advantage may be obtained by estimating this error source along with the position solution. The simple tests reported in this paper removed biases of several centimetres in height when estimating the residual tropospheric delay from GPS data recorded at an aircraft in flight. However, important limitations exist in the geometry of the satellite coverage which must be considered before the full reliability of the technique can be quantified.

A fully digital Airborne Integrated Mapping System (AIMS) for large-scale mapping and other precise positioning applications is currently under development at The Ohio State University Center for Mapping. AIMS is designed to be installed in an aerial platform and incorporates state-of-the-art positioning (differential GPS integrated with INS) and imaging (CCD) technologies. Preliminary test results show that platform position and orientation can be obtained with an estimated accuracy better than 7 cm and 10 arcsec, respectively, over long baselines.

Preliminary results obtained from the analysis of GPS data collected during an airborne photogrammetric project conducted in 1996 over the Island of Vulcano are described. The objective of this study is to determine an optimal processing procedure for obtaining high accuraacy projection center positions for large scale aerial photography applications. GPS measurements were collected using a multi antenna configuration both on the aircraft and on the ground. The performance of standard software for OTF differential kinematic processing was evaluated. In order to use redundant information both rigorous combination of independently computed solutions and simultaneous adjustment of GPS observation from multiple reference stations and mobile receiver were performed. GPS-derived camera projection centers are compared to the results from a aerotriangulation adjustment results. The results indicate that the GPS data can be helpful to preserve the accuracy level when there is a lack of ground control points.

The aim of this paper is to show that GPS signal distortions can significantly influence the accuracy of the results of a GPS survey. Signal distortions may occur if the satellite signals propagate through a diffuse obstacle such as a bush and the distorted carrier phases arrive at the GPS antenna. As a result errors in the horizontal position of up to 2 cm and in the vertical position of up to 5 cm can occur. The results of several experiments will be shown. Different GPS choke ring antennas and receivers as well as different GPS processing software packages were used in order to prove that this effect is hardware and software independent. The signal distortion was evident in all experiments. The detailed investigation of the double difference phase data allowed the removal of the affected data which subsequently yielded the correct GPS results.

To analyse and test the feasibility of GPS-aided approaches for civil aviation in the Swiss Alps, an internationally co-operating group, leaded by the Swiss Federal Office of Civil Aviation (FOCA) has been set up. During several test flights and dedicated ground missions various problems, such as satellite visibility, multipathing, GPS signal interference and the influence of the topography on the navigation and the GPS system were addressed. The tests showed the potential of satellite based systems for approaches, in particular for use in rugged terrain.

In Europe the densification and improvement of the high speed railway network has been made a priority, as it is generally accepted that railroads are most efficient for transporting people. In the near future high speed trains will connect all major towns with a speed of about 200 km/h. The Dept. of Engineering Geodesy, Technical University in Vienna, was involved in the development of fundamental concepts concerning geodetic networks and rail track alignment methodes.The investigations have shown that the fundamental technical geodetic networks should be established in four steps. For a most economic organization of the measurement sessions simulation calculations have to be performed in advance. Approximate coordinates taken from a map and IGS orbit data were used.For the railway track alignment a carriage was constructed carrying the rover of a DGPS system and other electronic sensors. The alignment calculations can be performed on-line, using algorithms based on Wiener filtering procedures.

This article presents results of repeated GPS surveys on the Greenland Ice Sheet for studying ice elevation changes, and for providing ground truth data for evaluation of remote sensing methods (e.g. satellite altimetry and SAR interferometry). The surveys have been carried out at several central sites of the Ice Sheet, as well as on local marginal ice caps, which are more sensitive to coastal climate changes. Kinematic GPS techniques have been used for surface profiling using snowmobile traverses and airborne laser altimetry, providing good agreements at the sub-metre level. Static GPS repeated surveys yield sinking and strain rates which are in good agreement with glacilogical mass balance models, indicating no major shrinking or growth of the Ice Sheet.

The Global Positioning System (GPS) has been proved to be a capable surveying tool for monitoring the stability of steep walls in open pit cut mines. However, due to restrictions on satellite geometry and severe environment-dependent errors, as well as the uncertainty of the a priori knowledge of system noise variance, sub-centimetric accuracies using existing algorithms cannot always be achieved. This paper proposes a reweighted filtering algorithm which treats GPS deformation monitoring in an open pit environment as a kinematic surveying system. Estimating techniques of a posteriori weights of predicted states and observations are developed to match the weights appropriately and obtain reweighted filter solutions. Two different weight functions to reweight the predicted states and observations are established based on the state residuals and observation residuals. The proposed algorithm can automatically assign the larger weights to the more accurate predicted states and observations and the smaller weights to the less accurate predicted states and observations. Results from simulated GPS deformation monitoring data are shown using the proposed algorithm. Comparisons with the traditional Kaiman filtering method indicate high levels of filter stability and accuracy for the proposed algorithm.

Ground deformation due to volcanic magma intrusion, crustal motion, ground subsidence, etc., are phenomenon ideally suited for study using GPS. The change in length, height difference and orientation of baselines connecting GPS receivers in a carefully monumented ground network can be monitored. This is done by repeatedly measuring the same baseline components to an accuracy commensurate to, but preferably much higher than, the expected baseline component changes. Such GPS techniques are based on the “campaign” principle: the periodic (often annual) re-survey of a network of control points. However, over the last half decade or so, there has been a growing interest in the deployment of permanent, continuous GPS monitoring networks. The factors responsible for this trend include the declaration a few years ago of Full Operational Capability of the GPS system, and the steady decrease in price, size and power requirements of GPS receivers. Important additional factors have been the high cost of annual GPS surveys (manpower, travel, logistics, etc.), as well as the fact that geo-scientific research can be fürthered because of the continuous measurement of a deformation phenomenon, rather than its periodic measurement.The GSI network in Japan, the SCIGN network in California, and the SWEPOS network in Sweden are examples of large scale, continuous GPS networks for near-real-time crustal motion monitoring. However, smaller scale GPS arrays such as those on the Augustine volcano (Alaska), the Popacatepetl volcano (Mexico), the Kilauea volcano (Hawaii), and the Rabaul volcano (Papua New Guinea) reflect a growing interest in local continuous GPS volcano monitoring systems. GPS is also increasingly used to monitor engineering structures such as dams, bridges, offshore drill platforms, etc.Developing an automatic GPS array system for such small scale monitoring applications is an engineering and software challenge. A network of permanent GPS receivers need to be deployed, often in an inhospitable environment in which they must operate reliably on a continuous basis. The GPS observations must be telemetered to a central computing facility where data processing occurs with minimum delay. Analysis of the time series of baseline results then takes place in order to detect any baseline component change between successive solutions which may be a precursor to failure or eruption. What are the bases of small scale monitoring systems? Often the solution has been simply to purchase commercial “off-the-shelf” real-time-kinematic (RTK) GPS systems. This is the high cost option, yet there are several hundred active volcanoes in the world, many located in the less developed countries, and the cost of GPS monitoring systems must be significantly reduced if the technology is to contribute to volcano hazard mitigation. An alternative approach is to develop a system based on single-frequency GPS receivers, integrated with communications and in-field computer sub-systems.This paper discusses the characteristics of GPS monitoring networks and considers such issues as monument design, network design, GPS hardware, communication links, power supply, and data processing strategies.

Algorithms for On-The-Fly ambiguity resolution have been modified for the deformation monitoring of a suspension bridge. Instantaneous relative positioning of a deformation network at a precision of about ±5 mm horizontally and ±10 mm vertically has been achieved using GPS LI phase observations. Particular attention has been paid to the modeling of relative tropospheric delay and to the phase center calibration between antennas of different types. Moving averages on station coordinates have also been applied to reduce multipath effects.The Pierre-Laporte bridge is a 6-lane, 1040-m-long suspension bridge which crosses the St. Lawrence river in Québec City. Three 48-hour GPS sessions have been conducted during the months of July and October 1996 and February 1997. For each session, 5 GPS receivers were observing at a data rate of 2 seconds.Daily variation in the vertical position of the antenna located at the deck center shows clear correlation with respect to temperature and vehicle loading. Transverse movement of the deck center has been monitored and correlated with (transverse) wind speed. Seasonal variation in temperature caused a vertical displacement of the deck center, a contraction of the towers as well as a displacement of the towers towards the river banks.

The correct management of water supply, both surface and ground water, has tremendous environmental implications, directly affecting the quality of life of the population in general, with economical and political consequences. Very recently, part of the Karst aquifer system, which passes under part of the Brazilian Southern Region, has become under scrutiny for solving part of the problem of water supply for Curitiba and surroundings. The project for its digital mapping aims to correctly understand the processes governing this aquifer allowing for its use without depleting it.This paper describes the GPS component of this project, applied to a test area of around 40 km2. This component is basically composed of the surveying of the artesian wells and the definition of a local geoid. For the latter, the strategy adopted was to reposition two diferent geoidal maps, one given by the program MAPGEO from IBGE, and the other presented by Sá & Molina with a far greater resolution. Bench marks of the fundamental vertical network were used as constraints in the adjustments carried out for both geoidal maps and as control points (“ground truth”). Results indicate that a better local geoid was obtained based on the second geoidal map.This project presents an application of Geodesy to sustainable development.

In a kinematic GPS positioning system, state parameters are usually estimated using the Kaiman filtering method. It has been noted that reliable ambiguity resolution, and estimation of other state parameters are highly dependent on the correct stochastic models for the GPS double difference measurements. In this paper, a real-time statistical procedure to estimate the covariance matrix of GPS double differenced measurements is presented. Test results indicate that with the proposed procedure, the reliability of ambiguity resolution and kinematic positioning results can be improved.