The Implementation of a new, common height datum in Denmark
THE NEW UPDATED ISO STANDARDS CONCERNING LEVELLING INSTRUMENTS
HISTORY AND EVOLUTION OF HEIGHT DETERMINATION TECHNIQUES
The past and the future of levelling networks in hungary
ACCURACIES ACHIEVABLE BY GPS IN PRACTICAL ENGINEERING AND SURVEYING APPLICATIONS
Some questions of the precise height measurement
OCEAN TIDE LOADING EFFECTS ON HEIGHTS
Levelling by GPS: the state of the art in France.
Different solutions adopted to modernise the height networks in the Baltic countries
The Swedish geodetic networks today and in the future
REQUIREMENTS ON THE HEIGHT BENCHMARKS IN THE THIRD PRECISE LEVELLING OF SWEDEN.
Manufacturing of High Precision Levelling Rods
Multi-Sensor Systems for Height Determination
Sensitive high-speed-railway track control
Checking, Testing and Calibrating of Geodetic Instruments
HEIGHTING WITH GPS: POSSIBILITIES AND LIMITATIONS
Status of the European height systems UELN and EUVN
The evolution of digital levelling techniques - limitations and new solutions
State of the art and present developments of a general approach for GPS-based height determination
BEST USE OF DIFFERENT HEIGHT DETERMINATION TECHNIQUES
ERROR SOURCES IN HIGH PRECISION LEVELLING HOW TO MINIMIZE THEIR EFFECTS ON THE HEIGHTS
Feasibility test of GPS-heighting in the Hungarian leveling network
PRECISE LEVELLINGS IN FINLAND
DIRECT ACCESS TO THE FRENCH DIGITAL HEIGHT DATABANK VIA MINITEL
The digital production line used in the third precise levelling of Sweden
Motorized Trigonometric Levelling (MTL) for precise levelling – the Swedish tests and results
Experience of technical co-operation in Zambia and Mozambique
Carl Zeiss: The development of levels at Carl Zeiss during the past 25 years with special focus on the optical level NI 002 and the digital Level DiNiÒ 11
REQUIREMENTS FROM URBAN USERS OF HEIGHTS, EXAMPLES FROM STOCKHOLM
High precision GPS makes challenges to realisations of Reference Systems.
The National Land Survey of Sweden today and tomorrow
New Automated Real-time 3D total station, possibilities and limitations.
Use of GPS in unification of vertical datums and detection of levelling network errors
The influence of heights on datum transformations.
Height Control of Construction Plant by GPS and GLONASS
COMPUTATION OF POSTGLACIAL LAND UPLIFT FROM THE THREE PRECISE LEVELLINGS IN FINLAND
The Use of Mororized Trigonometric Levelling (MTL) in Denmark
The Changing World of Geodesy and Surveying
Some Trends in Geomatics - Multi-sensor Systems and Global Georeferencing.
Result of test and Experiments with SDL30 Digital Level
Heights and Vertical Control in the Czech Republic - Evolution and Present State in Context of Current European Projects
Present State of Standardisation in Surveying Profession
The Future of Vertical Geodetic Control
Approximation of local geoid surface by artificial neural network
On behaviour of invar rods with aluminium frame used in Third Levelling of Finland
EXPERIENCE WITH VARIOUS DIGITAL LEVELS IN BOTH MOTORISED AND CONVENTIONAL „ON FOOT" PRECISE LEVELLING IN EAST GERMANY
National report and activities in Denmark.

The societies of today are becoming more and more complex, integrated and international. The demands of standardisation are in general increasing, such as introductions of geodetic datums. The paper will focus on different society aspects needed to consider in implementing a new, common height datum in Denmark.

Inside ISO, the International Standard Organisation, several Technical Commissions (TC59/SC4 and TC172/SC6) have produced different standards concerning the measuring instruments used for levelling purposes. Unfortunately, these standards made for the same instrument and the same purpose namely "Field procedures for determining the accuracy of surveying instruments" are often quite different because of different approaches and goals. TC59 looked from the building constructing views and TC172 from the instrument manufacturer views.

Since 1997, a Joint Working Group (JWG) common for both TC works on a harmonisation and updating of these standards. The goal is one standard for one instrument. One of the projects concerns levels and is chaired by J-M Becker. A reviewed draft proposal has been discussed in Berlin the 4 mars 1999 after examination of the comments from the national standard organisation. This paper will present the results from the JWG concerning the field procedures recommended for the determination of the accuracy of levelling instruments for different applications.

The surveying profession has been subject to many important changes during the last decades. We have seen a rapid technological evolution especially concerning the surveying techniques and instrumentation used for different applications.

The time of surveying with purely optical and mechanical instruments (steel meter, levels, etc) has rapidly been replaced with more sophisticated surveying techniques and equipment like motorised total stations, GPS, etc. However this was mainly for the purpose of positioning in 2D, plan co-ordinates.

Attempts to automate and make the levelling process more efficient have also been going on for a considerable time through both method and instrumentation developments. Today we use a new kind of survey systems based on "Black box", "Push Bottom" fully automate and producing digital height data in real time like the Swedish Motorised Levelling technique (ML). In the following report the author will present some of the most important development steps during his live time and illustrate this with examples from Sweden like the different motorised height determination techniques.

Because of its historical situation, Hungary has started to establish levelling networks four times up to now. The first network was built on the territory of the Austro-Hungarian Monarchy from 1873 until 1914. The second network was finished between the two world wars, but it was not adjusted and used at all. The current national levelling network – which is used on the half part of the country now, – was developed between 1948 and 1964. From the 1970s we started to develop the so-called Unified National Vertical Control Network (UNVCN), that is suitable also for crustal movement analysis, but the work has not been completed yet because of financial reasons. Only the first order network is ready for the whole territory of the country, the second and third order ones (which are necessary for practice) are finished for only the half of the country.

Using the strict regulations of precise levelling, and thanks to the appropriate adaptation of the new technology, Hungarian experts produced very good results.

Between 1995 and 1997 a new GPS network with 10 km long baselines was established. A question occurred, weather GPS technique can help to increase the density of the levelling network or not. There are research works concentrating on that how precisely the heights of the third order points can be determined with the knowledge of the GPS ellipsoidal heights and the geoid model. According to the initiatives the third order levelling network will be developed with GPS technique.

The paper gives a report on the current status and introduces special Hungarian solutions.

GPS is rapidly becoming the default measurement tool for a very large number of standard civil engineering and land surveying operations. These include classical activities such as control surveys, topographical detail collection for large scale plans, setting out, and monitoring of deformation - as well as new developments such stakeless surveying (i.e. automatic guidance for civil engineering plant) and futuristic applications such as on-site robotics. For all of these applications knowledge of the accuracy being delivered by GPS is crucial. It is, however, only very rarely the case that the quality measures output by the GPS data processing component of the operation (either in real-time or in post-processing mode) properly describe the real quality of the results.

The most significant error sources in some of the major GPS applications are reviewed and the critical factors for each application are highlighted. Emphasis is placed on the problems associated with multipath error estimation and analysis, as it is usually the single-most important problem in kinematic GPS applications. Typical sizes and patterns of multipath errors are presented and methods for modelling it in real time are reviewed. Finally a method is discussed that can, for slow moving antennas, reduce such errors by about 50% of their normal size and some possible engineering applications of the method are highlighted,

Evidence is presented for strong spatial correlation between phase measurements made to different satellites and it is shown the fact that this correlation is ignored, which is normal practice in GPS data analysis, is a major factor in the poor performance of current quality assessment methods and real-time quality control measures. A general discussion of the role of mathematical modelling in GPS data analysis is included.

Finally the issue of performance calibration of GPS systems is briefly discussed. This is a topic that has received little attention to date but is of increasing importance. It is easy to compare the performance of traditional surveying instruments such as EDM where accuracy and range are usually simply related (and where the environment is of rather little importance). For GPS there are many other factors to consider, such as time to first fix (for RTK), range for guaranteed ambiguity determination, performance in partially masked environments (e.g. under light tree canopies), susceptibility to multipath and accuracy in different kinematic scenarios. Recommendations are made for some of the components of a standard testing facility to compare the performance of commercially available (and other) GPS systems.

There are research works for many decades in the College for Surveying and Land Management, in connection with height measurement. In my paper I’d like to emphasise some results of these works.

20 years ago, when the crustal movement network had been under development, an idea occurred, to build a ring-shape polygon keeping away levelling lines from the town, because the levelling measurement in the inner city results worse precision due to the traffic. Around town Székesfehérvár we created a ring-shape polygon, in order to determine a higher precision. In this way the errors of the town levelling don’t influence the national crustal movement network.

During establishment of the Bodajk-Csókakõ test levelling line the measurement readings were collected directly into the hand computer. This made us possible to store more data and more quickly than manually. Because of the data processing was made by computer we could calculate some confidence factors too. We determined the standard deviation of the measurement for one point using all measured data, the average rate of instrument subsidence, the duration of the measurement for one station, the mean error of the one km levelling. This was the first time we used digital level NA3000.

The unique task is the levelling across the wide river: we made it across the river Tisza around town Szeged. We’ve developed a new kind of measuring system, which gave a more precise and faster solution than the previous one.

Another kind of accurate height measuring is the trigonometric levelling. We worked out an accurate table and method to reduce the mean error of height differences.

I’d like to mention two new point of views according to the data processing. When we model the building-subsidence using repeated precise levelling, it is worse to calculate the parameters of adjusting plane. We can separate the movements of the determining parts from the residuals. The residuals are represents the deformation of the foundation body.

When we model the surface deformations from repeated height measurements, the traditional way is to compare two different networks; the differences can be considered as the movements. We suggest the new approach: let’s determine the deformations directly based on the original measurements (as a surface from height differences) without building up two networks. This makes determine the moving surface more accurately. The changes between two measurements we can interpret as the changes of the surface tangent.

The paper outlines the theoretical principles behind Ocean Tide Loading (OTL) before concentrating on the practical implications for GPS positioning, giving examples of loading effects at both global IGS sites as well as at European Tide Gauge GPS stations. The paper describes how, with modern ocean tide models, it is possible in most areas to accurately model the loading effect, and also demonstrates how neglect of the effect can lead to aliasing of results, particularly when using short, episodic measurement strategies. This latter effect can be particularly important when assessing time series of height observations.

For GPS levelling applications, it is convenient to express accurately the height reference surface in a suitable geodetic reference system. This can be obtained through a set of levelled GPS points which realises efficiently for the end-user both the geodetic reference system and the height reference system. Unfortunately, available data are sparse. The paper shows how a gravimetric geoid can be used to interpolate the height reference surface issued from GPS and levelling. Both surfaces do not coincide exactly with each other for several reasons, which are explained. At this point, one must compare two realisations of the geoid, detect outliers, retrieve (if possible) the causes of the discrepancies and finally combine the two kinds of data. The paper presents a method to reach these objectives and how it was applied in France. Numerical tests show that levelling by GPS is feasible by standard methods, leading to a precision of 2-3 centimetres.

The history of precision levelling in the Baltic countries dates back to the second half of the 19th century. The analyses of high-precision re-measurements and concurrent studies performed by different institutions have identified the character and velocity of recent vertical crustal movements on the territory of the Baltic countries.

Considerably high values of vertical crustal movements (in different regions from -4 mm up to +2.5 mm/year) and destruction of the considerable amount of benchmarks require a continued and all-round attention to the problems of the height networks.

New political status of the Baltic states, participation in international study projects (BSL and EUVN GPS campaigns), implementation of new technologies and skills have created a need as well as possibilities for a co-ordinated improvement and modernisation of the height network in the Baltic region.

In Sweden there are hundreds of geodetic control networks because every city and municipally has more or less there own network. This paper gives a historical review of Swedish control networks at different levels and discusses some problems connected to this situation. It presents some ideas how a nation wide network of permanent reference stations for satellite techniques could act as a control network in combination with conventional networks.

The development in the field of surveying is very intensive today. The instrumentation as total stations, levelling instruments and GPS-equipment is constantly improved in order to be easier to use and to achieve better accuracy of the measurements. Calculation programmes are also improved for the same reasons.

However, most measurements are related to some kind of reference points in the terrain that should carry the coordinates or the heights calculated from the measurements.

So what happens if the benchmarks does not meet the standard of the measurements is that the good measurements performed are loosing a bit of its value. The efforts made to achieve high-class measurements are more or less wasted. Or can we really afford to make those accurate measurements just for the pleasure to see that nice low RMS in the result files from the calculation?

Not only the stability and accessibility of the benchmark is important, but also to be sure that the point used in the field is the right one. To guarantee that, it is necessary to have a reliable identification and documentation of all the points.

The quality of a measurement can easily be judged by calculation, and we can accept the measurement or remeasure, but how can the quality of a benchmark be judged? Unfortunately this can usually not be done until the point is used the next time, perhaps after several years, and then it is too late.

This paper mainly describes the requirements on the benchmarks in the third precise levelling of Sweden and what is done in order to meet those requirements. The actions taken here can however be applied to most kinds of networks.

In this paper, we describe the process of manufacturing and calibration of high precision levelling rods.

At NEDO in co-operation with University of Karlsruhe, Germany, a new method of manufacturing high precision levelling rods has been developed. Based on a type of steel which is characterised by a very small thermal length extension coefficient (a £ 1 × 10^-6 × K^-1) this new method was established. The basic idea is to use a high energy CO₂ laser source to generate markings on the surface of these steel tapes. Therefore, the tapes are first sprayed with black colour, then afterwards sprayed with yellow colour so that only yellow is visible on the surface. The high energy laser is used to remove yellow colour in those regions which are expected to be marked as black, either for barcode or standard scales. The light of this high energy laser is therefore formed by masks and focused by a complex system of lens.

To enable manufacturing of high precision levelling rods with lengths up to 3m, the steel tape is moved along the high energy laser at constant speed. By using a very high precision interferometer based calibrator, it is possible to measure the position of the steel tape in comparison to the high energy laser. A complex computer system uses the output of the calibrator and generates control signals for the high energy laser so that finally a black marking can be positioned very precisely.

A lot of experiments were necessary to get the best combination of parameters like thickness and type of colours, energy of CO₂-laser, optical values and masks geometry. Much effort was necessary like soft- and hardware development for the overall control of this manufacturing process. To ensure high quality, several steps has been introduced. First of all, a new kind of calibration method was developed at University of Karlsruhe which uses an electronic microscope to detect the beginning of black marked areas. In conjunction with an interferometer based calibrator like it is used in the process of production, it is possible to check precision of our levelling rods. This method has been improved by University of Munich (TUM), Germany, to enable automatic calibration of levelling rods under changing atmospheric conditions. Permanent calibration of our levelling rods at TUM ensure a high quality standard which results in random errors less than 0.007mm and a very stable thermal behavior.

NEDO is currently working on new methods of quality control using image processing and pattern recognition e.g. to detect missing or misplaced markings during manufacturing process nearly under real time conditions.

Over the past several years, extraordinary achievements have been made in the use of GPS for a wide range of applications, e.g., navigation/aviation, mapping, fleet management, surveying, meteorological monitoring and precise farming. A key element in the rapid growth and success of these applications has been the integration of GPS with other enabling technologies/sensors. Such growth and success is expected to continue in the next few years as accuracy and integrity from GPS are further improved. This paper describes the basic features of multi-sensor systems and their role in the height determination. Emphasis will be given to systems that integrate GPS with other technologies such as INS, barometer, gravimeters, and vision based systems. The conceptual system differences are compiled and accuracies are classified according to their error characteristics. System integration advantages and disadvantages are discussed. Some test results are also presented to support the analysis described in this paper.

e-mail: grafarend@gis.uni-stuttgart.de

High-speed-railways / bullet trains / Transrapids / need „extreme reliable control systems" (extasy) to avoid catastrophic events. Here we consequently focus on two sensitive problems related to extasy: (i) a local high resolution representation of the track design (clothoid, circle, straight line) in UTM map matching coordinates is given. (ii) the Mixed Model (universal Kriging) is used to discriminate measurement errors from track displacements. (E. Grafarend and B. Schaffrin: Ausgleichungsrechnung in linearen Modellen, B.I. Wissenschaftsverlag, 483 pages, §2f, Mannheim 1993)

Checking, testing and calibrating of measuring instruments are traditional activities of all engineers engaged into meteorological processes. The necessity of these investigations is justified by the liability to produce measuring results of best accuracy, according to the measuring task. The surveyor as meteorologist had as well the responsibility to follow these procedures. But in the last years new technologies and the increasing complexity of measuring systems demand completely new methods for investigating or checking these instruments. Additionally new aspects concerning the quality management system (ISO 9000 ff) have to be considered. The paper proposes some general recommendations and concepts of this special field, which will consider not only technical and economical constraints but also the requirements of an international recognised quality system.

Global Positioning System (GPS) surveying is now seen as a true three dimensional tool and GPS heighting can be a viable alternative to other more conventional forms of height measurement. This paper examines the limitations and possibilities of GPS heighting. The first part of the paper details the limitations of GPS heighting, including those factors that affect the GPS height measurement itself and the associated issues of geoid modeling and compatibility with the local vertical datum. The second part of the paper examines the possibilities for GPS heighting, focusing on three application areas currently generating interest for the practicing surveyor; deformation monitoring, real time GPS surveying and machine monitoring and guidance. These applications cover the range of achievable GPS heighting accuracy.

After a break of ten years, the work on the United European Levelling Network (UELN) resumed in 1994 under the name UELN-95. The objectives of the UELN-95 project being to establish a unified vertical datum for Europe at the one decimeter level with the simultaneous enlargement of UELN as far as possible to include Eastern European countries. More than 3000 nodal points were adjusted, constraint-free, in geopotential numbers linked to the reference point of UELN-73 (gauge Amsterdam). The new heights in the system UELN-95/98 are available for more than 20 participating countries.

The European Vertical GPS Reference Network (EUVN) is designed to contribute to the UELN project along with the connection of European tide gauge benchmarks as contribution to monitoring absolute sea level variations, the establishment of fiducial points for the European geoid determination, and the stepwise development of a European kinematic height reference system. The EUVN includes 195 points all over Europe. At every EUVN point, three-dimensional coordinates in ETRS89 and levelling heights primary in the system of the UELN-95 have to be derived. The GPS computations are finalised, though some levelling connections still have to be realized. At the tide gauge stations of EUVN additional sea level observations have to be included.

In the early nineties, as a result of long-time research and development activities, digital levelling has been introduced to geodesy. During the last years, four manufacturers have developed digital levels using different code solutions and image processing methods. At present, digital levelling is the standard in precise height determination and in other levelling tasks.

The present paper gives an overview of the various developments and an outlook to the future technologies in precise height determination, based on optical procedures. Obviously, the new levelling technology permits new applications such as permanent height monitoring with additional motorization of the instrument and target illumination. In conjunction with an incorporated encoder and electronic target acquisition this technique is feasible for 3-D monitoring. The motorised digital levelling as a successor of motorised optical and tacheometric levelling is another special application. In Sweden, motorised digital levelling was tested for the first time, its further development was influenced by the respective results and additional experiences.

Taking into consideration that refraction is the fundamental limitation of all optical observations, a new concept of refraction determination has been elaborated at the ETHZ – it shows the future of optical measurements including digital levelling. The new solution bases on the evaluation of image-distortion determined out of the digital image.

The contribution treats a sophisticated concept in the area of GPS-based height determination with components being appropriate to branch out into different classes of standard approaches, depending on the kind of data sources as well as the principal target. So, besides a GPS-based height determination, also a height system transformation may be set up. Basically any kind of height data, namely geoid models N (e.g. EGG97), heights H, levelling D H, GPS heights h and GPS baselines D h may be combined. Partly a Finite Element Model (FEM) is set up for the representation of height reference surfaces (HRS). This FEM is parametrized by bivariate polynomials sets, and continuity conditions guarantee a continuous transition of the FEM surface along the edges of neighbouring meshes in any area size. In opposite to digital terrain models, the nodes of the FEM mesh may differ from the position of the data used for the FEM determination.

The first part of the contribution treats the class of already practical working standard approaches, developed to transform in a statistical controlled way ellipsoidal GPS heights h into heights H of a standard height system. First the role of use, and the kind of a datum and systematics adaption, of geoid models N in a GPS height integration are discussed. The „geoid refinement approach" standard means that a datum adapted geoid model N is used as direct observation, while the FEM serves as additional overlay to improve the final representation of the HRS. The special case of the „pure FEM approach" arises, if the FEM representation of HRS is computed purely by geometric observations H, D H, h, D h. The „pure geoid approach" means, that only a datum adapted geoid model N is used in a GPS height integration. The three approaches provide a flexible area of models implemented in the software HEIDI2. Different pilot projects in several parts of Europe finished successfully, and the height integration concept is meanwhile used as a standard in some state survey agencies. The experiences show that a high precision level for a GPS based height determination up to 5 mm in rather large areas is achieved, e.g. using the EGG97.

The second part and class of approaches treats the application of the FEM component for the purpose of height system transformation (e.g. conversion of NN-heights to normal heights). First practical experiences and results of a pilot project with a German state survey department are given.

The third part of the presentation and class of approaches considers the so called general approach, where the HRS is completely established by a FEM, using different datum adapted geoid models N, terrestrial height information H and ellipsoidal GPS heights h as data sources. The result of the computation and "geoid mapping" respectively, leads to a Digital FEM Height Reference Surface (DFHRS). The DFHRS may be set up as database for a datum free direct GPS-based online heighting in DGPS networks. First experiences and results of a pilot project in the German SAPOS network are reported.

The surveyor, thanks to the rapid evolutions of the available equipments, has today a wide range of possibilities opened to him when he has to perform height determinations. The present paper presents some considerations regarding the optimisation in different situations.

High precision levelling methodologies are basically extremely simple, and easy to understand. But they present two specific aspects that are without any equivalent among any other topographic operations : (i) their accuracy may be extremely high, and (ii) they are based on operations that are reiterated so many times that the smallest systematic error should be carefully known, and (iii) to the opposite of many other geodetic processes, it is poorly overdetermined so that the internal consistency (loop misclosures) is not a very efficient indication about the quality of measurements.

We shall present here a synthesis about the errors in levelling, first the errors due to the imperfection of instruments, secondly the errors due to the use of these instruments in field conditions, and then to the atmosphere.

The GPS technique can be considered as a leading tool in horizontal surveying. Its application for height determination is hindered by two facts. Firstly, the vertical component cannot be determined as accurately as the horizontal co-ordinates because of the troposphere. Secondly, the ellipsoidal height provided by GPS being a pure geometrical quantity the geoid information is required to establish connection to the physical reality. As it is known the accuracy of the present geoid solutions in most cases does not fulfil the sub-centimetre requirements.

During the last few years considerable research work was focused to this topic all over the world, but the accuracy of the GPS-heighting is still well below that of the spirit levelling can provide. The only field for GPS-heighting may be the lower order levelling works.

In Hungary the research work was also assisted by practical standpoints. The Hungarian Levelling Network is not fully completed, in the western part of the network only the 1st order lines had been measured some 20 years ago. The completion of the network will be an important task for the next future. In September 1998 a GPS test campaign was performed within a carefully selected 2nd order levelling loop, where 34 selected benchmarks of all the included 3rd order lines were measured. The GPS data were parallel processed by BERNESE 4.0 and GPSurvey 2.1 softwares. A local geoid solution was also computed using an updated gravimetric database. The comparison of the original levelled and "GPS-levelled" heights were shown suprisingly good agreement. A technology had been elaborated, which is offered to be used in the future 3rd order levelling works. This paper summarises the results of the test campaign and introduces the technology worked out for GPS-heighting.

Since the end of the last century, two precise levellings have been made in Finland. The third national precise levelling started in 1978 and field observations have been estimated to be completed in 2001. In this paper the short description of these three levellings and the applied measuring techniques are given.

The French levelling network consists theoretically of 452 400 benchmarks distributed along 298 300 kilometres of railways, roads, waterways and rivers. The identification sheets of these benchmarks are captured in databank since 1983; the capture of the whole network is now achieved. The direct access to this databank begun in 1991 on Minitel. Minitel is a home terminal of the French telecommunications system, which has been developed in the eighties. Nowadays, about 50 000 benchmarks per year are asked by this way. The advantages for the users are the immediate and permanent access, the possibility of choosing criterions to get the benchmarks they need. The access by Minitel is well appreciated by the users and there is no question of giving it up but it has been decided to make the informations of the database also accessible, from 1999, on the Internet network.

The National Land Survey of Sweden (NLS) is deeply involved in the Third Precise Levelling of Sweden since the 1970s. The whole project represents more than 50 000 km double levelling (50 000 benchmarks) and the production work is supposed to be completed about year 2000.

This paper describes the unique production line that is used in the third precise levelling network. Most of the production line is digital and has been so since the start. We have been using the same production line almost throughout the project which mean that we have been trying to treat a specific type of data the same way during the 20 years that the project has been going on. Included in the production line is, besides the actual levelling, also the production of site descriptions, maps as well as storage of data in a suitable archive.

The production line can roughly be divided into five different phases that are synchronised in time. These are

1. Planning the network
2. Establishment of benchmarks
3. Storing information about the benchmarks into the archive as well as preparations for the levelling
4. Levelling
5. Computation, archiving and delivery of results.

The whole process takes for one region about four years of work to complete.

The third national precise levelling programme of Sweden – called "riksavvägningen" (RA) – has been in progress since 1979 and is estimated to be completed in 2003. The National Land Survey of Sweden, which is responsible for this project, has developed and used Motorized Levelling (ML) as the main technique to solve this task. Parallel to the main project, experiments and test has been done in order to improve the levelling techniques and make them more efficient, furthermore the tests has been used to find answers to different problems in the levelling program which have arisen during the project.

This paper describes experiments using trigonometric levelling, and specially MTL (Motorized Trigonometric Levelling) during the period 1985-1990. The different objectives of the tests were:

	Investigation of equipment and procedures with regard to accuracy
	Investigation of production capacity compared to ML
	Investigation of different systematic errors which had occured in ML
	Investigation of different discrepancies between the second and third National levelling programme
	The use of trigonometric levelling in rough terrain in mountain areas without roads

Swedesurvey is a state-owned company, which markets co-ordinates and provides services in land administration and surveying throughout the world, often in the form of institutional co-operation. We have been active on the international market since 1980 and our aim has been to provide services based on the requirements of the customer.

We have programmes of co-operation with cadastre and mapping organisations in many countries. The primary goal of our activities in this respect is institutional building through development co-operation and transfer of knowledge and technology. We also provide production services within our range of activities.

We mainly draw our resources from the National Land Survey of Sweden and have access to their full range of equipment and modern technology, and the competence of approximately 2,000 qualified professional and technical staff employed by the National Land Survey of Sweden. We provide consultants such as lawyers, cadastral experts, surveyors, geodesists, photogrammetrists, cartographers, valuation experts, computer programmers and administrators who are specialists with first hand practical experience and many of whom have overseas experience too. We also have close contacts with other private companies and governmental departments, we are partners in international consortia and have joint venture agreements with overseas companies.

Our fields of business include:

	Cadastral and Land Information Systems
	Surveying, Mapping and Geographic Information Systems
	Management, Organisation and Financing
	Training

After a look back on the development of levels at Carl Zeiss during the past 25 years, the author reviews technical details of the NI 002 and DiNiÒ instruments, analyzes their design and points to particulars of their operation, from which result special advantages in various applications. The analysis also includes the instruments' application in Motorized Levelling.

Nach einem historischem Rückblick zur Entwicklung von Nivellieren in der Firma Carl Zeiss in den letzten 25 Jahren werden technischen Details der Instrumente NI 002 und DiNiÒ betrachtet. Die konstruktiven Gestaltung der Instrumente wird analysiert und Besonderheiten in der Benutzung aufgezeigt. Daraus ergeben sich anwendungstechnischen Besonderheiten und vorteilhafte Einsatzmöglichkeiten. Die Analyse schließt auch die Anwendung der Instrumente für das bekannte Motorisierte Nivellement ein.

Après un bref aperçu sur le développement des niveaux de la société Carl Zeiss dans les dernières 25 années, l'auteur regarde de plus près quelques détails techniques des instruments NI 002 et DiNi^®. Il analyse non seulement leur principe de construction et leurs particularités d'emploi, il présente aussi les possibilités d'application préférées. Cette analyse inclut également l'utilisation des instruments pour le Nivellement motorisé connu.

In a city like Stockholm you can find a great need for height information within the building and construction industry. For existing buildings, heights are needed to control vertical deformations. Of great importance is the possibility to document the three-dimensional position of water pipes, sewer systems, electric cables, etc. It is vital that an unambiguous height system is used for all applications, especially when data are stored and exchanged in a data base. The Stockholm City Planning Administration is responsible for providing such a height system and as such maintains a series of benchmarks throughout the city, making connection to the system possible. Different users have different requirements of accuracy and availability of the benchmarks.

Several city administrations and other organisations require access to a large scale digital base map. Such a product covering the entire city is available from and maintained by the City Planning Administration. For example, city planners are using the base map to visualize new building projects. Now there are requirements of three-dimensional visualisation using computer animation. The height information in the base map is today limited to contour lines and some street elevations. Several methods have been tested in an attempt to find the best way to convert the base map into a three-dimensional city model.

Through out the latest 5 years The Danish Road Directorate has employed high precision GPS for solving a wide area of surveying tasks. During this work we have often learned that the realisation of the national reference system and den national geoid model has to be treated cautiously in order to obtain results fitting existing surveys.

Some examples of those problems and how they have been treated in the particular situation will be given in the paper. The basis of the discussion will be formed on two main subjects:

* The need of a high local precession for road work, with special focus on the heights for pavement.

* The establishing of a common reference system for the fixed link between Sweden and Denmark (Øresundsforbindelsen).

The National Land Survey of Sweden, originating from 1628. The mission is to give support for creating an efficient and sustainable use of Sweden's real property, land and water. The combination of geographic information, land information, property formation and geographic information technology gives us unique possibilities to meet the users' needs.

Change in society is rapid and affects us all. To continue to provide high levels of service, we need to anticipate and plan for change in the needs of those who depend upon our data, information and services. We do this in the context of a transformation in the very nature of the public service: over the last years, there has been enormously increased emphasis upon efficiency and measuring value for money in these services. The operation accounts for the last years show that we have had problems in the adoption of the organisation to new circumstances, but that is now history. In 1998 we started to earn money again, and we plan to do still better. Not primarily for our own sake, but for our customers. To meet the users' needs it is essential to have resources for development of new products and services as well as staff competence.

In the introduction of the seminar on Geodesy & Surveying in the future I will summarise a vision of where we intend be in five years time. I will also give examples of the strategies and plans which will translate the vision into reality.

Among the different processes and equipment's for 3D measurements, the new high accurate motorised total station Geodimeter 650 S (GDM) from Spectra Precision gives an interesting contribution to help the users. Servo-motorised, the GDM 650 S can be equipped with an electronic and optical device called Tracker, able to detect the infrared emission at a specified frequency sent by the IR diod of an "active prism" Remote Target (RMT) located at the point to be measured. The GDM 650 S Tracker automatically and continuously aims at the diode of the RMT and locks on it during the moving of the RMT holder. The GDM 650 S also measures the distance on the RMT prism.

For height determinations it follows at the combination of:

A continuous "Autolocking" on a diode making the pointing on the vertical angles-trigonometric height determination - very stable and independent of the operator's eye skill,

together with

The building of an arithmetic mean value of distance integrating the small and sudden variations of the refraction, and mean value of angles, on.line corrections for the axis verticality alterations,

Give a very good height difference determination.

The instrument GDM 650 S Autolock has a resolution of 0.1 mm, thus giving 4 decimals on distances, height differences, X,Y,Z, etc. Angels reading in D at 0.1 cc (0.00001 gon) on Hz and 1 cc (0.0001 gon) on V.

We have made tests of height difference determination and angles on sight lines of respectively 73, 138 and 215 metres, checked with the accurate Zeiss Ni002 level

	After 7-8 seconds, a perfect stability and repetitive of the results were obtained.
	Agreements of 0.1 mm (at 73 metres), 0.4 mm (at 138 metres), 1.0 mm (215 metres) were obtained compared with the obtained values with the Zeiss Ni002. They correspond well to the manufacturer's specifications, and are even better.

Specified angles 3cc (0.0003 gon), distances: 1 mm + 1 ppm accuracy.

Furthermore, the technique of continuous automatic locking on special RMT targets sending out modulated IR light, together with accurate and/or quick angles and distance measurements on prism reflectors can be used for Position Monitoring, Machine Control, Position Tracking. This family of equipment developed by Spectra Precision called ATS: Automatic Tracking System, is also a useful contribution to field measurements.

We discuss the interconnection of vertical networks using GPS and gravimetric geoid models, and give some examples how this can be done in practice. We also discuss several error sources of GPS-based methods which degrade the accuracy of height determination. GPS can be used for interpolation of heights, and some medium/low precision spirit levelling may be replaced by GPS. It can also be utilised in maintaining country-wide levelling networks or connecting national levellings. However, many ordinary spirit levelling errors cannot be controlled by GPS and it will not replace precise levelling over relatively short distances.

Computing parameters for transformation between a global reference frame and a local datum presupposes knowledge of the geodetic co-ordinates in both systems for a number of control points. A common problem is to assign accurate geodetic heights above the local geoid. Errors of several meters might occur due to deficiencies in local geoid and/or inaccurate heighting. Under certain circumstances, these errors might have a bad influence on the horizontal fit. The paper gives an example of this effect and how it can be avoided.

Construction plant control and guidance is a very hot topic. Already laser levelling has been used to enable automated construction plant height control. RTK GPS is an alternative method of controlling the blade’s height in real time.

The paper details trials conducted by the IESSG whereby GPS and GPS/GLONASS receivers were situated upon a bulldozer, allowing kinematic carrier phase positioning of multiple locations upon the plant. The kinematic positioning was then compared to the on board laser guidance system as well as a digital level. The results show that precision in the order of a few millimetres is possible.

We have calculated vertical velocities and their change in time using the First, Second and Third Levellings of Finland, with central epochs 1902, 1946 and 1986. Models with and without simultaneous determination of heights give the same results for velocities, but different error estimates. Observed velocity change in time is statistically significant, but the significance could just as well be explained by assuming that the loop misclosures underestimate levelling error. It also can be shown to depend on the interpretation of the empirical covariances between the levellings. We therefore consider the evidence inconclusive and further research is required. A new land uplift map, drawn from all three levellings assuming constant velocities in time, reproduces the main features of earlier maps from First and Second Levellings only, but does not support some spatial irregularities in them.

The MTL-technique has been used for the last 10 years to achieve height determination of low precision. The equipment applied and the measuring procedures, including the error bounds, are described. The adjustment is outlined. Some practical field experiences are reported and an evaluation of the technique is given.

The field of geodesy and surveying has been changing rapidly during the past twenty-five years. This change was mainly driven by new measurement systems which required new mathematical tools for handling high-rate kinematic data and new operational procedures to optimally use the new measurement tools. Twenty-five years ago, after almost two decades of satellite geodesy, this development and its impact on practical surveying could not be easily predicted. If anything, the predictions were much too pessimistic in terms of the speed of the development. Today, when looking at the next twenty-five years, some trends can be distinguished, but it seems as difficult as it was twenty-five years ago to predict the future with any degree of certainty. The objective of this paper is twofold. In the first part, some of the global trends will be briefly outlined and their impact on the world of surveying and mapping will be discussed. These trends indicate major shifts from static models to dynamic models, from stationary observation techniques to kinematic techniques, and from sparse data handling with geometric constraints to high-rate data handling requiring spatio-temporal filtering techniques. They are described by keywords, such as time-variable reference systems, global georeferencing of local measurements, and the increasing use of kinematic multi-sensor system and their impact on operational and numerical procedures. In the second part, these trends will be specifically looked at in terms of a global height system. Here, the definition of a global vertical datum and its change in time, the impact of some of the upcoming gravity satellite missions on accurate global geoid determination, the effect of airborne gravimetry, and the development of fast and accurate surveying of height profiles (roads) and areas (topography) will be discussed. The paper will conclude with a cautious extrapolation of these trends.

Recent developments in kinematic measurement technology have made it possible to conduct local surveys from moving vehicles and express the results directly in an accurately defined global reference systems without the use of terrestrial ground control. Such systems can be deployed everywhere on the globe without the need for identifying existing ground control or establishing new control. They can be used on airborne or vehicle platforms and provide a mobile survey tool to rapidly cover large areas including urban centers. Systems of this type have become known as mobile multi-sensor systems. It is the objective of this paper to discuss the basic concepts behind such systems, outline the underlying mathematics, review the state of the art by presenting some of the existing systems, and show their potential by looking at some new applications. The presentation is didactical in nature and will emphasize the principles. A detailed bibliography will help the interested reader to further study specific applications.

Electrification of surveying systems began with Electronic Distance Measuring Instruments in the 1950s, followed by Angle measurement devices, then Total Station Systems integrating Digital processing systems, and finally GPS systems which now claim a large share of the market for surveying systems. Though height measurement systems remained in the classical style for a long time, improvement of electronic parts has rapidly changed the situation of the level system recently. Several advanced manufacturers have already introduced new models of digital levels to the surveying market in this decade. Sokkia, working at the forefront of surveying technology has recently introduced a new model of digital level, the SDL30, that combines accuracy and ease of operation at a very reasonable price.

The system employs a finer pitch (8 micron/cell) CCD sensor, high performance microprocessor (32bit RISC), and new bar code system called RAB-Code that utilizes a digital sensing method for highly reliable and wide ranging height measurements.

The graduation patterns on the staff are read by the CCD sensor in two ways; for longer distance measurements images are recognized as 3 kinds of mark to cover the lack of optical resolution, and in the case of shorter distance measurements they are converted into 6 kinds of mark to make up for the insufficiency of position information limited by the narrow field of view of the telescope.

The new system was tested in severe field conditions for brightness, partial obstruction, ground vibration, and air scintillation. Accuracy tests were performed in conformable with ISO DIS 12857-1.

SDL30 can be used not only for ordinary leveling, but also for measuring and monitoring applications.

After a brief review of the evolution of the vertical control in the Czech Republic from 1918 to the beginning of nineties, an outline is given of the present Czech activities towards incorporation of the national vertical control into the European structures currently established within the international projects like UELN and EUVN. An attempt to compile a map of annual velocities of vertical movements of the earth surface for a part of Central Europe is also mentioned. Finally, a feasibility of GPS heighting is discussed in the context of modelling high resolution quasigeoid for the Czech Republic and some recent results are presented.

Four important aspects influence the present situation in standardisation of surveying activities:

	standardisation of quality assurance and certification according to the series of Standards ISO 900x
	standardisation of geographic information and geomatics as represented by ISO/TC 211 and CEN/TC 287
	standardisation of accuracy requirements of various engineering professions
	standardisation of meteorological procedures for verification of surveying instruments

Information technology and globalisation changed the surveying profession and surveyors all of a sudden have to face the fact, that the profession considered until recently an "art of positioning and mapmaking" is not any more an art but in most cases a „push button technology" which can be carried out "by any one with a good common sense and basic knowledge of IT".

Except for the meteorological procedures, the initiative in the creation of new standards in the fields of quality assurance, geographic information and accuracy requirements has been taken over by other professions. This concerns not only proposal of new standards but also creation of completely new terminology.

Since the FIG PC Meeting in Berlin in 1995 the FIG realised the seriousness of the situation and created a liason status with ISO/TC 211 Geographic Information and Geomatics. CERCO established in 1997 a WG on Quality and exactly at this moment is meetinf to discuss the "Good reasons for introduction of the Quality Control according ISO 9000". CLGE are preparing a document on quality assurance and certification of companies and individuals acting in the surveying businesses. IAG participates actively in the final revision of the ISO/WD 15046-11.8 Geographic Information/Geomatics - Part 11: Spatial referencing by co-ordinates. FIG established a Task Force for standardisation.

This presentation aims to inform on the present situation in field of standardisation and tries to define the role surveyors should play in this field.

The phenomenal progress made by motorized leveling in the past two decades has not changed the status of vertical geodetic control (VGC) world wide. Only a few countries have been able to establish and to maintain (re-observe at reasonably short time intervals) their national VGC networks at first order accuracy's. The establishment of a national VGC network by precise levelling has been indeed a technical possibility for many years and for some countries it has been even economically affordable. We claim that today those national VGC networks are not indispensable. Due to their many drawbacks such as high cost, slow pace, obstacle-crossing impotence and monumentation problems, we believe that they are going to be replaced in the near future by a more attractive alternative. The alternative we envision is a national VGC network based on GPS measurements which forms an integral part of a 3-D spatial network. The reference surface is the WGS84 (or any future equivalent) ellipsoid and thus the system of heights is ellipsoidal and not orthometric. We believe that the switch from orthometric (OVGC) to ellipsoidal (EVGC) height systems at the national (or at the continental or even at the world wide) level is imminent and inevitable. OVGC networks may continue to be used in the future but that only in the form of floating islands with a limited geographical extent. Viable candidates for such orthometric islands are regions and projects where the use of orthometric heights remains mandatory. There are cases where the island can be ellipsoidal in character or in other words form a densification layer beneath the national EVGC network. Typical examples of such orthometric or ellipsoidal islands are urban areas, low-shore areas, sensitive water system projects etc. Each island maintains its own independent datum based on those points through which it is attached to the national EVGC network. In Israel we are already experimenting with the above flexible scheme of vertical control. The national EGVC network is composed of two layers: a primary layer of ten permanent GPS stations at distances of 30-90 kilometres and a secondary layer of 150 extremely well monumented points (the G1 geodynamic network) at distances of 10-20 km.

Four orthometric islands based on the above secondary layer are at various stages of construction:

1. The Mediterranean island: a narrow strip of land along the coast covering an area of 2000 km²;
2. The Mount Carmel island: with 67 points over an area of 600 km²;
3. The City of Eilat island: by the coast of The Red Sea with 20 points over an area of 20 km²;
4. The City of Modyin island: and its environs (still in the early stages of planning).

The geoid surface determination is based on the difference between geoid point positions and reference ellipsoid point positions. The local geoid surface could be modelled by the approximation function, usually the first order, the second order surface or spline function. The least squares collocation may be applied in the above mentioned approximations. The choice of the approximating function depends on the shape of the actual geoid and the number of points where the geoid heights are known. An alternative method of approximation is the use of artificial neural networks (ANN). Unlike in other approximation methods, the neural networks need to be trained to simulate the desired functional relationship. The ANN approximation has been tested on existing astrogeodetic geoid of the chosen area. It has been concluded that ANN may efficiently approximate the real geoid heights on the chosen area. It has been found that the local geoid approximation by the ANN is consistently better than the first order approximation functions and comparable to the second order approximation functions.

The vertical rod comparator of the Finnish Geodetic Institute (FGI), automated in 1996-1997, applies the HP Laser Interferometer 5529A with the HP5519A Zeeman stabilized HeNe laser as a length standard and the Cohu CCD-camera with an area sensor for determining the position of the rod lines. The comparator is housed in the FGI´s laboratory in Masala, especially designed for rod calibrations. The atmospheric conditions in the room can be regulated by an airconditioner for temperature from 5°C to 50°C and for humidity from 5% to 100%.

Since 1988 the behaviour of the invar rods with aluminium frame was examined by determining their linear thermal expansion and analysing the results. Several irregularities in length of invar rods were found, mostly indicating malfunction of rod compensators.

The Swiss instrument manufacturer Leica introduced the first digital level, the WILD NA 2000 in 1990 after discarding a number of prototypes. The introduction of the NA 2000 heralded a new phase in the development of efficient levels because it was the first instrument to allow the automatic measurement and storage of levelling data. The high degree of automation thus achieved resulted in very large gains in productivity in the determination of heights. Leica introduced the first digital precise level, the NA 3000 at the end of 1991. Two further high precision levels were introduced by other manufacturers in 1994: the German firm Carl Zeiss with the DiNi 10, designed and produced in Jena, and the Japanese firm TOPCON with the DL 101. The large number of these and various digital engineering levels sold indicate that digital levels have achieved a wide acceptance and have replaced optical levels.

All three of the above-mentioned firms have subsequently made further developments in their high precision instrument. The NA 3003 with the software 4.2, the DiNi 11 and the DL 101C are considerably better than their predecessors are. Some instrument of each of these levels have been repeatedly and thoroughly tested under practical and comparable conditions in the Faculty of Surveying and Cartography at the Hochschule für Technik und Wirt-schaft Dresden (FH) - University of Applied Sciences (HTWD) over the last three years. They were tested in respect of their accuracy, reliability, and ease of use and in respect of the influence of external factors on the levelling results. As far as external factors are concerned, particular attention was paid in the investigations to the influence of:

- changes in temperature on the stability of the line of collimation,

- vibration and influences of wind on the reliability and accuracy of the results,

- various lighting conditions and shadow patterns on measurement ability and on the results,

- partial obscuring of the staff and staff inclination.

In general, the precise digital levels currently produced by all three manufacturers fulfil the requirements of precise levelling on foot without major problems. This has been proved by the adjustment results of two precise levelling networks, which were levelled a number of times with different instruments. No significant differences in accuracy or efficiency were found. These results are discussed in this paper, together with those of the special investigation into the above mentioned external influences. The best result in the latter investigations was obtained from the Carl Zeiss and Leica levels, where differences between the two were small.

The second part of the paper is devoted to the presentation of two new methods of motorised levelling which have been developed in East Germany over the years, and of their results. The author was involved in the development of both of these methods. The motorised levelling method developed by the mining firm Lausitzer Braunkohle AG uses the NA3003 for the measurements and vehicles made by the German manufacturer Volkswagen. In 1998, about 650 km of double levelling had been completed with the equipment by October, at a rate of 1,5 km of single levelling per hour. The accuracy achieved (0.5-0.6 mm/km) was well within the specifications.

The Survey Agency of the State of Saxony has used a second method of motorised levelling since 1996. It involves the use of a NA 3000 or NA 2000 level and Renault vehicles. The application of the DiNi 11 level in this method was successfully tested in the Saxonian first order level network during an undergraduate thesis at the HTWD in the spring of 1997.

The presentation will deal with all the Motorized levelling activities in Denmark, including geometric as well as trigonometric levelling, used in precise and in second order levelling. A small part of the presentation will deal with the geoid aspects, trying to fit the geoid to the old as well as the new height systems in Denmark.