A number of works have presented evidence that the seasonal terms of the Earth’s rotation are variables from year to year. In this paper, introducing a parameter associated to a mean seasonal kinetic energy we show evidence of about 4 year and 7 yr periodicity. This result is briefly discussed.

This year (1980) the U. S. Department of Defense has scheduled to have in operation six satellites of the Global Positioning System (GPS), which will provide timing and three dimensional position recovery potential to North America during certain segments of a day. By the mid-eighties, continuous timing and three dimensional recovery from 18 GPS satellites are planned. Although the GPS is designed for fast position recovery to the 10-meter level, extended data collection periods could yield subdecimeter relative positioning on a routine basis.

Classical horizontal geodetic networks are commonly combined with space observations, mostly satellite Doppler, in order to optimize the accuracy of geodetic control points and, thus, satisfy as many types of users as possible. Since satellite Doppler observations refer to a fully defined three-dimensional reference system and terrestrial observations, through the presence of Laplace stations (astronomical longitude and azimuth), contribute also to the pole and longitude orientations, it is imperative to ensure the highest possible degree of compatibility between the astronomical and satellite Doppler systems to maintain optimization of the accuracy of control points. Since gravity and geopotential (in the form of spherical harmonics) data are usually combined to evaluate geoid undulations and deflections of the vertical which are in turn used to reduce terrestrial angular and range observations, it is equally imperative to ensure that the satellite Doppler system and that of the geopotential solution are truly geocentric and thus compatible with the gravity data which should refer to a single equipotential surface. In order to estimate the degree of compatibility in terms of longitude orientation between satellite Doppler and geodetic astronomical systems as realized by current observations, astrogeodetic (based on CIO pole, BIH longitudes, and NWL9D satellite Doppler system) and gravimetric deflections of the vertical were compared at several hundred stations of the Canadian geodetic framework and U.S. transcontinental traverse. It was found that, when using the U.S. data subset only, incompatibility between the zero geodetic meridian plane of the NWL9D system and the zero astronomic meridian plane of the BIH was of the order of 08, which is in good agreement with previous results. However, inter-comparisons between various North American subsets revealed inconsistencies between areas of up to 08 (between Canadian and U.S. geodetic astronomical longitude observations). These results are based on the assumption that gravimetric deflections are bias free. The geocentricity of the NWL9D system with respect to other systems such as the Goddard Earth Models and SAO Standard Earths is also analyzed by comparing satellite Doppler derived geoid undulations with GEM and SAO SE undulations. An incompatibility of 4 m in Z (axis) exists between the origin of the NWL9D system and that of the other systems.

Present knowledge of the number, distribution, proper motion and structures of extragalactic radio sources indicates that there should be no problem in defining a celestial reference frame with stabilities of a few milliseconds of arc over time spans of the order of a decade. One of the limiting factors appears to be the structure of the sources. By measuring and monitoring these structures, the stability could probably be improved by as much as one or two orders of magnitude. Even without this improvement, a network of properly distributed fixed observatories making regular interferometric observations of these radio sources could be used to define a terrestrial coordinate system that could be maintained at the few centimeter level over indefinitely long time periods. Such a stable terrestrial reference system would be useful for a host of modern geodetic and geodynamic applications, including, in particular, studies of the time varying deformations and relative motions of lithospheric plates. The National Geodetic Survey has already begun work on a three station base network of permanent observatories under project POLARIS as a first step toward implementing the new celestial and terrestrial reference frames. It is hoped that others will join in the effort and make the new reference frames a reality by the middle of this decade.

Single measurements of gravity cannot give sufficient information about the position of measuring points with respect to some terrestrial reference system. Only a set of gravimetric stations all over the Earth combined with a determination of their coordinates allows one to determine (from the solution of Molodensky’s problem) the heights of these stations with respect to a level ellipsoid with center at the geocenter. Given in addition their heights above some reference ellipsoid, whose position in the Earth’s body is fixed through a set of reference points on its surface, the position of the geocenter in the same reference system may be obtained.

In active tectonic zones relative earth surface displacements of mm-order and, relative gravity disturbances of 0.1 mgal size may be expected between two survey epochs. To monitor and separate these small effects significantly special local free networks have to be designed which are measured by relative observations of highest accuracy (mm-and ygal-level). For the connection of the local nets to a global terrestrial reference system a suitable local datum for point coordinates (geometric datum) and gravity field parameters (gravimetric datum) has to be defined within the terrestrial reference system.

The pole coordinate obtained by the Doppler observation of Earth satellite shows good results, but there are some important difference from those of the astronomical observation, for example, that of the IPMS.

The Doppler observation of the Navy Navigation satellite has proved to be precise enough to supersede the astronomical observation in the geodynamics. (Anderle, 1965, 1966, 1970, 1971a, 1971b, 1971c, 1974). Yumi et al. (1979) showed that the local terms in the astronomical observation was not found in the Doppler latitude obtained at Mizusawa. This shows the superiority of the Doppler observation to the astronomical one.

In order to compute satellite motions with centimeter accuracy, the reference system, to which they are referred, should be carefully chosen. In fact there are many kinds of the reference systems. In this paper advantages and disadvantages of various reference systems are discussed.

The center of mass of the Earth is commonly taken as origin for the coordinate systems used in satellite geodesy. In this paper the notion of the “geocenter” is discussed from the point of view of mechanics and geophysics. It is shown that processes in and above the crust have practically no impact on the position of the geocenter. It is possible however that motions of the inner core may cause variations of the geocenter of the order of 1 m. Nevertheless the geocenter is the best point for the origin of a coordinate system. Mather’s method of monitoring geocenter motion is discussed, and some other possibilities are mentioned. Concerning the scale problem, the role of the constant GM and time measurements in satellite net determinations are briefly discussed.

The possibility of determining the variations of the Earth’s cejiter (expressed in terms of the geopotential harmonic coefficients , , or of some functions of them), using geostationary satellites, was discussed by Burŝa and Ŝíma (1978) after discerning the difference between observed and theoretical positions of libration points. Although their effect on geostationary orbits cannot be expected to be of high magnitude, the question of determination of these terms arose. It was suggested that the “resonant phenomenon” observed here might be analogous to the resonant effects in close satellite orbits, because the satellite is at the 1/1-resonance (1 nodal revolution per 1 sidereal day).

The use of spherical functions in dynamical problems is very common. As a rule, they arise in perturbing functions. It is well known that passing from one reference frame to another is accompanied by a double transformation of the perturbation function. That is why problems lose their simplicity and elegance. The problem of two solid bodies is a typical example in this respect.

In the present paper the questions connected with the transformation of the spherical functions when passing from one reference frame to another frame are considered. Traditional functions are generally unsuitable as they introduce a series of difficulties in the problems. That is why complex spherical functions are used. The transformation of spherical functions due to rotation of the coordinate frame is made by means of the Wigner’s functions. When translating the frame the Clebsch-Gordon’s coefficients are used.

In order to describe geometrical facts in geodesy, geodynamics, or astronomy, one must have suitable reference frames. This paper deals with some basic concepts for the definition of both terrestrial and celestial systems which make it possible to describe position and motion of mass elements of the earth or stars, respectively. It is suggested to use such frames with respect to which coordinate changes of points on the earth’s surface or of stars, respectively, become minimal. The rotation vector should be defined by applying the rotation operator on the velocity field.

Scandinavia rests upon one of the most stable crustal platforms on the Earth, the Baltic shield. None the less the study of crustal movement is of great importance. Inparticular studies of vertical movement induced by post glacial rebound have been carried out for over 100 years. These have been accomplished usinq water level variations, and more recently in this century using precise levelinq methods. The uplift measurements of Fennoscandia has led to a clearer understanding about the rheology of the crust and upoer mantle of the Earth.

In an Earth-fixed reference frame, polar motions (precession, lunisolar nutation, free nutation) introduce small apparent forces in the equations of motion of an Earth satellite. We discuss the possibilities (a) of integrating the orbit in an Earth-fixed frame when tracking data are used for geophysical applications, and (b) of determining from orbital data a set of unknown parameters describing the long-period wandering of the pole.

Project MERIT is a special programme of international collaboration to Monitor Earth-Rotation and Intercompare the Techniques of observation and analysis. It was conceived in 1978 at IAU Symposium No 82 on Time and the Earth’s Rotation and a draft proposal was prepared by a working group set up by the Presidents of IAU Commissions 19 and 31. The proposal was endorsed at the IAU General Assembly at Montreal in 1979 August and at the IUGG General Assembly at Canberra in 1979 December, when the organisation and membership of the Working Group were modified accordingly. The Group is affiliated to the Commission on the International Coordination of Space Techniques for Geodesy and Geodynamics (CSTG), which is sponsored by the International Association of Geodesy (IAG) and by COSPAR. Project MERIT has received the support of the International Council of Scientific Unions and of many national organisations and observatories throughout the world.

The underlying method of forming the adopted Fundamental Reference System is the compilation of observations referred to some (independent) well defined coordinate systems. If the observations themselves are used to define the independent coordinate system, they may be used to help establish the Fundamental System. If the observations are differential, they may be used to refer star positions to the Fundamental System and/or smooth systematic irregularities within the Fundamental System.

With the accuracy of modern observations the relativistic treatment of the basic astronomical reference frames only requires the consideration of comparatively simple types of metrics such as heliocentric Schwarzschild metric, geocentric Schwarzschild metric and metric of the Earth-Sun system. Dynamical (related to the motion of the bodies) and kinematical (related to the light propagation) characteristics of these metrics enable one to perform the accurate relativistic reduction of astronomical measurements. In this reduction, the choice of specific quasi-Galilean coordinates may remain arbitrary. This paper presents expressions for the main relativistic terms in coordinates of the principal planets and Moon using the PPN formalism parameters β, γ and coordinate parameter α. General formulae for the reduction of radar, radio-interferometric and astrometric observations of planets and for the interpretation of lunar laser ranging are given. For estimating the actual magnitude of relativistic effects, the ephemeris data should be expressed in terms of physically measurable quantities.

In the past twenty years there has been a great amount of growth in radiometric observing methods, as well as in classical optical observations. Through radar ranging and Doppler observations of the planets and spacecraft, we have been able to improve our knowledge of the location and motion of the planets by several orders of magnitude and have succeeded in planning and executing space missions which would have been difficult if not impossible to plan and to perform utilizing the classical ephemerides. We will outline the goals and methods employed by the Jet Propulsion Laboratory in its effort to develop improved ephemerides which accurately reflect the motions of planets in an inertial system.

The IAU (1976) System of Astronomical Constants, the FK5 and new lunar and planetary theories are being introduced in 1984. The investigation and planning for the transition has revealed the complex interdependencies between observational techniques and the reference systems, and their strong link to the rotating and orbiting Earth. The inaccuracies in our knowledge of the star positions, astronomical constants and the rotation and motion of the Earth are embedded in subtle ways in the observations and the reference coordinate systems. For example, the FK4 reference system in 1950.0 coordinates rotates with respect to an inertial system. Details are given for the conversion to the new system.

The possibilities of tying the geometrical and dynamical reference frames by use of lunar occultations are reviewed. Actually, the best accuracy available is about 0”.4, essentially limited by stellar positions and limb corrections. One can expect an improvement of one order of magnitude with the accomplishment of POLO and HIPPARCOS Projects.