A global, monthly snow depth data set has been generated from weather satellite (Nimbus 7) observations using passive microwave remote-sensing techniques. In this paper we analyzed five years of data, 1980–1984, to compute the snow-load excitation of the annual wobble of the Earth's rotation axis. A uniform sea-level decrease has been assumed in order to conserve water mass. The result shows dominant seasonal cycles. The prograde component of the annual excitation is Ψ+ = (5.0 milliarcsec, −110*) and the retrograde component Ψ− = (5.0 milliarcsec, −31*). These computed values are compared with previous groundwater estimates, as well as the inferred values from ILS and LAGEOS polar motion measurements. The importance of accurate data is stressed and future plans proposed.

We have undertaken a computation of water storage fluctuations and their contribution to polar motion excitation for the period 1900–1982. The calculation is based mainly upon monthly mean observations of precipitation at individual stations around the globe. Seasonal average variations are computed for over 500 individual drainage basins, permitting a comparison with observed annual polar motion.

A new research for the secular drift of the Earth's pole was made based on nine long sequences of latitude observations and led to the following conclusions.

1. 1. During this century, the Earth's pole has been moving with a mean rate of about 0″.0016/yr along the meridian about 70° (W). This drift rate is much less than the 0″.0035/yr derived from the ILS sequence.

2. 2. Relative to the North American continent, the Ukiah station located on the west coast of the U.S. shows a local drift of about 6 cm/yr northward, which coincides well with that determined by new techniques.

3. 3. Referring to the Europe-Asia plate, the whole North American continent shows a drift northward with a rate of about 8 cm/yr. The Mediterranian shows a similar drift of about 6 cm/yr. Perhaps these drifts are the consequences of plate motion and/or deflection of local vertical. It is useful for ascertaining the sources of the drifts to intercompare longer sequences observed with different techniques, including classical and new ones.

4. 4. Three of five ILS stations, Ukiah, Gaithersburg, and Carloforte, show significant local drifts. Therefore, the Conventional International Origin (CIO), which is defined by the 1903.0 mean latitudes of five ILS stations, is far from fixed on the Earth's surface. It is necessary to re-define an origin of the pole of the Conventional Terrestrial System (CTS).

5. 5. The quasi-30-year libration showed by the ILS data is not the real pattern of the Earth's polar motion, but results from both the irregular polar motion over some period and the local motion of Ukiah.

Recent BIH-data from September 1979 - September 1986 have been deconvolved. The excitation function has been considered in detail and a spectral analysis indicates that there are atmospheric contributions; moreover, the September 19, 1985 Mexican Earthquake has been related to polar motion and a possible tidal triggering mechanism has been investigated.

The effect on the Earth Rotation Parameters (ERP) of all the large earthquakes that occurred during 1977–1985 is evaluated. It is found that they cannot have caused the variations observed in the ERP during this time period.

According to a vast amount of both seismological and astronomical data obtained from China, it has been understood that: 1, A large number of anomalies may appear before and after the major earthquake and the anomalous area is usually very large; 2, And therefore the anomalous residuals in the observations of time and latitude before the shock are of profound significance in geophysics.

Within the framework of Newtonian kinematics the local vorticity vector is introduced and averaged with respect to global earth geometry, namely the ellipsoid of revolution. For a deformable body like the earth the global vorticity vector is defined as the earth rotation. A decomposition of the Lagrangean displacement and of the Lagrangean vorticity vector into vector spherical harmonics, namely into spheroidal and toroidal parts, proves that the global vorticity vector only contains toroidal coefficients of degree and order one (polar motion) and toroidal coefficients of degree one and order zero (length of the day) in the case of an ellipsoidal earth. Once we assume an earth model of type ellipsoid of revolution the earth rotation is also slightly dependent on the ellipsoidal flattening and the radial derivative of the spheroidal coefficients of degree two and order one.

A fundamental dynamical scale, ET(solar), was formed from absolute, meridian circle observations, from 1835 to 1982. At 1967.0 the ET(solar) and SI seconds were equal to 1 part in 109 and, from Morrison's (1979 a) ET(lunar) scale, the ET(ILE) and SI seconds were equal to 1 part in 1010. The long term fluctuation in the Earth's rotation has been parabolic, instead of sinusoidal, for at least 350 y.

An analytical formula for the time transformation TDB–TDT valid over a few thousand years around J2000 has been computed with an accuracy at the 1 ns level. The coefficients for a formula accurate at the 100 ns level are provided here. The numerical and analytical procedures to compute this transformation are discussed. We note that these procedures cannot comply with recommendation 5 of the 1976 IAU meeting. Furthermore, these procedures yield different units for the corresponding TDB time scales. We also note that this transformation is independent of the two PPN parameters γ and β and of the 3 most commonly used coordinate systems (isotropic, standard-Schwarzschild, Painlevé) at least at the 1 ns level.

Time transfer data, derived from single frequency observations of GPS satellites, have shown a unique bowing characteristic in the residuals with an amplitude which sometimes amounts to about 80 ns. Investigations attempting to explain this characteristic have so far been unsuccessful. The studies have been concerned with hardware and software effects. The hardware studies have:

(a) compared data obtained with single frequency receivers with data obtained simultaneously with dual frequency receivers which eliminate the effects of the ionosphere;

(b) compared data obtained with different kinds of single frequency receivers; and

(c) compared data obtained simultaneously with different kinds of antennas, including one designed to minimize the effects of multipath.

The Mark IIIA correlator system is currently operating at the U. S. Naval Observatory in Washington, D.C. in support of VLBI geodetic measurements being made by NASA, NGS, NRL, and USNO. This correlator system, developed at Haystack Observatory, is a second-generation version of the original Mark III correlator, and adds significant new capabilities such as double-speed operation, longer integration periods, and improved internal modelling.

The Mark IIIA correlator can simultaneously process up to 10 baselines of data from 5 stations. Experiments including more than 5 stations may be processed with multiple passes through the correlator. The architecture of the correlator allows a future expansion to a maximum of 16 stations. Due to a simple modular design and the low cost of required computer-support equipment, expansion is straightforward and relatively economical.

Although current geodetic VLBI observations are made using data from distant natural continuum radio sources, the Mark IIIA correlator was designed to also support processing of data collected from pulsars, artificial earth satellites, and from earth-orbiting antennas receiving signals from natural sources. These capabilities may be important in the future as efforts continue, for example, to tie satellite-related reference frames to current VLBI reference frames.

The design approach of the Mark IIIA correlator will be discussed, including its performance with respect to systematic and random errors which may affect geodetic VLBI data.

The outline of the planned Japanese VLBI system for the Earth rotation study and astrometry (VERA) is described. As a result of simulation study, it is concluded that precision and accuracy of the VLBI estimates of astronomical and geophysical parameters are remarkably improved when VERA participates in the present IRIS network.

The National Radio Astronomy Observatory is constructing a Very Long Baseline Array (VLBA) consisting of ten antennas across the USA, between Hawaii and the Virgin Islands. The array will be a dedicated instrument for producing milliarcsecond images of radio sources. However, exciting geophysical applications of the array are obvious and, furthermore, detailed geodetic, astrometric and environmental parameters are necessary for the VLBA to achieve its full sensitivity, resolution and positional capabilities.

Evidence is found that in IRIS experiments the gravitational time delay due to the gravitational field of the Earth can be seen. Influences of this time delay upon baseline determinations are discussed.

The Kiev-Geodynamics-2 program complex was used for determination of earth-rotation parameters and station coordinates from the full set of the MERIT SLR data. The results have been compared with other realizations that have been reported to the Coordinating Centre at the BIH. Other researches in the reduction of SLR data and in instrumen-tation are considered.