Before combining the coordinates of the pole obtained by the Dahlgren Polar Monitoring Service (DPMS) with those of the Bureau International de l'Heure (BIH) we investigated the systematic differences between the two series and their relative weights.

In this paper the objective of the ATS-F/NIMBUS-E Tracking Experiment, the first of such kind, is presented. Specifically, this experiment has a two-fold purpose: First, to gain experience in the practical use of satellite-to-satellite range and range rate data for very accurate orbit determination (this was its original intent); and second, to evaluate the real usefulness of such a technique for geodetic studies despite the fact that the 1000 km NIMBUS orbit is not ideally suited for such a purpose.

The accuracies of the tracking systems of the satellite-to-satellite and satellite-to-ground link (ATS-F to the Rosman, N.C. ground station) will be ~ 0.035 cm/s in range rate and ~ 1 m in range – utilizing a 10 s integration time. With these values one obtains, based upon performed error analyses, orbit height errors in the order of 0.1 to 0.3 m for the near earth orbiting NIMBUS spacecraft. This experiment will therefore hopefully prove to be a significant first step for future Earth applications spacecraft carrying altimeters systems for measuring ocean height variations.

In addition, laser corner reflectors placed on board the ATS-F will make a total independent position determination of this spacecraft to approximately 15 m to 30 m possible, assuming that the location errors of four laser tracking stations used to determine the orbit are about 3 m or 5 m in each component respectively, with laser ranging system noise errors of 1.2 m and bias errors of 0.15 m. A small position error of the ATS-F, the ‘orbiting tracking station’ is essential in order to make full usage of the small satellite-to-satellite tracking errors mentioned.

For purpose of geodetic studies, one of the final goals, range rate variations of 0.1 cm/s or less corresponding to surface gravity anomaly of 20 mgal or less (over a half-width of 100 km on the Earth surface), will have to be measured during a time interval of approximately, say, 30–60 s. These values are within the range of the planned tracking systems accuracies for ATS-F and NIMBUS-E.

For three months in 1970, NASA, Goddard Space Flight Center conducted an experiment to try and detect the motion of the Earth's pole of rotation by laser ranging to an Earth satellite. Two laser tracking stations were used, one at Goddard Space Flight Center and one near Seneca in New York State, 408 km away and on the same meridian as the Goddard station. The objective of the experiment was to precisely measure the orbital inclination of the Beacon Explore C spacecraft, on a daily basis and with a resolution of a few hours, to a few hundredths of an arc second by the simultaneous tracking of the spacecraft as it passed between the two stations. During the experiment over fifty thousand range measurements on over two hundred revolutions of the satellite were obtained and on fifty occasions the satellite was simultaneously tracked at both stations. The noise accuracy of the tracking varied between 20 and 60 cm during most of the experiment and the orbital inclination obtained from the measurements showed a systematic change of about 25 arc sec, due primarily (about 80%) to the gravitational perturbation of the orbit by the Sun and the Moon. Perturbations by solar radiation pressure and the solid-Earth tides appear to account for most of the remaining systematic changes in the inclination and from which it has been possible to derive a value of Love's number for the Earth as A preliminary determination of the motion of the pole in the meridian containing the stations by subtracting the known perturbations (including tides) from the observed perturbation suggests that an accuracy of 3 to 5 m has been obtained, over periods of about 6 hr, during the latter part of the experiment. The experiment has also demonstrated that the present laser tracking systems are able to measure the chord distance between the two tracking stations with a repeatability of 25 to 30 cm.

The present analysis of the data is directed towards an improved extraction of the gravitational perturbations and an improved determination of the Love number k2.

A full account of the experiment and its results so far will be published elsewhere.

The methods of using close earth satellites for determining the motion of the Earth's axis of rotation are discussed. Particular emphasis is given to the possibility of using laser range observations. Typical error sources in orbit computations are discussed and their influence on the resulting pole positions estimated. Within a few years pole coordinates with an accuracy of about 20 cm and with a resolution of 1 day or even 12 hr can be routinely expected.

Long-arc orbit computation of highest accuracy can provide pole positions. Optical Baker-Nunn and laser range observations of several satellites are combined. The accuracy of the pole position is comparable to that of the mean satellite-tracking station coordinates (±5 m) when sufficient tracking data are available. Exploitation of the technique requires more accurate tracking data.

A theoretical investigation is made to explore the possibility of obtaining corrections to star catalogues from the observations of artificial satellites. A model is established to represent the system formed when such corrections are determined simultaneously with corrections to the geocentric position of the observer. The covariance matrix is established analytically for the two-dimensional case. The resulting correlations and weights are discussed. Since this method presents an independent way of obtaining the systematic errors in star catalogues, it is related to the measurement of all quantities which depend on star positions.

The motion of the pole has been observed by means of visual telescopes, employed by the ILS, photographic zenith tubes and astrolabes. It is pointed out the difficulties of combining the results of different types of instruments employing several observation techniques.

The new astronomical methods of Doppler observations on satellites and laser pulses sent to the Moon give another possibility of determining polar motion with greater precision. It is forecast that the results will not agree with the older ones and, in order to compare the methods, it is suggested that stations should be equipped with both techniques of detecting polar motion.

Automatic operation of the PZT located at the U.S. Naval Observatory Time Service Substation, Richmond, Florida was initiated February 1, 1970. A comparison of one year's automatic operation with the previous year's operation has shown a significant increase in the number of time and latitude sights obtained.

Based on an improved ventilating system within the PZT, the probable error of a single night's observation has been reduced.

A comparison of the Universal Time and local latitude deduced from the Astrolabe observations since 1966 to 1970, with respect to similar results obtained from BIH datas, show significative annual differences, specially in time, of a stable pattern. The consistence of the latitude differences (Astrolabe – BIH) and latitude group corrections, computed by the chain method, prove that these differences could be explained by the influence of a systematic error in the FK4 of Δδα type. On the other hand the inconsistence of the strong annual differences in time with the time group corrections, show that these differences can not be explained by catalogue systematic errors. The group corrections in time are compared with an investigation of the FK4 Δαα error based in the observations made with a transit instrument at Santiago.

The author propose the inclusion in observational program of latitude services a number of Talcott pairs with greater zenithal distances (refraction pairs). It is specially advisable for the stations, forming a chain, where the comparison of local latitude data is necessary. Refraction pairs could be very useful for investigations of the local anomalous refraction influences.

The method is given for the analysis of the anomalous refraction in the zenith zone. Connected with this the observations of the same zenith stars in the meridian and the prime vertical are used. The moment of passage through the eastern and western part of the prime vertical is registered, while the difference of zenith distances in the meridian passage is measured – always in two instrument position.

The method practically eliminates coordinate errors influences of observed stars. For the succesful employment of this method, we must use the high quality instruments and, especially, correct apparatuses for the carreful determination of the inclinations of instrumental parts.

The pendulum astrolabe with an electronic transit detector after the principle proposed by I. Tsubokawa has comparable accuracy to those of the Danjon astrolabe and P.Z.T. even in the field. After the same principle, an instrument for observatory use is being constructed, in which all observational processes are done automatically under the predetermined programs. The necessary mechanisms for automatic operation are as follows:

(1) Azimuth setting mechanism for respective stars observed.

(2) Mechanism which turn the telescope through 180° during the passage of a star near the mean altitude.

(3) Tracking mechanism in azimuth direction.

(4) Tracking mechanism in altitude direction.

(5) Read-out mechanism of the times of passages of a star on respective altitude circles.

Of these, the altitude tracking mechanism directly affects the observational accuracy. The rotation of a plane parallel glass compensates the diurnal motion of a star and keeps its image in the state of repose on the edge of a prismatic knife of the electronic detector. The rotating mechanism gives to the plane glass combined rotation of the two parts; the servo-system of the electronic detector and the system of generating the variable rotational velocity corresponding to the diurnal motion of a star. The extreme accuracy in the earth's atmosphere is expected.

The purpose of this work is to investigate changes in the rotation of the Earth in the past few thousand years. Since most available observations involve the Moon, study of the Earth's rotation is inseparable from study of the Moon's motion. Since it is doubtful that present theories of tidal friction account for the present acceleration (Spencer Jones, 1939; Van Flandern, 1970; Pariisky et al., 1972) of the Moon, we cannot safely assume that consequence of the theories which says that tidal friction has been almost constant.

Rotational accelerations of the Earth arise from changes in speed of rotation: periodic, irregular, and secular. Changes are caused by winds, Earth tides, ocean tides, and motions of the core. The semimonthly lunar body tide produces the maximum acceleration.

Researches on random variations in the Earth rotation, published in the last decade, are analyzed. The basic conclusions are that both mechanical and electromagnetic interactions between mantle and core have definite and measurable influence in the rate of rotation; turbulent motion within the core and non-periodic flows in the mantle must also be taken into account; atmospheric and irregular tidal effects have possible influence. A statistical approach based only upon observations and a simple description of the above phenomena is proposed as an extension of previous works.

The usual procedures for estimating the motion of the Earth's pole use a least squares data reduction procedure to estimate the coefficients in a time series solution for the coordinates of the pole. Whether optical data or radar tracking data from near Earth satellites is used, the presence of random accelerations in the equations which describe the motion of the Earth's pole will lead to errors if the data is reduced in the classical least squares manner. This investigation presents a technique for estimating the polar motion in the presence of unmodeled accelerations. The unmodeled acceleration is represented by a first order stationary Gauss-Markoff process which can be separated into a timewise correlated component and a purely random component. Using this model, a sequential estimation procedure is developed for estimating the orientation of the pole, the components of Earth's angular velocity and the magnitude of the components of the unmodeled acceleration. The application of the method using both optical and radar tracking data from near Earth satellites is discussed.

This paper investigates the possibility that the boundaries between the Earth's mantle and core and between the inner and outer core might show temporal changes. The evolution of the Earth's core is not discussed, but the question is raised as to whether these boundaries might not undergo small changes both in size and shape over comparatively short time intervals.

The Earth with a rigid core and a rigid mantle is considered and a recovering force is assumed between the constituents.

The equations of motion in canonical form are derived in terms of the Andoyer's variables. Applying the perturbation theory, the phenomenon of the polar motion is investigated.

The polar motion obtained with this model is a circular motion of the rotational axis on the mantle's surface with its radius gradually changing.

After the advent of artificial Earth satellites, an accurate knowledge of the harmonic coefficients of the Earth's potential has enabled us to determine the size and shape of the Earth with extraordinary accuracy.

The gravitational force between the core and the rest of the Earth, which makes both centers of mass tend to coincide, is so enormous that there is a strong possibility of a very slow motion of the core forcing its way into the mantle. It includes secular variations in both latitude and the time of rotation of the Earth.

Kimura (1902) pointed out that there might be an unknown cause other than polar motion which would produce an apparent latitude variation, and introduced the Z-term as, Δϕ = Xcos λ + + Ysin λ + Z.

The theoretical results by Jeffreys-Vicente (1957) and Molodensky (1961) have shown that the effect of a liquid core of the Earth may increase the coefficients of the semi-annual solar nutation term (2 ⊙) which is involved in the diurnal nutation (the so-called Oppolzer term) by 0″.02. It is reasonable to accept this correction which will appear in the Z-term with an argument of (2 ⊙ −α).

From comparison of the observed amplitudes and the phase angles of the annual Z-terms derived from the ILS data, it is concluded that the argument of the principal annual term in Z is (2 ⊙ −α) and not ⊙. The following results were obtained for the annual Z-term for 1955-1966 from the analysis of data by two independent methods (Wako, 1970): 0″.0137 sin (2 ⊙ −α + 2°.2), 0″.0203 sin (2 ⊙ −α + 4°.3).

Melchior (1970) proposed another effect of the Earth's liquid core for the annual nutation in obliquity, thus a term such as a sin (⊙ +α + A) would appear in the Z-term and it might cause a part of semi-annual Z-term. For the determination of these corrections, analysis of Kimura's Z-term is the most effective method.