A rigorous program of adjustment for any geodetic network must have a useful file of geoidal heights in its own to adjust strictly observations by GPS for space vectors. The program, by which geoidal heights can be automatically computed without missing its fine undualtion at every point where vertical deflection is observed, was newly added to the Universal Program of geodetic net-adjustment. The conventional method to find geoidal heights successively along routes linking neighboring vertical deflection points seems not elegant mathematically, because selections of routes are too arbitrary. It ishighly mathematical to find out geoidal heights at all vertical deflection points badjusting simultaneously all observation equations of vertical deflections which are ex pressed as the function of their geoidal heights. Although it seems geophysically reason able that geoidal heights are expressed by a curved surface with the mathematical formula written by coordinates on the surface of the earth, we are afraid that it might miss local fine undulation of geoid. In order to adjust rigorously geodetic network including GPS-observations, it is desirable that every vertical deflection point has its more reliable geoidal height. Vertical deflection at a point is decided by a local shape of the geoid around it. Then we think a new method as follows : Q We make a local geoid by using geoidal heights (rough height+small unknown correction) at several vertical deflection points inside a circle drawn around a vertical deflection point. Q2 We make both observation equations for I(north-south component) and (east-west component) of vertical deflection at thecentral point by considering the difference between geodetic longitude (latitude) and computed astronomical longitude (latitude) to be the direction of normal to the local geoid through the point. (3) We can find better geoidal heights at all vertical deflection points by solving simultaneously all observation equations by means of the method of least squares. How to derive a local geoid mathematically is the most important problem in abovementioned method. A curved surface with higher order terms requests many surrounding points. Reliability of local geoid made by using remote points deteriorates. Vertical deflection points are distributed locally in remarkable high and low densities. We adopted the following sophisticated way with much freedom after considering those peculiarities above.

View full abstract

Precise levelling has been carried out annually from 1977 to 1987 except in 1978 on a levelling loop across the Hanaori Fault, one of the active faults in Southwest Japan. Some of bench marks located on the levelling loop show significant elevation changes except for monotonously sinking ones, and it is probable that the change of under ground water level has no influence upon a tilt motion of the levelling loop. An annual tilt rate observed seems to have a negative correlation with a ratio of the number of earthquakes in the area concerned (approximately 1010 km²) to the number of those in the area along the Hanaori Fault.

View full abstract

A new method of astronomical observaion is developed. We have carried out astro nomical survey for over 30 years by G. S. I . type astrolabe which is equipped with E. T. D. (Electronic Transit Detecter). E. T. D, is an electronic apparatus which deter mines the transit of a star by using a knifeedge and photomultipliers . Formerly, the output of the E. T. D, is recorded on a pen oscillograph with time signal and the transit time is determined by measuring the record . The new method adopts a laptop computer with AD converter instead of the pen oscillograph to deter mine the transit time automatically as soon as it is observed . Firstly, the mechanism of the new method will be discribed . Secondly, the result by both methods will be compared.

View full abstract

During the 1970's, several authors reported the so-called ‘migrating crustal move ments’ observed in some tectonic regions such as the south Kanto and central Tohoku districts, Japan, and Peru. In this paper, we consider one of these phenomena, that is observed in the central Tohoku district, by means of numerical experiments. The experiments are carried out with a two-dimensional finite element model of a vertical cross-section of the central Tohoku district. The upper-crust is assumed to be elastic, while the lower-crust and the upper-mantle are both assumed to be visco-elastic. The responses of the elastic upper-crust to sinusoidal boundary shear stresses applied to the plate boundary near the Japan trench clearly show migrations of strain (or stress) waves. The velocities of migration obtained from numerical experiments (about 130 km/yr at the period of 5 yr) and from observations (about 40 km/yr at the almost same period) are, however, different from each other. This discrepancy is considered to be mainly due to uncertainty in the crustal structure as well as in the physical parameters such as viscosity. Even though there is some discrepancy between the experimental and observational results, we can expect the existence of the migrating crustal movements if the interaction between the oceanic and continental plates varies gradually with time.

View full abstract

A method how to accurately estimate the drift of gravimeter is one of important but difficult problems in the analysis of gravimetric data, especially for data obtained through gravimetric connections which were carried out taking a long duration of time. It was clearly recognized for some gravimeters, in international gravimetric connections along the Circum-Pacific zone performed in 1979-1982, that positive and negative drifts of the gravimeter occurred when the gravimeter was brought from places with large and small gravity values, respectively. It was, however, difficult to conclude which affects the drift of the gravimeter more, ambient temperature or gravity value of measured stations. A similar tendency was found for both gravimeters G-196 and G-605, but the opposite tendency for the gravimeter G-484 through an international gravimetric connection between Japan and China. This fact may suggest that a thermal effect on the gravimeter is more effective on the drift of LaCoste & Romberg gravimeter than a stress relaxation of its spring system.

View full abstract

Tidal strains and tilts are investigated using the data of 3 extensometers, 3 watertube tiltmeters and 2 pairs of horizontal-pendulum tiltmeters installed at the Miyazaki Crustal Movement Observatory in the east coast of Kyushu, Japan. Successive analyses by means of the least-squares method with 30 or 60 days data are carried out by shifting the origin of analysis by a 5 or 10 days step. The data with disturbance due to rainfall are removed from the calculation. The results from 1 year data in 1987 show that the amplitudes of MZ and Ki constituents at each component vary over the standard errors, but their phases are stable with some exceptions . Deflections of the amplitude from the typical value in each component appear mainly at rainy season in spite of the removal of the disturbed data. But we cannot find relations on all deflective values to effects of rainfall

View full abstract