1980 Volume 32 Issue 12 Pages 721-739
Intensive observations of the geomagnetic field have been carried out since 1976 in the Izu Peninsula, Japan. An array of observation sites covers the area of anomalously high microearthquake activity persisting since 1975. The array also extends over the region of crustal uplift amounting to 15cm or thereabouts which was revealed at the beginning of 1976. At present an observation system consisting of four types of stations is in operation. The total intensity of the geomagnetic field has been measured continuously by proton precession magnetometers at four stations. At two other temporary stations the total intensity is to be measured whenever earthquake activity becomes high in their vicinity. Three components of geomagnetic field variations have also been measured by a flux-gate magnetometer at another station. Finally, 45 observation sites have been established for repeated surveys of the total intensity. In addition to the repeated surveys at these sites, a small-scale experiment based on a synchronized measurement of difference in the total intensity between station pairs has been under way at an array of 14 sites located near the center of the crustal uplift.
During the period from 1976 to 1978, four major earthquakes took place within a distance of 30km from the array: a M5.4 event of August 1976, a M7.0 one of January 1978, a M5.0 one of November 1978, and a M5.4 one of December 1978. In association with these earthquakes, some changes were detected in the total intensity and also in short-period geomagnetic variations. In the case of the largest event of magnitude 7.0, the difference in the total intensity between the two stations, where continuous measurements had been made, underwent a change about two months prior to the shock. The change was very similar in time sequence to that in the electric self-potential observed near one of the above two stations. Amplitudes of short-period geomagnetic variations also changed about two months before the shock. The result of repeated surveys disclosed a striking pattern of spatial distribution of presumably coseismic changes which is quite opposite to that obtained during a post-earthquake period. About two months prior to the M5.0 earthquake which occurred several kilometers east of another station for continuous measurements, the total intensity remarkably decreased by 5nT or so at the station. The field abruptly recovered after its occurrence; a coseismic change amounted to about 5nT. In the case of the M5.4 event of 1978, no coseismic change was detected at a temporary station located a few kilometers west of the aftershock area. Changes in amplitude of short-period geomagnetic variations also appeared before this M5.4 earthquake.
