PROGRESS OF SCIENCE

Wanganui Chronicle, Volume 74, Issue 148, 25 June 1931, Page 11

PROGRESS OF SCIENCE

WARNINGS OF EARTHQUAKES FORECAST STUDIES On October 28, 1891, a great earthquake devastated the two provinces of Alino and Owari in the main island of Japan. In its central regions it was as strong as the destructive earthquake of 1923 and it was felt over an area twice as large. A few weeks later the House of Peers petitioned the Government to appoint a committee for the study of earthquakes, and in the following June an Imperial Ordinance decreed the foundation of such a body. Two objects were set before the Earthquake Investigation Committee. One was to ascertain whether there are anyways of predicting earthquakes, the other to study what can bo done to lessen tho disastrous effects by the choice of proper structures, materials and building sites. The committee, first under the guidance of Professor 8. Sekiya and then under that of Professor F. Omori and Professor A. Imamura, did not feel themselves bound within these definite limits. They studied earthquakes and earthquake phenomena in every way that seemed desirable. In the first line of research—that of forecasting earthquakes—no progress was made for more than 30 years, and it is worthy of notice that the methods that now seem so full of promise arose, as it were, accidentally from researches that appeared at the time to have no practical value. In forecasting other natural phenomena within their range, the committee more than Justified their existence. Early in 1914 slight shocks of rapidly increasing frequency w’ere recorded in

the seismological observatory of Kagoshima. They were soon recognised as closely resembling those that precede a volcanic eruption. The island contained more than 23,000 inhabitants, and the authorities, on the recommendation of Professor Omori, ordered the immediate removal of the whole population. It was a race against time in which the Government won, for all the people were safely removed from the island, and the only persons who lost their lives during the eruption were three officials who in tho course of their duty remained too long on the island.

To return to the forecasting of earthquakes. Among the earthquake centres that surround the city of Tokyo, one is of great importance, for within it occurred the destructive shock that ruined the capital, Yedo, as it was then called, in 1855. Three-quarters of a century have passed, and now, once more, the centre is beginning to show ominous signs of coming action.

The river Sumida, the Thames of Tokyo, flows southward through the city into Tokyo Bay. Immediately to the east lies a block of tho earth’s crust, about four miles wide, bounded on both sides by faults. The earthquake of 1855 was probably caused by a settling of this block, especially along its eastern fault-margin and practically within the area of the city. Two months after the great earthquake of September 1, 1923, the Military Land Survey measured the levellings along two lines crossing the block in question. The last survey of the district had been carried out in 1918. When the two sets of figures were compared it was found that the block between tho faults had subsided, in one place by as much as 25 inches. As no other earthquake of any consequence had occurred within 25 miles of the capital during these five years, it would seem fair to associate the depression with the earthquake of The movement of the block did not, however, cease with last year. Since then, with the thoroughness that is characteristic of the Japanese race, the survey of the district has been twice repeated—in January, 1926, and in March, 1930. During each interval the block has undergone a marked subsidence, and the curves of equal depression in both intervals bear a close resemblance to one another and to the curves for the interval 1918-23. And now we come to the point that is of most interest in these measurements. During the first, interval, 192326, no earthquake of any importance originated within 25 miles of Tokyo. But, during the second one, 1926-30, an earthquake occurred on May 2.1, 1928, that caused some damage to property in the low-lying country to the east of Tokyo. Tho centre of the earthquake was near tho sea and close to the eastern fault. Thus the block-move-ment. during the earlier interval seems to have prepared the way for the slip that caused the earthquake, while that during the later interval was in all probability a direct result of that displacement. For an earthquake forecast to be of any use we must know not only the place where it is likely to occur but within fairly narrow limits the time when it may be expected. A forecast so rough as that given by the earlier pair of levellings described iibovo possesses but little practical value, except to show where instruments should be erected that will provide the. essential clement of tho time. Many years ago a pair of horizontal pendulums were set up in Tokyo in order to detect changes in the inclination of the ground. When their records were examined it was found that suck changes were constantly going on. Both daily and yearly variations were detected that were clearly connected with changes in tho temperature of the air. But occasionally there were other remarkable tilts that could not be attributed to atmospheric changes, and one of these it was noticed occurred just before a local earthquake on August 3, 1926. Led by this coincidence the Tokyo seismologists examined the records that were, provided by the same instruments for six weeks before) thi- great earthquake of September 1.

1923. Besides the changes due to atmospheric variations there were two remarkable tilts revealed, one during the first half of August that may have been a forerunner of the earthquake, the second on the morning of September Since then observations have been made in other disturbed districts with tiltmeters specially designed to record the slightest change in the inclination of the ground. Tho tiltings to be measured are exceedingly minute. For instance, that which occurred during the eight hours before the earthquake of 1923 was no more than would be experienced if one end of a straight line ID miles long were raised by one inch. But there can be no doubt about their existence, or about their value in pro viding trustworthy warnings that before many years have passed may be tho means of saving thousands of lives in a country so subject as Japan is to disastrous earthquake shocks.