This is an archive story, published in the March 1940 edition of Geographical.
All facts, figures and statistics were accurate at the time of original publication. The text has been edited solely for current house style reasons but otherwise remains unchanged. Language and terminology used reflects the writing of the time and has been left preserved as a historical record. Imagery used is purely illustrative of the theme and not taken from the archived issue except where indicated.
For access to the entire Geographical archive dating back to 1935, check out our digital subscription options. Every issue of Geographical right at your fingertips!
It is rare for a day to pass without an earthquake happening somewhere in the world and on many days there are three or four. Some of these are small earth tremors not felt by man and only recorded by instruments so delicate as to be capable of measuring ground movements of two thousandths of an inch. Such a one happened in Paris about 1.30am on Wednesday 17 January, 1940, and was recorded at the Paris Observatory. Some are of terrific magnitude but fortunately happen miles from the nearest human dwelling, thus doing no damage. Such a one was the Aleutian Islands shock of Thursday 10 November, 1938, which was recorded by seismographs throughout the world, caused a seismic sea wave which rolled up on the beach at Hawaii, and did so little damage that the Times in a leading article wished ‘that all upheavals could spend their force with so little damage: Thundering like ramping hosts of warrior horse, To throw that faint thin line upon the shore, and so pass innocuous into silence’. But such is not to be, as we have lately seen in the earthquake disaster which has overcome Turkey.
Scales of seismic intensity such as the Rossi-Forel Scale and the modified Mercalli Scale of 1931 enable seismologists to classify earthquakes according to their intensities and to determine roughly the areas from which they started (epicentral area). But probably one of the greatest desires of every student of earthquakes is to be able to forecast the exact time and place of the next great earthquake. Unfortunately it is not yet possible to do so. That there are certain periodicities in the occurrence of earthquakes is not generally accepted by seismologists, and though foreshocks do sometimes occur before a large earthquake, as on 23 November, 1939, and other days in Anatolia before the Turkish disaster of 26 December, 1939, yet they are not often recognised as such, but only considered as separate individual earthquakes and tremors. There is a known tendency of earthquakes to recur in areas where they have once happened but the time or magnitude of the next occurrence cannot be foretold.
Long before the great Chilean Earthquake of 24 January, 1939, the ground was known to be tilting gradually and this was recorded and commented upon by the Chilean seismologists, but it was impossible to say when or exactly where the earthquake would occur or whether the ground tension and compression would become dissipated in bending and in a series of small shocks which would do no material damage. In Japan changes in earth currents of electricity have been shown to precede certain earthquakes but these changes sometimes occur without earthquakes happening afterwards. Certain areas near the Himalayas have been found by the Survey of India to be deficient in the force of gravity. According to the hypothesis of Isostasy these areas are uncompensated and thereby unstable. Earthquakes have occurred near these places and there is without doubt some correlation between the facts though, as with the ground-tilting in Chile, the time, exact place and magnitude of possible earthquakes cannot be estimated in advance. The forecasting of earthquakes is one of the enigmas.
From the earliest times men have sought to explain why the solid ground beneath their feet should suddenly and without apparent warning begin to shake in all manner of complicated ways. Very early these were mere speculations often bound up with the focal mythology. In Japan the shakings were supposed to be due to the movement of a subterranean mythical earth spider or jishin mushi which later in history became a cat-fish. There is a rock at Kashima some 60 miles northeast of Tokyo which was said to rest on the head of this creature to keep it quiet. In Mongolia the earth-shaker is a subterranean hog. In Scandinavia there was supposed to be a god of evil genius, by the name of Loki, who killed his brother Baldur. As punishment he was bound to a rock, face upwards, so that the poison of a serpent could drop on his face. Loki’s wife usually intercepted the poison in a dish but when she went to empty the dish the poison dropped on the prostrate deity causing him to writhe in agony and shake the earth. The philosophies of Aristotle, Pliny, and others, including the Chinese, concerning earthquakes, based on the fact that they had seen steam and other gases arising from volcanic vents, held that earthquakes were due to the workings of wind or imprisoned vapour beneath the earth’s crust.
In more recent times a great many more observations have been made and theories concerning earthquake origins must take account of all these multitudinous facts. It is now generally accepted that earthquakes have been known to have had their origins from points in the Earth’s surface down to points at depths of at least 364 miles. Light local earthquakes and earth tremors have been found to be due to the collapse of underground caverns and workings and to the slipping along local fault planes: for instance, those felt regularly at Comrie in Perthshire owing to the slipping of the Highland Boundary Fault, the mine shake of 12 November, 1938, at Brierley in Yorkshire and those felt on relatively rare occasions in the Settle district in Yorkshire owing to movements of the Craven Fault.
NORMAL AND DEEP FOCUS EARTHQUAKES
Minor earthquakes are also known to have been due to the activities of volcanoes such as those of the volcano Kusatu-Sirane from November 1937 to February 1939. The swarm of shocks at Montserrat in the West Indies are alternatively thought to be correlated with the geyser activity or the collapse of underground caverns in the limestone. Normal earthquakes, however, are at present thought to originate at depths of between 5.25 and 16 miles below the Earth’s surface, intermediate ones between depths of 32 and 160 miles and deep focus ones below this.
Professor Gutenberg, of the California Institute of Technology, is of the opinion that the large shear waves observed in deep focus earthquakes leave no doubt that the causative mechanism cannot be of an explosive character, but, as in the case of normal shocks, must depend on the release of strains. Dr KE Bullen, of Auckland University College, New Zealand, has noted that at depths of about 364 miles there appears to be a rapid increase in velocity of seismic waves, a rapid increase in the density of material, and an increase in electrical conductivity, all pointing to a change in the properties of the material of which the earth is composed at this depth.
Really deep focus earthquakes have been found to occur chiefly between the Japanese Islands and the Dutch East Indies, between Japan and a point to the north of New Zealand, and a few in western South America. Intermediate and normal earthquakes have been found to occur chiefly along the well established belts shown in the map.
These belts are conspicuously related to the surface geology and fall close to the tectonic lines of Tertiary or recent mountain building. In the East Indies they are close to the areas of large gravity anomalies.
It thus appears probable that the problem of the origin of normal earthquakes is bound up with the problem of the origin of mountain chains. From the time when it was first imagined that the Earth had originated from a molten body, the effect of the cooling of this was seen to produce shrinkage. Since the solid crust formed, this crust would have to adapt itself to the shrinking interior. The forces in the crust would be mostly horizontal, but, since the elastic properties of the crust are not uniform, the forces need not be wholly horizontal. As soon as these stresses become greater than the breaking strength of the material, the material ruptures and earthquakes result; should the viscous strength be reached first, plastic flow sets in and although mountains are produced no earthquake happens.
The discovery of radioactively generated heat just below the Earth’s surface has led some to throw doubt on the efficacy of this cooling process to produce mountains and thus earthquakes, at any rate during recent geological times.
Other long period forces discussed have been the movement of the continents (Polflucht of Wegener), tidal and other friction, secular movements of the Earth’s axis and changes in latitude. Changes which have occupied long periods of time so far as a human life is concerned, but short periods according to geological history, are: deviation from hydrostatic equilibrium of the crust; chemical processes; cosmic sources of energy; thermal differences between ocean bottoms and continents; changes in air-pressure; body tides; erosion; sedimentation; forming or melting of ice; variation in sea-level due to storms and tides; freezing of ground; seasonal changes in the temperature of the ground; vertical movements of blocks in the Earth’s crust and such like.
Various authors have discussed the possibility of convection currents within the region immediately under the Earth’s crust. Professor Daly, of Harvard University, has advanced a hypothesis whereby large blocks of the Earth’s crust slide under the action of gravitational forces to fold and thrust the more superficial, geosynclinal rocks thus forming the mountain chains and incidentally causing earthquakes. Science, requiring more data, still regards the long-range cause of earthquakes as an enigma. Whatever this cause or these causes may be, the process of the accumulation of strain prior to an earthquake is very insidious and it remains for some ‘trigger’ action at the last moment to precipitate the shock. This trigger action may be one of those forces already mentioned or in some cases it is an earthquake in another part of the world which sends out elastic waves in all directions. The second earthquake may then be termed a sympathetic earthquake. Some of these may have occurred in Los Angeles, in San Salvador, in the Rand near Johannesburg and in Tangier following the recent Turkish earthquake.
RECENT GREAT EARTHQUAKES
Some of the major earthquakes within living memory have undoubtedly been the Kwanto (Japan) earthquake of 1923 which nearly destroyed Tokyo and Yokohama; the Tango (Japan) earthquake of 1927; the Idu (Japan) earthquake of 1930; the Hawkes Bay (New Zealand) earthquake of 1931 ; in India, the Bihar earthquake of 1934 and the Quetta earthquake of 1935; the Chilean earthquake of 1939; and the Turkish earthquake of 1939. All of these occurred in well established seismic zones where lighter earthquakes are by no means uncommon. It is very rare for earthquakes even as great as the Baffin Bay earthquake of 1933, to occur in places other than the usual zones. These zones must therefore be considered as inherent lines of weakness in the Earth’s crustal and subcrustal structure where the numerous and varied forces acting within and without the world finally cause disruption of the crust.
The Hawkes Bay earthquake of 3 February, 1931, occurred without any warning at 10.47am New Zealand Summer Time in the region of Napier and Hastings. At Napier eyewitnesses saw buildings sway to an alarming extent, telegraph poles lean over at critical angles, cracks and fissures appear both in the walls of buildings and on the surface of streets, whilst control of motor cars became a matter of great difficulty owing to the peculiar nature of the Earth movement. Those who were not pinioned in buildings by the first shock had just reached the footpath when there came a second downward bump and many buildings collapsed. Owing to the scattering of fires in buildings numerous fires broke out in the town. About 260 people were killed and many were injured.
The Quetta earthquake of 31 May, 1935, was one of the most destructive to human life of all known Indian earthquakes. Chiefly owing to the badly constructed buildings, 26,000 out of a population of 40,000 were killed, and after the earthquake the city was cleared of survivors and ‘sealed’ for a year as a precaution against the spread of disease. The disturbed area covered a band 70 miles long and 15 miles wide, running from Quetta through Mastung to midway between Mastung and Kalat.
The Chilean earthquake of 24 January, 1939, occurred at about 11.35pm (local time). It had an epicentre near Chillan which was practically wrecked. The disastrously affected area reached Parral in the north and Concepcion in the west, and upwards of 20,000 people were killed, many more being injured. Such earthquakes as these have led to the formulation of schemes for earthquake insurance and to the better design of earthquake-proof buildings.
THE TURKISH DISASTER
The greatest earthquake ever recorded in Turkey happened at about 1.57am (local time) on 27 December, 1939, when an area in Anatolia including Erzinjan and Erbaa was completely devastated, and the surrounding district including Sivas and Tokat badly damaged. The loss of life was tremendous, estimates varying from the official figure of 23,131 dead and 7,994 injured to unofficial estimates of 45,000 killed and 20,000 injured. The high death roll was in part due to thick-walled mud houses of the locals collapsing on them whilst they were in bed.
This earthquake was recorded at observatories throughout the world on seismographs, many of which in effect are horizontal pendulums carrying a heavy inertia mass. The relative movement of the mass and the rest of the apparatus causes a pen at the end of the pendulum to write on paper fixed to a rotating drum the record of the earthquake. Time marks are also made on the paper. I am indebted to Rev. JP Rowland for the Stonyhurst record of the recent Turkish earthquake reproduced above. It was obtained with a Milne-Shaw seismograph registering optically with a galvanometer. The first waves to arrive have come through the body of the Earth. They are the P (Primary or Push) waves. Then come the S (Secondary or Shake) waves and finally the waves which have come round the surface of the Earth called L and M waves, for the knowledge of which we owe so much to Dr R Stoneley. It will be noticed that for this particular earthquake the surface waves are extremely large. They exceeded the limits of registration at Kew and unhinged the seismographs at West Bromwich, Faenza (Italy) and some of the Swiss observatories. Knowing the speeds at which the P and S waves travel, it is possible by observing the time interval between the arrival of these at an observatory to estimate the distance of the epicentre of the shock from the station. Doing the same for several observatories enables the epicentre to be determined approximately by drawing the arcual distances on a globe. Thence the original time of the earthquake may be determined.
Using observations from 15 observatories, Monsieur A Hee has found the epicentre of the Turkish earthquake to be near latitude 39°.5 N., longitude 38°.2 E. and the time when it happened to be near 23h. 57m. 23s. Greenwich Mean Time.
In any particular earthquake, should the arcual distances not quite intersect, the earthquake focus is known to be at some distance below the Earth’s surface and the depth of focus may be calculated. Should the records of the earthquake exhibit large P and S waves and no L and M (surface) waves the earthquake is thought to have had a very deep focus. In that case waves peculiar to deep focus earthquakes are also discernible, and the arcual distance together with the initial time and depth of focus may be found by using special graphs prepared by Rev. GJ Brunner, SJ, and Rev. JB Macelwane, SJ, of the St Louis University, USA.
By identifying other waves seen on seismograms, students of seismology have drawn conclusions concerning the properties of the materials deep down in the Earth and the structure of the Earth. It appears that if we take the Earth to be a sphere of nearly 4,000 miles radius, the core has a radius of approximately 2,200 miles and is made of more dense material than the remainder. The core of the Earth appears to possess some of the properties of a liquid in that, according to general acceptance, it will not transmit S waves directly through the centre. Waves from an earthquake near to an observatory enable the crustal structure to be determined. Underneath the sedimentary surface rocks of variable thickness there appears to be from the surface downwards first a layer of granitic material estimated from 6.25 to 7.75 miles thick, then a layer of intermediate material from 9.5 to 10.5 miles thick, and underneath that basaltic material becoming more dense towards the core, with possible discontinuities.
Artificial earthquakes caused by exploding charges of dynamite sunk in the ground have been used to determine the thickness of the ice-cap in Greenland and other places; to determine the position and size of salt domes in prospecting for oil; and to determine the geological structure including the depth of the sedimentary rocks in various parts of the world. This work is done by portable seismographs and in the case of prospecting for oil has saved the companies thousands of pounds in avoiding unnecessary drilling.
Get the best of Geographical delivered straight to your inbox by signing up to our weekly newsletter and get a free collection of eBooks!