THE INTERRUPTED CHARACTER OF EARTH MOVEMENTS: EARTHQUAKES AND SEAQUAKES (Concluded)

Experimental demonstration of earth movements.—The study of the Alaskan earthquake of 1899 showed that during this adjustment within the earth’s shell some of the local blocks moved upward and by larger amounts than their neighbors, and that still others were actually depressed so that the sea flowed over them. It must be evident that such differential vertical movements of neighboring blocks at the earth’s surface can only take place if lateral transfers of material are made beneath it. From under those strips of coast land which were depressed, material must have been moved so as to fill the void which would otherwise have formed beneath the sections that were uplifted. If we take into consideration much larger fractions upon the surface of our planet, we are taught by the great seaquakes which are now registered upon earthquake instruments at distant stations that large downward movements are to-day in progress beneath the sea much more than sufficient to compensate all extensions of the earth’s surface within those districts where the land is rising in mountains. From under the offshore deeps of the ocean to beneath the growing mountains upon the shore, a transfer of earth material must be assumed to take place when disturbances are registered.

Within the time interval that separates the sudden adjustments of the surface which are manifested in earthquakes, the condition of strain which brings them about is steadily accumulating, due, as we generally assume, to earth contraction through loss of its heat. It seems probable that the resistance to an immediate adjustment is found in the rigidity of the shell because of the compression to which it is subjected. To illustrate: a row of blocks well fitted to each other may be held firmly as a bridge between the jaws of a vice, because so soon as each block starts to fall a large resistance from friction upon its surface is called into existence, a force which increases with the degree of compression.

It is thus possible upon this assumption crudely to demonstrate the adjustment of earth blocks by the simple device represented in plate 4 A. The construction of this experimental tank is so simple that little explanation is necessary. Wooden blocks of different heights are supported in water within a tank having a glass front, and are kept in a strained condition at other than their natural positions of flotation by the compression of a simple vice at the top. Held firmly in this position, they may thus represent the neighboring blocks within the earth’s outer shell which are supported upon relatively yielding materials beneath, and prevented from at once adjusting themselves to their natural positions through the compression to which they are subjected. Held as they now are, the water near the ends of the tank is forced up beneath the blocks to higher than its natural level, and thus tends to flow from both ends toward the center. Such a movement would permit the end blocks to drop and force the middle ones to rise. The end blocks are, let us say, the sections of Alaskan coast line which sunk during the earthquake, as the center blocks are the sections which rose the full measure of 47 feet. Upon a larger scale the end blocks may equally well be considered as the floor of the great deeps off the Alaskan coast, whose sinking at the time of the earthquake was the cause of the great sea wave. Upon this assumption the center blocks would represent the Alaskan coast regarded as a whole, which underwent a general uplift.

Though we may not, in our experiment, vary the tendency to adjustment by any contractional changes in either the water or the blocks, we may reduce the compression of the vice, which leads to the same general result. As the compression of the vice is slowly relaxed, a point is at last reached at which friction upon the block surfaces is no longer sufficient to prevent an adjustment taking place, and this now suddenly occurs with the result shown in plate 4 B. In the case of the earth blocks, this sudden adjustment is accompanied by mass movements of the ground separated by faults, and these movements produce successional vibrations that are particularly large near the block margins, and other frictional vibrations of such small measure as to be generally appreciated by sounds only. The jolt of the adjustments has thrown some blocks beyond their natural position of rest, and these sink and rise subsequently in order to readjust themselves with lighter vibrations, which may be repeated and continued for some time. In the case of the earth these later adjustments are the so-called aftershocks which usually continue throughout a considerable period following every great earthquake. Gradually they fall off in intensity and frequency until they can no longer be felt, and are thereafter continued for a time as rumblings only.

Derangement of water flow by earth movement.—The water which supported the blocks in our experiment has represented the more mobile portion of the earth’s substance beneath its outer zone of fracture. The surface water layers in the tank may, however, be considered in a different way, since their behavior is remarkably like that of the water within and upon the earth’s surface during an earth adjustment. At the instant when adjustment takes place in the tank, water frequently spurts upward from the cracks between the sinking end blocks; and if in place of one of the higher center blocks we insert one whose top is below the level of the water in the tank, a “lake” will be formed above it. When the adjustment occurs, this lake is immediately drained by outflow of the water at its bottom along one of the cracks between the blocks (Fig. 76).

Such derangements of water flow as have been illustrated by the experiment are among the commonest of the phenomena which accompany earthquakes. Lakes and swamp lands have during earthquakes been suddenly drained, fountains of water have been seen to shoot up from the surface and have played for some minutes or hours before their sudden disappearance in a sucking down of the water with later readjustment. During the great earthquake of the lower Mississippi valley in 1811, known as the New Madrid earthquake, the earlier Lake Eulalie was completely drained, and upon the now exposed bed there appeared parallel fissures on which were ranged funnel-like openings down which the water had been sucked. In other sections of the affected region the water shot up in sheets along fissures to the tops of high trees. Areas where such spurting up of the water has been observed have in most cases been shown to correspond to areas of depression, and such areas have sometimes been left flooded with water. During the Indian earthquake of 1819 an area of some 200 square miles suddenly sank and was transformed into a lake.

Sand or mud cones and craterlets.—From a very moderate depth below the surface to that of several miles, all pore spaces and all larger openings within the rock are completely filled with water, the “trunk lines” of whose circulation is by way of the joints or along the bedding planes of the rocks. The principal reservoirs, so to speak, of this water inclosed within the rock are the porous sand formations. When, now, during an earthquake a block of the earth’s shell is suddenly sunk and as suddenly arrested in its downward movement, the effect is to compress the porous layers and so force the contained water upward along the joints to the surface, carrying with it large quantities of the sand (Fig. 77).

Ejected at the surface this water appears in fountains usually arranged in line over joints, or even in continuous sheets, and the sand collecting about the jets builds up lines of sand or mud cones sometimes described as “mud volcanoes” (Fig. 78). The amount of sand thus poured out is sometimes so great that blankets of quicksand are spread over large sections of the country. Most frequently, however, the sand is not built above the general level of the surface, but forms a series of craterlets which are largely shaped as the water is sucked down at the time of the readjustment with which the play of such earthquake fountains is terminated (Fig. 79). Subsequent excavations made about such craterlets have shown them to have the form of a trumpet, and that in the sand which so largely fills them there are generally found scales of mica and such light bodies as would be picked out from the heterogeneous materials of the sand layers and carried upward in the rush of water to the surface (Fig. 80).

The earth’s zones of heavy earthquake.—Since earthquakes give notice of a change of level of the ground, the special danger zones from this source are the growing mountain systems which are usually found near the borders of the sea. Such lines of mountains are to-day rising where for long periods in the past were the basins of deposition of former seas. They thus represent the zones upon the earth’s surface which are the most unstable—which in the recent period have undergone the greatest changes of level.

By far the most unstable belt upon the earth’s surface is the rim surrounding the Pacific Ocean, within which margin it has been estimated that about 54 per cent of the recorded shocks of earthquake have occurred. Next in importance for seismic instability is the zone which borders both the Mediterranean Sea and the Caribbean—the American Mediterranean—and is extended across central Asia through the Himalayas into Malaysia. Both zones approximate to great circles upon the earth’s surface and intersect each other at an angle of about 67°. It has been estimated that about 95 per cent of the recorded continental earthquakes have emanated from these belts.

The special lines of heavy shock.—Within any earthquake district the shocks are not felt with equal severity at all places, but there are, on the contrary, definite lines which the disturbance seems to search out for special damage. From their relations to the relief of the land these lines would appear to be lines of fracture upon the boundaries of those sections of the crust that play individual rÔles in the block adjustment which takes place. More or less masked as these lines are beneath the rounded curves of the landscape, they are given an altogether unenviable prominence with each succeeding earthquake. At such times we may think of the earth’s surface as specially sensitized for laying bare its hidden structure, as is the sensitized plate under the magical influence of the X rays.

When, at the time of an earthquake, blocks are adjusted with reference to their neighbors, the movements of oscillation are greatest in those marginal portions of direct contact. Corners of blocks—the intersecting points of the important faults—should for the same reason be shaken with a double violence, and this assumption appears to be confirmed by observation. Upon the island of Ischia, off the Bay of Naples, the shocks from recent earthquakes have been strangely concentrated near the town of Casamicciola, which was last destroyed in 1883. This unfortunate city lies at the crossing point of important fractures whose course upon the island is marked by numerous springs and suffioni (Fig. 81).

Seismotectonic lines.—The lines of important earth fractures, as will be more clearly shown in the sequel (p. 227), are often indicated with some clearness by straight lines in the plan of the surface relief (Fig. 82). Lines of this nature are easily made out upon the map of the West Indies, and if we represent upon it by circles of different diameters the combined intensities of the recorded earthquakes in the various cities, it appears that the heavily shaken localities are ranged upon lines stamped out in the relief, with the most severely damaged places at their intersections (Fig. 83). These lines of exceptional instability are known as seismotectonic lines—earthquake structure lines.

The heavy shocks above loose foundations.—It is characteristic of faults that they soon bury themselves from sight under loose materials, and are thus made difficult of inspection. The escarpment which is the direct consequence of a vertical displacement upon a fault tends to migrate from the place of its formation, rounding the surface as it does so and burying the fault line beneath its deposits (Fig. 43, p. 60).

This is not, however, the sole reason why loose foundations should be places of special danger at the time of earth shocks, for the reason that earthquake waves are sent out in all directions from the surfaces of displacement through the medium of the underlying rock. These waves travel within the firm rock for considerable distances with only a gradual dissipation of their energy, but with their entry into the loose surface deposits their energy is quickly used up in local vibrations of large amplitude, and hence destructive to buildings.

The essential difference between firm rock and such loose materials as are found upon a river bottom or in the “made land” about our cities may be illustrated by the simple device which is represented in Fig. 84. Two similar metal pans are suspended from a firm support by bands of steel and “elastic” braid of similar size and shape, and carry each a small block of wood standing upon its end. Similar light blows are now administered directly to the pans with the effect of upsetting that block which is supported by the loose braid because of the large range or amplitude of movement that is imparted to the pan. The “elastic” braid, because of these large vibrations of which it is susceptible, may represent the loose materials when an earthquake wave passes into them. In the case of the steel support, the energy of the blow, instead of being dissipated in local swingings of the pan, is to a large extent transmitted through the elastic metal to materials beyond. The steel thus resembles in its high elasticity the firmer rock basement, which receives and transmits the earthquake shocks, but except when ruptured in a fault is subject to vibrations of small amplitude only.

Construction in earthquake regions.—Wherever earthquakes have been felt, they are certain to occur again; and wherever mountains are growing or changes of level are in progress, there no record of past earthquakes is required in order to forecast the future seismic history. Although the future earthquakes may be predicted, the time of their coming is, fortunately or unfortunately, still hidden from us. If one’s lot is to be cast in an earthquake country, the only sane course to pursue is to build with due regard to future contingencies.

The danger, from destructive fires may to-day be largely met by methods of construction which levy an additional burden of cost. Though the danger from seismic disturbances can hardly be met as fully as that from fire, yet it is true that buildings may be so constructed as to withstand all save those heaviest shocks in the immediate vicinity of the lines of large displacement. Here, also, a considerable additional expense is involved in the method of construction, in the case of residences particularly.

From what has been said, it is obvious that much of the danger from earthquakes can be met by a choice of site away from lines of important fracture and from areas of relatively loose foundation. The choice of building materials is next of importance. Those buildings which succumb to earthquakes are in most cases racked or shaken apart, and thus they become a prey to their own inherent properties of inertia. Each part of a structure may be regarded as a weight which is balanced upon a stiff rod and pivoted upon the ground. When shocks arrive, each part tends to be thrown into vibration after the manner of an inverted pendulum. In proportion, therefore, as the weights are large and rest upon long supports, the danger of overthrow and of tearing apart is increased. In general, structures are best constructed of light materials whose weight is concentrated near the ground. Masonry structures, and especially high ones, are, therefore, the least suited for resisting earthquakes, of which the late complete destruction of the city of Messina is a grewsome reminder. Despite repeated warnings in the past, the buildings of that stricken city were generally constructed of heavy rubble, which in addition had been poorly cemented (Fig. 49, p. 67). Such structures are usually first ruptured at the edges and corners, since here the vibrations which tend to tear the building asunder are resisted by no supports and are reËnforced from neighboring walls.

An advantage of the first importance is evidently secured if the rods of the pendulum, of which the building is conceived to be composed, have sufficient elasticity to be considerably distorted without rupture and to again recover their original position. This is the supreme advantage of structural steel for all large buildings, which is coupled, however, with the disadvantage that the riveted fastenings are apt to be quickly sheered off under the vibrations. Large and high buildings, when sufficiently elastic, have fortunately the property of destroying the earth waves by interference before they have traveled above the lower stories.

For large structures in which wood cannot be used, strongly reËnforced concrete is well adapted, for it has in general the same advantages as steel with somewhat reduced elasticity, but with a more effective binding together of the parts. This requirement of thorough bracing and tying together of the several parts of a building causes it to vibrate, not as many pendulums, but as one body. If met, it removes largely the danger from racking strains, and for small structures particularly it is the requirement which is most easily complied with. For such buildings it is therefore necessary that the framework should be built in a close network with every joint firmly braced and with all parts securely tied together. Especial attention should be given to the fastenings of floor and partition ends. The house shown in Fig. 85 could not have been subjected to heavy shocks, for though the walls are thrown down, the floors and partitions have been left near their original positions.

This tendency of the walls, floors, partitions, and roof to act as individual units in the vibration, is one that must be reckoned with and be met by specially effective bracing and tying at the junctions. Otherwise these larger parts of the structure may act like battering rams to throw over the walls or portions of them (Fig. 86).

Reading References for Chapters VII and VIII

General works:—

John Milne. Seismology. London, 1898, pp. 320.

C. E. Dutton. Earthquakes in the Light of the New Seismology. Putnam, New York, 1904, pp. 314.

A. Sieberg. Handbuch der Erdbebenkunde. Braunschweig, 1904, pp. 362.

Count F. de Montessus de Ballore. Les Tremblements de Terre, GÉographie SÉismologique. Paris, 1906, pp. 475; La Science SÉismologique. Paris, 1907, pp. 579.

William Herbert Hobbs. Earthquakes, an Introduction to Seismic Geology. Appleton, New York, 1907, pp. 336.

C. G. Knott. The Physics of Earthquake Phenomena. Clarendon Press, Oxford, 1908, pp. 283.

E. Rudolph. Ueber Submarine Erdbeben und Eruptionen, BeitrÄge zur Geophysik, vol. 1, 1887, pp. 133-365; vol. 2, 1895, pp. 537-666; vol. 3, 1898, pp. 273-536.

Descriptive reports of some important earthquakes:—

C. E. Dutton. The Charleston Earthquake of August 31, 1886, 9th Ann. Rept. U. S. Geol. Surv., 1889, pp. 203-528.

B. KotÔ. On the Cause of the Great Earthquake in Central Japan, 1891, Jour. Coll. Sci. Imp. Univ., Tokyo, Japan, vol. 5, 1893, pp. 295-353, pls. 28-35.

John Milne and W. K. Burton. The Great Earthquake of Central Japan. 1891, pp. 10, pls. 30.

R. D. Oldham. Report on the Great Earthquake of 12th June, 1897, Mem. Geol. Surv. India. Vol. 29, 1899, pp. 379, pls. 42.

A. C. Lawson, and others. The California Earthquake of April 18, 1906, Report of the State Earthquake Investigation Commission, three quarto vols. (Carnegie Institution of Washington); many plates and figures.

Italian Photographic Society, Messina and Reggio before and after the Earthquake of December 28, 1908 (an interesting collection of pictures). Florence, 1909.

R. S. Tarr and L. Martin. Recent Changes of Level in the Yakutat Bay Region, Alaska, Bull. Geol. Soc. Am., vol. 17, 1906, pp. 29-64, pls. 12-23.

William Herbert Hobbs. The Earthquake of 1872 in the Owens Valley, California, BeitrÄge zur Geophysik, vol. 10, 1910, pp. 352-385, pls, 10-23.

Faults in connection with earthquakes:—

William H. Hobbs. On Some Principles of Seismic Geology, BeitrÄge zur Geophysik, vol. 8, 1907, Chapters iv-v.

Expansion or contraction of the earth’s surface during earthquakes:—

William H. Hobbs. A Study of the Damage to Bridges during Earthquakes, Jour. Geol., vol. 16, 1908, pp. 636-653; The Evolution and the Outlook of Seismic Geology, Proc. Am. Phil. Soc., vol. 48, 1909, pp. 27-29.

Earthquake construction:—

John Milne. Construction in Earthquake Countries, Trans. Seis. Soc., Japan, vol. 14, 1889-1890, pp. 1-246.

F. de Montessus de Ballore. L’art de bÂtir dans les pays À tremblements de terre (34th Congress of French Architects), L’Architecture, 193 AnnÉe, 1906, pp. 1-31.

Gilbert, Humphrey, Sewell, and SoulÉ. The San Francisco Earthquake and Fire of April 18, 1906, and their Effects on Structures and Structural Materials, Bull. 324, U. S. Geol. Surv., 1907, pp. 1-170, pls. 1-57.

William H. Hobbs. Construction in Earthquake Countries, The Engineering Magazine, vol. 37, 1909, pp. 1-19.

Lewis Alden Estes. Earthquake-proof Construction, a discussion of the effects of earthquakes on building construction with special reference to structures of reËnforced concrete, published by Trussed Concrete Steel Company. Detroit, 1911, pp. 46.

---