CHAPTER VII

INSTABILITY OF THE EARTH'S CRUST

The crust of the earth is unstable. To the modern student of geology the old notion of a "terra firma" is outworn. The idea of an unshakable, immovable earth could never have emanated from the inhabitants of an earthquake country. In general we may recognize two types of crustal movements--slow and sudden. To most people the sudden movements accompanied by earthquakes are more significant and impressive because they are more localized and evident, and often accompanied by destruction of property, or quick, though minor, changes in the landscape. But movements which take place slowly and quietly are often of far greater significance in the interpretation of the profound physical changes which have affected the earth during its millions of years of known history.

A few well-known examples will serve to prove that upward, downward, and differential movements of the earth's crust have actually taken place not only in the remote ages of geologic time, but also that such movements have geologically recently taken place, and that similar movements are still going on. It is very important that the reader thoroughly appreciate the fact that crustal disturbances, often profound ones, do take place, because this is one of the most fundamental tenets of geologic science. Let us consider the case of the Hudson-Champlain-St. Lawrence Valley region. That the whole region was once notably higher (at least 1,000 feet) than at present is proved by the drowned character of the Hudson Valley, in which tidewater extends northward for 150 miles to near Troy. Where the New York City Aqueduct passes under the Hudson River near Newburgh, the bedrock bottom of the old river channel is now about 800 feet below sea level as determined by drilling. This old channel is there filled up nearly to sea level with glacial and postglacial rock débris, which shows that the old channel must have been cut before the oncoming of the ice of the great Ice Age. Before the Ice Age, then, the lower Hudson Valley must have been considerably more than 800 feet higher than at present, because it then contained a river with sufficient current to be an active agent of erosion, carving out the canyonlike valley in the vicinity of West Point. This conclusion is strongly reenforced by the fact that the old valley of the Hudson River has been definitely traced as a distinct trench across the shallow sea bottom for about 100 miles eastward from the entrance to New York harbor. Toward the eastern end of this trench the depth of water is now considerably over 1,000 feet, and thus it is obvious that, preceding the Ice Age, the earth's crust in the vicinity of New York City must have been much higher than at present, so that the Hudson River was able to erode its now completely drowned channel. Somewhat similar evidence has also established the fact that the lower St. Lawrence Valley region was much higher before the Ice Age. It is evident, therefore, that the general Hudson-St. Lawrence Valley region is now notably lower with reference to sea level than it was before the Ice Age. That this was caused by actual sinking of the earth's crust rather than by a rise of sea level is proved by the fact that similar changes of level between land and sea did not take place at the same time even along the Atlantic and Gulf coast of our Southern States.

We shall now proceed to the next step in the geologically recent history of earth-crust movements in the Hudson-Champlain-St. Lawrence Valley region by asserting that, since the Ice Age, the land was actually notably lower than at present. In fact, the land was enough lower to allow tidewater to extend up the St. Lawrence Valley into the Ontario basin, and all through the Champlain-Hudson Valley. Many beaches, bars, and delta deposits formed in these arms of the sea are still plainly preserved, in some cases with shells and bones of marine animals in them, now hundreds of feet above sea level. These marine deposits are highest above sea level in the northern portion of the Champlain Valley, where they lie at an altitude of 700 feet or more and their altitude steadily diminishes southward to about 300 to 400 feet in the general vicinity of Albany, and to near sea level in the general vicinity of New York City. Obviously, then, the land stood lower during part or all of the interval of not more than a few tens of thousands of years since the Ice Age than at present. This leads us to the third important conclusion regarding earth movements in this region, namely, that still later the land has undergone a differential uplift, the rate having steadily increased toward the north where the total uplift is many hundreds of feet. We have discussed this region somewhat in detail because the principles of slow up and down movements of the earth's crust are there so plainly recorded.

Among many other regions where earth movements similar to those above described have taken place, brief mention may be made of Norway. The great fjords of Norway were, just before the Ice Age, stream-cut valleys which were then more or less modified by glacial erosion, and after the Ice Age the rivers in them were drowned due to land subsidence. The kind of evidence is like that above given for the lower Hudson River. Since the subsidence there has been partial reelevation, as proved by the fact that along the sides of the larger fjords marine terraces and beaches may be traced with gradually increasing altitude for many miles (150 or more) back into the country where they are hundreds of feet above tidewater.

Scandinavia is of still further special interest because very appreciable earth movements have there come under human observation. Marks carefully placed along the shores of Sweden by the government have proved that during the last 150 years the southern end of the country has actually subsided several feet, while from Stockholm north the land has risen in increasing amount, reaching a maximum of seven or eight feet. In southern Sweden, at Malmo, a certain street now at times becomes covered by wind-driven high water, and during excavations made some years ago an older street eight feet below the present one was found.

A theory which appears to be in perfect harmony with the facts to account for the subsidence and partial reelevation of central eastern North America and Scandinavia since the beginning of the Ice Age is that the great weight of ice during the Ice Age pressed the land down, and that since the removal of the ice there has been an appreciable tendency for the land to spring back.

Certain crustal movements which have occurred about the Bay of Naples are of very special interest because actual human history dates can be placed upon them. Most remarkable are the records in connection with the temple of Jupiter Serapis which was built near the shore before the Christian era. The land sank about five feet and a new pavement had to be constructed; then, by the middle of the third century A. D., the temple rose to well above sea level. By about the ninth century the land had subsided fully thirty feet, so that marble columns of the temple were bored full of holes as high as twenty-one feet above their bases by marine-shelled animals, species of which still live in the bay. Then a slow uplift of twenty-three feet began, bringing the bases of the columns two feet above sea level by 1749. Since that time a slight sinking has taken place and this seems to be still going on. Three of the marble columns with the borings still stand in upright position.

While the movements just described were taking place, the island of Capri, twenty miles across the Bay of Naples, has slowly sunk to an amount estimated at thirty or forty feet as proved by evidence from the famous Blue Grotto. About the beginning of the Christian era a large ancient wave-cut cave, part of which is now called the Blue Grotto, had its floor above sea level, and it was used by certain Romans as a cool place to retire to from the heat. In order to obtain better light an opening was cut through its upper portion. The land has sunk so much that at the present time even part of the artificial opening (through which tourists pass) is now under water.

By way of illustrating remarkable contrasts in direction of crustal movements on very considerable scales in a given region, we shall briefly mention some facts regarding part of the coast of southern California and the neighboring islands of Santa Catalina and San Clemente, respectively twenty-five and fifty miles offshore. Those movements were not, however, checked up by human history records. The mainland at San Pedro has clearly risen 1,240 feet, as proved by the presence of unusually perfect coast terraces (so-called "raised beaches"), while San Clemente has risen 1,500 feet as proved by the raised beaches into which deep, youthful V-shaped stream-cut valleys have been sunk, and a shore line characteristic of recent notable uplift. It is a remarkable fact that at the same time the intervening island (Santa Catalina) has notably sunk, as proved by the nature of its shore line, and the distinctly more mature character of its topography.

We are, however, by no means dependent upon lands along sea shores for evidences of slow rising and sinking of land. Thus, by careful measurements it has been shown that the general region of the Great Lakes is now differentially rising toward the northeast at the rate of about five inches per 100 miles per century. At Chicago the rise of water is estimated at about nine inches per century, which means increase of flowage through the Chicago Canal. At this rate the upper lakes would, in some thousands of years, drain through this canal to the Mississippi. A well-preserved shore line of the large ancestor of Lake Ontario shows a steady increase in altitude at the rate of several feet per mile toward the northeast from near Niagara to the St. Lawrence Valley, thus proving a tilting of the land since the shore line was formed.

Shore lines of the great ancestor of Great Salt Lake also show warping of the earth's crust, some parts of a definite shore line being several hundred feet higher than others.

Very significant evidence pointing to profound crustal movements consist in the finding of fossil remains of marine animals in the strata high above sea level, very commonly from one to three miles, in many parts of the world, especially in the high mountains. In Wyoming, nearly horizontal strata of the Mesozoic Age carrying marine fossils lie two miles or more above sea level. The fact that given formations, carrying marine fossils representing certain definite portions of geologic time, are found at various altitudes up to several miles in many parts of the world, shows that the land in those places has really risen relative to sea level.

It should not be presumed from the above discussion that the sea level itself has never changed. Thus, the vast areas of thick ice sheets in both North America and Europe during the Great Ice Age represented sufficient water withdrawn from the sea to very appreciably lower its level. All land-derived materials, carried into the sea mainly by rivers, displace sea water, with consequent rise of its level. If all existing lands were worn down and carried into the sea, its level would be raised some hundreds of feet. Subsidence of any part of the ocean bottom would cause a lowering of sea level. There is a strong reason to believe that some such shiftings of sea level have occurred during the vast lapse of geologic time. During certain periods erosion of the land predominated, and during other periods building up of the land predominated, as pointed out in the chapters on geologic history. It is not thought that shifting of sea level has ever amounted to more than a few hundred feet, at least not during the millions of years of the more clearly recorded earth history.

We have thus far considered slow upward and downward movements of the earth's crust without notable structural changes in the rocks. Another type of crustal disturbance causes more or less profound changes in the structures of the rocks themselves. Just how the earth originated is a matter of uncertainty, but we can be sure that for many millions of years it has been a shrinking body. The outer, or crustal, portion of the earth, in adjusting itself to the contracting interior, has had many pressures, stresses, and strains set up within it. As results of such forces the rocks at and near the earth's surface have in various places, and at various times, been broken (faulted) and subjected to sudden movements (see discussion beyond), while those well within the crustal portion, that is to say a few miles or more down, have, in many cases, been bent (folded), or even crumpled. For these reasons the surface and near-surface crustal portions are called the "zone of fracture," while the more deeply buried portions comprise the "zone of flowage." In the zone of flowage the rocks, where subjected to great lateral pressure, act like plastic materials and therefore bend rather than break, because of the great weight of overlying materials. Laboratory experiments have confirmed the findings of geologists in this regard. Small masses of rocks properly inclosed in nickel-steel cylinders have been subjected to slow differential pressures equivalent to those which obtain twenty to forty miles within the earth. Under such conditions rocks have been made to change shape very notably without fracturing. Both geological observations and experiments have led us to conclude that not even small fractures or crevices can remain open at a depth greater than ten or twelve miles even in the hardest rocks.

From time to time, during the long history of the earth, forces of lateral pressure have been slowly exerted along more or less localized zones or belts within the earth's crust, and the rocks have been deformed chiefly by bending or folding, especially in those regions where mountains of the folded type have developed. Movements of this type are considered beyond in the chapter on mountains. Rock folds vary in size from microscopic to miles across, and they exhibit many shapes. Plate 7 will give the reader a good idea of actual rock folds of common sizes and shapes in various places. Folded structures are most clearly discernible in sedimentary rocks, because of their stratified (layered) arrangement. Since folds in hard rocks rarely, if ever, develop except at a depth of some miles within the earth, they show at the surface only where great thicknesses of overlying materials have been stripped off by erosion.

From the standpoint of our consideration of slow earth-crust movements, it is important to bear in mind that lateral pressure in the zone of flowage has not only notably deformed rocks, but that, as a result of the buckling forces, given rock masses have, in many cases, been notably shifted downward or upward--mainly upward--from their original positions. Folded strata carrying shells of sea animals are commonly found thousands of feet above sea level in many of the great mountain ranges of the world. During the process of folding on a large scale the crust of the earth is very appreciably shortened at right angles to the direction of applied pressure, due to squeezing or bending of the strata. In the case of the Appalachian mountains of Pennsylvania it has been estimated that such shortening amounts to about twenty-six miles or, in other words, that the strata originally spread out horizontally across an area whose width was about 100 miles have been squeezed or folded into an area whose width is twenty-six miles less.

We shall now turn to a consideration of sudden earth movements and some of their effects, including earthquakes. Mention has already been made of the fact that, when pressures and strains are set up in the outer portion ("zone of fracture") of the earth's crust, the rocks yield mainly by breaking or fracturing because the rocks not being under a great load of overlying material are there brittle. The earth's crust has been fractured on small and large scales in many places during the long space of geologic time. Where one block of earth's crust has slipped or moved past another along a fracture we have what is called a "fault." Such displacements of rock masses vary in amount from less than an inch to some miles, and they constitute one of the most important features of the architecture of the outer portion of the earth. There are two types of faults fundamentally different as to cause. In one type (so-called "normal fault") the rocks suddenly yield to a force of tension; a fracture develops and the earth block on one side of the fracture or fault drops with reference to that on the other. In the other type (so-called "thrust faults") the rocks yield suddenly to a force of compression or lateral thrust, and one block of earth is pushed or thrust partly over another along the surface of fracture or fault. (See Plate 8.)

Faults range in length up to hundreds of miles, those from one to twenty miles in length being very common. Where an earth block has been displaced thousands of feet along a fault surface, it is not to be understood that the whole displacement resulted from a single movement, but rather from a series of sudden movements separated by greater or less intervals of time. Each sudden movement along a fault surface produces a vibration of the earth near by. Many such sudden movements are known to have caused violent earthquakes. Displacements of twenty to fifty feet, as a result of single movements, are definitely known to have taken place in various regions during the last fifty years; and rarely, if ever, has any sudden displacement of as much as several hundred feet occurred. Cliffs and steep slopes very commonly result from faulting, but, because of the long lapse of time required for the repeated movements in the case of great faults, the cliffs or steep slopes begin to wear back and become more or less subdued long before the last of the movements take place. In regions where movements along great faults have long since ceased, the original steep slopes may be completely obliterated by erosion.

How does the geologist determine the actual amount of displacement, especially in the case of a large fault in stratified rocks? First, the various formations of the region, where unaffected by faulting, are carefully studied, especially in regard to the character and thickness of each, and their relative geologic ages or normal order as they were deposited one layer above the other. Then, in the simple case of a normal-fault surface at right angles to horizontal strata, it is only necessary to find out what two formations or parts of formations come together along the fault fracture, and the actual amount of displacement is readily determined. Where strata and normal fault surfaces lie at various angles, and also in thrust faults, those angles must be determined in addition to the data above named. In many mining regions, where valuable deposits are affected by faulting, accurate knowledge of the direction and amount of displacements of faults is of great economic importance.

A few examples of normal faults from well-known districts will now be briefly described. The whole eastern front of the central and southern Sierra Nevada Range of California is a great, steep fault slope, from a few thousand to ten thousand or more feet high and hundreds of miles long, of such recent geologic age that it has been only moderately affected by erosion. In fact, it is well known that the southern two-thirds of the range is a great tilted fault block, the total displacement having resulted from repeated sudden movements since about the middle of the present geologic era. A great fault also extends along the eastern base of the great Wasatch Range of Utah and the steep slope thousands of feet high is a fault scarp only slightly modified by erosion. Renewed movements along this profound fault have very recently taken place as proved by the presence of fresh fault scarps in loose deposits which have accumulated across the mouths of some of the canyons, as, for example, near Ogden. In fact, practically all of the north-south ranges of the Great Basin from Utah to California are essentially a series of tilted fault blocks. Another great fault, less conspicuous from the topographic standpoint, is hundreds of miles long in the Coast Range Mountains of California. At the time of the San Francisco earthquake of 1906 there was a renewed sudden movement along this great fracture. The eastern one-half of the Adirondack Mountains of New York is literally a mosaic of hundreds of fault blocks. Many of these faults are from two to thirty miles long and they commonly show displacements of from a few hundred to 2,000 or more feet. A glance at the geological map (in colors) of the vicinity of the great copper mines at Bisbee, Arizona, shows most of that region to contain a network of normal faults which separate it into a mosaic of fault blocks. In each of the examples of faults just given a block of earth has sunk relative to the other, or in other words, each is a "normal fault."

We shall now turn to some large scale cases of faults in which great masses of earth have been pushed one over another--so-called "thrust faults." In the southern Appalachian Range, and especially well exhibited in the vicinity of Rome, Georgia, one portion of the mountain mass has literally been shoved over another, at a low angle over a fault surface many miles long, for fully seven miles westward. Both the tremendous weight of rock material actually translated and the number of sudden movements required in the operation stagger the imagination. It is safe to say that during the long time of this great operation violent earthquakes were not uncommon. In the Rocky Mountains of the northern United States and southern Canada there is the greatest known thrust fault on the continent. It is hundreds of miles long, and the actual displacement is commonly at least several miles. In the Glacier National Park of Montana it has been established that the front range portion of the Rockies has actually been pushed at least seven miles, and possibly as much as twenty miles, eastward over a fault surface, and out upon the Great Plains. In some cases rocks of the Prepaleozoic Age have there been pushed upon rocks of the late Mesozoic Age, thus locally upsetting the geologic column.

The Wasatch Range of Utah, in addition to the great normal fault along its western base, contains a remarkable system of thrust faults. In the region now occupied by the Wasatch Mountains a number of parallel (thrust) faults were developed close together and the broken pieces of the earth's crust between them were pushed up, the rocks on one side of each crack riding up over those on the other side until a great mountain range was formed where once lay a plain. In the Ogden Canyon one great earth block of Prepaleozoic (Algonkian) Age has been shoved thousands of feet over late Paleozoic (Carboniferous) rock, which latter has in turn been thrust over early Paleozoic (Cambrian) rock. This thrust faulting was accomplished before the development of the geologically recent normal fault along the western base of the range.

Any sudden movement of part of the crust of the earth, due to a natural cause, produces a trembling or shaking called an earthquake. Though earthquakes are generally classed among the most terrifying of all natural phenomena, those which have occurred during human historic times have had scarcely any geological or topographical effects of real consequence on the face of the earth. Locally, the effects may be notable and the destruction of life and property may be great. The earth may be locally cracked and rent, small fault scarps may develop, landslides and avalanches may result from the shaking of the earth, buildings may be demolished, and sea waves may be rolled upon the land. On the other hand, many earthquakes, called "tremors," are too slight to be noticed by people, though they are recorded by specially constructed instruments called "seismographs." We have already stated that actual sudden displacements causing earthquakes have amounted to twenty or even fifty feet right along fault fractures, but during the vibrations or quakings, which are often so destructively sent out into the neighboring country, the earth's surface rarely actually moves more than a small fraction of an inch. Because of the suddenness of the movement objects on the surface may be moved inches or even feet. Violent shocks may last one or two minutes and cause the whole earth to tremble, though at distant points only seismographs record the movement. It is probably true that some part of the earth is shaking all the time.

Studies during the last fifty years have made it certain that the main cause of earthquakes is the sudden slipping of earth blocks past each other along fault fractures, the sudden slipping furnishing the impulse which sends out the vibrations into the surrounding more or less elastic crust of the earth. The low rumbling to roaring sound, which sometimes immediately precedes an earthquake, is probably due to the grinding of the rocks together below the surface.

Earthquakes generally accompany volcanic outbursts of the violent or explosive type, and in such cases subterranean explosions cause both the eruptions and the quakings of the earth. It is well known that the principal volcanic districts or belts of the earth are also the belts of most frequent earthquakes, but this does not mean that volcanic action causes most of the earthquakes. Active volcanoes and earthquakes are so commonly associated in the same belts because those belts no doubt represent portions of the crust which are now most actively yielding to the forces directly resulting from the shrinkage of the earth. Within the volcanic belts many earthquakes take place unaccompanied by any volcanic action, and many others take place far from volcanoes. Some earthquakes have been caused by the impact of great landslides or avalanches, or by the sudden caving in of underground openings.

Brief descriptions of a few typical carefully studied earthquakes during recent years will serve to make the main features of earthquakes still clearer to the reader.

The violent Japanese earthquake of 1891 was caused by the sinking of a block of earth forty miles long from two to thirty feet below that on the other side of a fault fracture. There was also considerable horizontal shifting, and cracks developed in the adjacent region. A distinct fault scarp, fifteen to twenty feet high, developed, and in some cases extended right across cultivated fields.

The San Francisco earthquake of 1906 was produced by renewed movement along the great fault which extends lengthwise through the Coast Range Mountains for several hundred miles. It is literally correct to say that, for 250 miles along this great earth fracture, one part of the Coast Range instantaneously slipped from two to twenty-two feet past the other. More or less of the movement extended at least several thousand feet down into the earth. In this case both sides slipped and the movement was very largely horizontal rather than vertical. The land on the east side of the fault moved south and that on the west side moved north, the amount diminishing away from the fault on each side so that some miles out the actual crustal movement was only a few inches. When one thinks of the tremendous volumes of earth material involved in this shifting of the earth's crust, is it any wonder that such destructive earthquake waves were produced? Many buildings were wrecked, several hundred people were killed, the disastrous San Francisco fire resulted, water mains were broken, and fences and roads crossed by the fault were dislocated as much as fifteen to twenty feet.

During the great earthquake on the coast of Alaska in 1899 notable changes took place along the shore for some miles, one portion having suddenly risen as much as forty-seven feet, while another portion sank below sea level.

In 1886 the earthquake centering near Charleston, S. C., was preceded by rumbling and roaring noises and the slight quaking increased to violent shaking which lasted more than a minute. Eight minutes later a rather violent earthquake shock took place, followed during the next ten or twelve hours by less severe shocks. Most buildings in the city were wrecked or more or less badly damaged, and some people were killed. The shocks were so violent that the quaking was actually felt by people over an area of more than 2,000,000 square miles, the disturbance having spread at the rate of about 150 miles per minute. Near Charleston openings and fissures were formed through which sand and muddy water were ejected, but the cause of the disturbance was most likely slipping of the old very hard rocks below the loose deposits of the Coastal Plain.

From 1811 to 1813 a series of violent earthquakes developed in the general vicinity of New Madrid, Missouri. In an area of over 2,000 square miles, now called the "sunk country," many portions suddenly sank giving rise to small fault scarps or cliffs, and various lake basins were formed. Development of a fissure caused a local change in the course of the Mississippi River.

In 1897, Assam, India, was shaken by an earthquake of unusual magnitude, which lasted 2-1/2 minutes. An area of 150,000 square miles was disastrously shaken, and the shocks were distinctly felt over an area of 750,000 square miles. A number of notable fault scarps developed, the movement on one having been thirty-five feet.