Geology: The Science of the Earth's Crust

CHAPTER III

Chapter 44,886 wordsPublic domain

STREAM WORK

Most streams are incessantly at work cutting or eroding their way into the earth's crust and carrying off the products of weathering. By this means the general level of lands is gradually being reduced to nearer and nearer sea level. Base level of erosion is reached when any stream has eroded to its greatest possible depth, and a whole region is said to be base-leveled when, by the action of streams, it has been reduced to a practically flat condition. A region of this kind is known as a "peneplain."

To one who has not seriously considered the matter, the power of even moderately swift water to transport rock débris seems incredible. A well-established law of transportation by running water is that the transporting power of a current varies as the sixth power of its velocity. For example, a current which is just able to move a rock fragment of a given size will, when its velocity is merely doubled, be able to move along a piece of similar rock sixty-four times as large! That this must be the case may be readily proved as follows: A current of given velocity is just able to move a block of rock, say, of one cubic inch in the form of a cube. A cubic block sixty-four times as large has a face of sixteen square inches. By doubling the velocity of the current, therefore, twice as much water must strike each of the sixteen square inches of the face of the larger block with twice the force, thus exerting sixty-four times the power against the face of the larger block, or enough to move it along. This surprising law accounts for the fact that in certain floods, like the one which rushed over Johnstown, Pennsylvania, in 1889, locomotives, massive iron bridges, and great bowlders were swept along with great velocity. It is obvious, then, that ordinarily swift rivers in time of flood accomplish far more work of erosion (especially transportation) than during many days or even some months of low water.

Few people have the slightest idea as to the enormous amount of earth material which the rivers are carrying into the sea each year. The burden carried by the Mississippi River has been carefully studied for many years. Each year this river discharges about 400,000,000 tons of material in suspension; 120,000,000 tons in solution; and 40,000,000 tons rolled along the bottom. This all represents earth material eroded from the drainage basin of the river. It is sufficient to cover a square mile 325 feet deep, or if placed in ordinary freight cars it would require a train reaching around the earth several times to contain it. Since the drainage basin of the Mississippi covers about 1,250,000 square miles, it is, therefore, evident that this drainage area is being worn down at the average rate of about one foot in 3,840 years, and this is perhaps, a fair average for the rivers of the earth. The Ganges River, being unusually favorably situated for rapid erosion, wears down its drainage basin about one foot in 1,750 years. It has been estimated that nearly 800,000,000 tons of material are annually carried into the sea by the rivers of the United States. According to this the country, as a whole, is being cut down at the rate of about one foot in 9,000 years. In arriving at this figure it should, of course, be borne in mind that the average level of hundreds of thousands of square miles of the western United States, particularly the so-called Great Basin, is practically not being reduced at all because none of the streams there reach the sea.

Deposition of sediment is an important natural consequence of erosion. The destination of most streams is the sea, and where tides are relatively slight the sediments discharged mostly accumulate relatively near the mouths of the rivers in the form of flat, fan-shaped delta deposits. Some rivers, like the Ganges, which carry such unusual quantities of sediment, are able to construct deltas in spite of considerable tides. Deltas also form in lakes. In most cases, however, rivers enter the sea where there are considerable tides and their loads are more widely spread over the marginal sea bottom. But in many cases some of the sediment does not reach the mouth of the stream. It is, instead, deposited along its course either where the velocity is sufficiently checked, as is the case over many flood-plain areas of rivers, or where a heavily loaded, relatively swift stream has its general velocity notably diminished. An excellent example of the latter type of stream is the Platte River, which is swift and loaded with sediment in its descent from the Rocky Mountains, but, on reaching the relatively more nearly level Nebraska country, it has its current sufficiently checked to force it to deposit sediment and build up its channel along many miles of its course, and this in spite of the fact that it still maintains a considerable current. In a mountainous arid region a more or less intermittent stream at times of flood becomes heavily loaded with rock débris and rushes down the mountain side. On reaching the valley floor the velocity is greatly checked and most of the load is deposited at the base of the mountain, successive accumulations of such materials, called alluvial cones or fans, having not uncommonly built up to depths of hundreds, or even several thousand feet.

Any newly formed land surface, like a recently drained lake bed or part of the marginal sea bottom which has been raised into land, has a drainage system developed upon it. In the early or youthful stage of such a new land area lying well above sea level, under ordinary climatic conditions a few streams only form and these tend to follow the natural or initial slope of the land. These streams carve out narrow, steep-sided valleys, and all of them are actively engaged in cutting down their channels, or, in other words, none of them have reached base level, and flood plains and meandering curves are therefore lacking. During this youthful stage there are no sharp drainage divides; gorges and waterfalls are not uncommonly present; and the relief of the land in general is not rugged. A good example of youthful topography is the region around Fargo, North Dakota, which is part of the bed of a great recently drained lake. The Grand Canyon of the Yellowstone River is an excellent illustration of a youthful valley cut in a high plateau of geologically recent origin. (Plate 2.)

As time goes on, a region in youth gradually gives way to a region in maturity, during which stage the maximum number (usually a network) of streams in broader V-shaped valleys have developed; divisions of drainage are sharp; the maximum ruggedness of relief has developed; the larger streams only have cut down so near base level that winding (meandering) courses and flood plains are well developed along them; and waterfalls and gorges are rarely present. An almost perfect example of a region in maturity is that around Charleston, West Virginia.

The old-age stage develops next in the history of the region, during which only a moderate number of streams remain, most of these being at or close to base level so that sweeping curves or meanders (Plate 4) and cut-off meanders or "ox bows" and wide flood plains are characteristic and common. The relief is greatly subdued and the term "rolling country" might be applied to the moderately hilly region. Divisions of drainage are, of course, not at all sharp and the valleys are wide and shallow. Oxbow lakes are common, but gorges and waterfalls are absent. A region typical of old-age topography is that around Caldwell, Kansas.

Finally, after the remaining low hills have been cut down, the region is in the condition of a broad monotonous plain, practically devoid of relief, over which the sluggish streams pursue very winding, more or less shifting or indefinite courses. For the attainment of this final stage (called a "peneplain") in the normal cycle of erosion a proportionately very long time is necessary, because the rate of erosion becomes slower and slower as the region is being cut down. Then, too, some change of level between the land and the sea is very likely to take place before the peneplain stage is reached. It is doubtful if any extensive region was ever brought to the condition of a perfect peneplain. Some masses of more resistant or more favorably situated rocks are almost sure to maintain at least moderate heights above the general plain level. Geologically recently upraised, fairly well developed peneplains are southern New England and the great region of eastern Canada. The remarkably even sky lines of these regions mark the peneplain level before the uplift took place, and occasionally masses, called "monadnocks" from Mount Monadnock in southern New Hampshire, rise above the general level. The valleys in such an uplifted peneplain region have been carved out by streams since the uplift began. We have positive evidence that more or less well-developed peneplains of considerable extent existed in various parts of the earth at various times during the many millions of years of known earth history.

The normal cycle of erosion which, as outlined above, tends toward the peneplain condition may be interrupted at any stage by other processes. An excellent case in point is the upper Mississippi Valley, which had reached the old-age stage, even approximating a peneplain, just before the great Ice Age. Then, during the withdrawal of the vast sheet of ice from the region toward the close of the Ice Age, extensive deposits (moraines, etc.) of glacial débris were left irregularly strewn over the country, giving rise to many low hills, lake basins, and altered drainage lines, in some cases with resultant gorge development. Some distinct features of a youthful topography are, therefore, plastered over what was otherwise a region well along in old age. The general district around the Dells of Wisconsin River well illustrates this principle.

Changes in level between land and sea which take place during the erosion of a region may also disturb the normal cycle of erosion. For example, a region in old age may be considerably upraised so that the streams have their velocities notably increased. Such a region is said to be "rejuvenated" and the streams, which are revived in activity, begin to cut youthful valleys in the bottom of the old ones and, after a time, the general surface of the region is subjected to vigorous erosion and a new cycle of erosion will be carried out unless interfered with in some way, as by relative change of level between the land and the sea. In this connection the history of the topography in the general vicinity of Harrisburg, Pennsylvania, may be of interest by way of illustration of the principle just described. The long, narrow, parallel Appalachian Mountain ridges there rise to about the same level, causing a remarkably even sky line as viewed from one of the summits. This even sky line marks approximately the surface of what was a peneplain late in the Mesozoic era. Early in the succeeding Cenozoic era, the broad peneplain was notably upraised to nearly the present altitudes of the ridge tops. The revived Susquehanna River left the old course which it had on the peneplain surface, and began to carve out its present valley, while tributaries (subsequent streams) to it developed along belts of weaker rock and thus they formed the present parallel valleys separated by belts of more resistant rocks which stand out as ridges. In this way, the mature stage of topography was reached. Very recently, geologically, the region has been rejuvenated enough to cause the larger streams to appreciably sink their channels below the general valley floors. The reader will find a general discussion of movements of the earth's crust in a succeeding chapter.

If, for example, a region along the seaboard has reached the mature stage of erosion, and the land notably subsides relative to sea level, the tidewater will enter the lower valleys to form estuaries and the valleys are said to be "drowned." The large streams, or at least their lower courses, are thus obliterated and also the general erosion of the region is distinctly diminished. The recently sunken coast of Maine well illustrates the idea of "drowned valleys." The drowned valley of the lower Hudson River is another fine example.

What is termed stream "piracy" is of special interest in connection with stream work. By this is meant the stealing of one stream or part of a stream by another. We shall here explain only one of the various ways by which stream capture may be effected. One of two fairly active streams, flowing roughly parallel to each other, is more favorably situated and has cut its channel deeper. Its tributaries are, therefore, more favorable to extension of headwaters and, in time, one of its tributaries eats back far enough to tap a branch of the less favorably situated stream so that the waters of this branch are diverted into the more favorably situated stream. The Shenandoah River of Virginia has been such a pirate. This river developed as a tributary of the Potomac. By headward extension toward the south, the Shenandoah finally tapped and diverted the upper waters of the smaller, less favorably situated Beaverdam Creek. The notch or so-called "wind gap" through which the upper waters of Beaverdam Creek formerly flowed across the Blue Ridge is still plainly visible. Such abandoned water gaps, known as "wind gaps," are common in the central Appalachian Mountain region.

A remarkable type of river is one which has been able to maintain its course through a barrier, even a mountain range, which has been built across it. Thus, the Columbia River, after flowing many miles across the great lava plateau, has maintained its course right across the growing Cascade Range by cutting a deep canyon while the mountain uplift has been in progress. In a similar manner the Ogden River of Utah has kept its westward course by cutting a deep canyon into the Wasatch Range which has geologically recently, though slowly, risen across its path. In no other way can we possibly explain the fact that such a river, rising on one side of a high mountain range, cuts right across it.

A feature of minor though considerable popular interest is the development of "potholes" by stream action. Where eddies occur, in rather active streams, rock fragments of varying sizes may be whirled around in such manner as to corrode or grind the bedrock, resulting in the development of cylinder-shaped "potholes." Such holes vary in diameter up to fifty feet or more in very exceptional cases. In the production of large "potholes" many rock fragments are worn away and new ones supplied to continue the work. Locally, some stream beds are honeycombed with "potholes."

Strikingly narrow and deep valleys, called gorges and canyons, are rather exceptional features of stream action. Most wonderful of all features of this kind is the Grand Canyon of the Colorado River in Arizona. In fact, this canyon takes high rank among the most remarkable works of nature. The canyon is over 200 miles long, from 4,000 to 6,000 feet deep, and from 8 to 15 miles wide. Contrary to popular opinion, this mighty canyon is not a result of some violent process, such as volcanic action, or the sudden sinking of part of the earth's crust. Nor is it the result of the scouring action of a great glacier. It is simply a result of the operation of the ordinary processes of erosion where the conditions have been exceptionally favorable. Some of the favorable conditions have been, and are, a large volume of very swift water (Colorado River) continually charged with an abundance of rock fragments for the work of corrasion, and a great thickness of rock which the river must cut through before reaching base-level. Aridity of climate also tends to preserve the canyon form. The whole work has been accomplished in very late geological time, and the tremendous volume of rock which has been weathered and eroded to produce the canyon has all been carried away by the Colorado River and accumulated in the great delta deposit near where the river empties into the Gulf of California. Even now the canyon is growing deeper and wider because the very active Colorado River is still from 2,000 to 3,000 feet above sea level. Standing on the southern rim near Grand Canyon station at an altitude of nearly 7,000 feet, and looking down into the canyon, one beholds a vast maze of side canyons, high, vertical rock walls which follow very sinuous courses, giving rise to a steplike topography, and countless rock pinnacles, towers, and mesas often of mountain-like proportions. The side canyons are the result of erosion by tributaries to the main river which have gradually developed and worked headward as the main river has cut down. The mountain-like sculptured forms which rise out of the canyon are erosion remnants, or, in other words, masses of rock which were more favorably situated against erosion by either the main river or any of its tributaries. All of the rocks of the broader, main portions of the canyon are strata of Paleozoic age, arranged as a vast pile of almost horizontal layers, including sandstone, limestone, and shale. Some of these layers, being distinctly more resistant than others, stand out in the canyon wall in the form of conspicuous cliffs, in some cases hundreds of feet high. The very striking color bands (mostly light gray, red, and greenish gray), which may be traced in and out along the canyon sides, represent the outcropping edges of variously colored rock layers. Far down in the canyon lies the steep-sided, V-shaped inner gorge, or canyon which is fully 1,000 feet deep. The rocks are there not ordinary strata, but rather metamorphic and igneous rocks, mostly dark gray, not in layers, and about uniformly resistant to erosion. There is reason to believe that this inner gorge has developed mainly since a distinct renewed uplift (rejuvenation) of the Colorado Plateau after the river began its canyon cutting. The narrow, steep-sided inner gorge may thus be readily accounted for and the general lack of steplike forms on its sides is due to essential uniformity of the rock material as regards resistance to erosion.

The wonderful King's River Canyon of the southern Sierras in California is remarkable for its combined narrowness and depth. It is a steep V-shaped canyon whose maximum depth is 6,900 feet, carved out in mostly solid granite by the action of weathering and running water. Some idea of the vast antiquity of the earth may be gleaned from the fact that this tremendously deep canyon has been produced by erosion in one of the most resistant of all known rocks in very late geologic time! Conditions favorable for cutting this canyon have been volume and swiftness of water and a liberal supply of grinding tools.

Among the many other great canyons of the western United States brief mention may be made of the Grand Canyon of the Yellowstone River in the National Park. The plateau into which the river has cut its steep-sided, narrow, V-shaped canyon, with a maximum depth of 1,200 feet, has been geologically recently built up by outpourings of vast sheets of lava. The large volume of very swift water, aided by decomposition and weakening of the ordinarily very hard rock by the action of the hot springs, has been able to carve out this deep canyon practically within the last period of earth history. The deepening process is still vigorously in progress. The wonderful coloring of the rock, mostly in tones of yellow and brown, is due to the hydrated iron oxides developed during the decay of the iron-bearing minerals of the lava, the chemical action having been greatly aided by the action of the hot waters. (See Plate 2.)

In regard to its origin, the marvelous Yosemite Valley, or canyon, falls in a somewhat different category, and it is discussed beyond in connection with the work of ice. Suffice it to say here that running water has been a very important factor in its origin.

In New York and New England there are many gorges which have developed by the action of running water since the Great Ice Age. Famous among these are Ausable Chasm and Watkins Glen of New York, and the Flume in the White Mountains of New Hampshire.

Before leaving our discussion of the work of running water, we should briefly consider waterfalls. True waterfalls originate in a number of ways. Most common of all is what may be termed the "Niagara type" of waterfall. Niagara Falls merit more than passing mention not only because of their scenic grandeur, but also because of the unusual number of geologic principles which their origin and history so clearly illustrate. Niagara Falls are divided into two main portions, the Canadian, or so-called "Horseshoe Fall," and the "American Fall," separated by a large island. The crest of the American Fall is about 1,000 feet long and nearly straight, while the crest of the Canadian Fall is notably curved inward upstream, and it is about 3,000 feet long. The height of the Falls is 167 feet. Downstream from the Falls there is a very steep-sided gorge about 200 feet deep and seven miles long. The exposed rocks of the region are nearly horizontal layers of limestone underlain with shales. Relatively more resistant limestone forms the crest of the falls, and directly underneath are the much weaker shales. Herein lies the principle of this type of waterfall because, due to weathering and the swirling action of the water, the weaker underlying rocks erode faster, thus causing the overlying rock to overhang so that from time to time blocks of it already more or less separated by cracks (joints), fall down and are mostly carried away in the swift current. Thus the waterfall maintains itself while it steadily retreats upstream. Careful estimates based upon observations made between 1827 and 1905 show that the Canadian Fall retreated at the rate of from three to five feet per year, while the American Fall retreated during the same time at the rate of only several inches per year. It has been well established that Niagara Falls came into existence soon after the ice of the great Ice Age had retreated from the district. The falls started by plunging over a limestone escarpment, situated at what is now the mouth of the gorge seven miles downstream from the present falls. If we consider the rate of recession of the falls to have been always five feet per year, the length of time required to cut the gorge would be something over 7,000 years. But the problem is not so simple, because we know that, at the time of, or shortly after, the beginning of the falls, the upper Great Lakes drained farther north and not over the falls; and that this continued for a considerable, though unknown, length of time. During this interval the volume of water in Niagara River was notably diminished, and hence the recession of the falls must have been slower. On the other hand, judging by the width of the gorge, the length of the crest of the falls has generally been considerably less than at present, which in turn means greater concentration of water over the crest and more rapid wear. Various factors considered, the best estimates for the age of the falls vary from 7,000 to 50,000 years, an average being about 25,000 years. Although this figure is not precise, it is, nevertheless, of considerable geologic interest because it shows that the age of Niagara Falls (and gorge) is to be reckoned in some tens of thousands of years, rather than hundreds of thousands or millions of years. Although waterfalls of the Niagara type are the most common of all, it is by no means necessary that the particular rocks should be limestone and shale.

Another common kind of waterfall may be termed the "Yosemite type," so named from the high falls in the Yosemite Valley of California. At the great falls of the Yosemite, the rock is a homogeneous granite and the undermining process does not operate. Yosemite Creek first plunges vertically over a granite cliff for 1,430 feet to form the Upper Falls, which must rank among the very highest of all true water falls. The water then descends about 800 feet by cascading through a narrow gorge, after which it makes a final vertical plunge of over 300 feet. A brief history of the falls is about as follows. A great steep-sided V-shaped canyon several thousand feet deep had been carved out by the action of the Merced River which now flows through the valley. Then, during the Ice Age, a mighty glacier plowed through the canyon, filling it to overflowing. The granite of this district having been unusually highly fractured by great vertical joint cracks was relatively easy prey for ice erosion. Due to its great weight, the erosive power of the ice was most potent toward the bottom, successive joint blocks were removed, and the valley was thus widened and the sides steepened or even commonly made practically vertical. (See Plate 6.) After the melting of the ice, certain of the streams, like Yosemite Creek and Bridal Veil Creek, were forced to enter the valley by plunging over great perpendicular, granite cliffs which are in reality joint faces. This type of waterfall does not retreat, but it constantly diminishes in height by cutting into the crest. A number of other high falls of this kind occur in the Yosemite region, and also in other mountain valleys which formerly contained glaciers, as in the Canadian Rockies, the Alps, and Norway, the rocks in these regions being of various kinds.

In the case of the "Yellowstone type" of waterfall a different principle is involved, namely, a distinctly harder or more resistant mass of rock which extends vertically across the channel of the stream. At the Great Falls of the Yellowstone a mass of relatively fresh, hard lava lies athwart the course of the river, while just below it the lava has been much weakened by decomposition. The harder rock therefore acts as a barrier, while, in the course of time, the weak rock on the downstream side has been worn away until a waterfall 308 feet high has developed. This waterfall does not retreat very appreciably, but it is probably increasing in height, due both to the scouring action of the water at the base of the fall and the unusual clearness of the river water here, thus causing little wear at the crest. It should be noted, in this connection, that the channel just on the upstream side of the barrier cannot be cut down faster than the top of the barrier itself.

The famous Victoria Falls of the Zambezi River in South Africa, represents a relatively uncommon type of waterfall. Considering height of the fall, length of crest, and volume of water, this is perhaps the greatest waterfall in the world. The Zambezi River, a mile wide, plunges over 400 feet vertically into a chasm only a few hundred feet wide and at right angles to the main course of the stream. The general country rock is hard lava, but locally a narrow belt of the rock has been highly fractured vertically, due to earth movements or faulting (see explanation beyond) and therefore weakened and more subject to weathering than the general body of the lava rock. This belt of weakened rock has been easy prey for erosion by the Zambezi River and the chasm has there developed. In fact the chasm is still being increased in depth. Leaving the chasm toward one end, the river flows through a narrow zig-zag gorge whose position has been determined by big joint cracks. The mile-wide crest of the falls is interrupted by a good many ledges and even small islands. The thundering noise of this great waterfall is most impressive, but a good complete view is impossible because most of the chasm is constantly filled with dense spray.

Still another type of waterfall develops by the removal of joint blocks by the action of running water. Falls of this type are fairly common though they seldom attain really great heights. Where the rock in the bed of a stream is traversed by well-developed vertical joint cracks, slabs of rock cleaved by the joints may fall away due to weathering or they may be pushed away by pressure of the water. Such a fall retreats upstream by removal of joint blocks even in comparatively homogeneous rocks. Taughannock Falls, 215 feet high, in southern central New York, has developed by this manner in a shaly sandstone. The several falls (one 50 feet high) in the famous gorge at Trenton Falls, in central New York, have developed in this way in limestone.