Fluvial Processes in Geomorphology

DOVER SCIENCE BOOKS

THE DEVELOPMENT OF MATHEMATICS, E. T. BELL. (0-486-27239-7)

507 MECHANICAL MOVEMENTS: MECHANISMS AND DEVICES, HENRY T. BROWN. (0-486-44360-4)

A REFRESHER COURSE IN MATHEMATICS, F. J. CAMM. (0-486-43225-4)

FLAWS AND FALLACIES IN STATISTICAL THINKING, STEPHEN K. CAMPBELL. (0486-43598-9)

HARMONIC PROPORTION AND FORM IN NATURE, ART AND ARCHITECTURE, SAMUEL COLMAN. (0-486-42873-7)

THE CURVES OF LIFE, THEODORE A. COOK. (0-486-23701-X)

A SHORT HISTORY OF TECHNOLOGY: FROM THE EARLIEST TIMES TO A.D. 1900, T. K. DERRY AND TREVOR I. WILLIAMS. (0-486-27472-1)

MATHOGRAPHICS, ROBERT DIXON. (0-486-26639-7)

EINSTEIN’S ESSAYS IN SCIENCE, ALBERT EINSTEIN. (0-486-47011-3)

EXPERIMENTAL RESEARCHES IN ELECTRICITY, MICHAEL FARADAY. (0-486-43505-9)

THE GREAT PHYSICISTS FROM GALILEO TO EINSTEIN, GEORGE GAMOW. (0-486-25767-3)

THIRTY YEARS THAT SHOOK PHYSICS: THE STORY OF QUANTUM THEORY, GEORGE GAMOW. (0-486-24895-X)

ONE TWO THREE . . . INFINITY: FACTS AND SPECULATIONS OF SCIENCE, GEORGE GAMOW. (0-486-25664-2)

GRAVITY, GEORGE GAMOW. (0-486-42563-0)

FADS AND FALLACIES IN THE NAME OF SCIENCE, MARTIN GARDNER. (0-486-20394-8)

RELATIVITY SIMPLY EXPLAINED, MARTIN GARDNER. (0-486-29315-7)

THE GEOMETRY OF ART AND LIFE, MATILA GHYKA. (0-486-23542-4)

UNDERSTANDING EINSTEIN’S THEORIES OF RELATIVITY: MAN’S NEW PERSPECTIVE ON THE COSMOS, STAN GIBILISCO. (0-486-26659-1)

FROM GALILEO TO NEWTON, A. RUPERT HALL. (0-486-24227-7)

MECHANICAL APPLIANCES, MECHANICAL MOVEMENTS AND NOVELTIES OF CONSTRUCTION, GARDNER D. HISCOX. (0-486-46886-0)

1800 MECHANICAL MOVEMENTS, DEVICES AND APPLIANCES, GARDNER D. HISCOX. (0-486-45743-5)

MAKERS OF MATHEMATICS, STUART HOLLINGDALE. (0-486-45007-4)

THE DIVINE PROPORTION, H. E. HUNTLEY. (0-486-22254-3)

ENGINEERING AND TECHNOLOGY, 1650-1750: ILLUSTRATIONS AND TEXTS FROM ORIGINAL SOURCES, MARTIN JENSEN. (0-486-42232-1)

SHORT-CUT MATH, GERARD W. KELLY. (0-486-24611-6)

MATHEMATICS FOR THE NONMATHEMATICIAN, MORRIS KLINE. (0-486-24823-2)

THE FOURTH DIMENSION SIMPLY EXPLAINED, HENRY P. MANNING. (0-486-43889-9)

VIOLENT PHENOMENA IN THE UNIVERSE, JAYANT V. NARLIKAR. (0-486-45797-4)

BASIC MACHINES AND HOW THEY WORK, NAVAL EDUCATION. (0-486-21709-4)

EXCURSIONS IN GEOMETRY, C. STANLEY OGILVY. (0-486-26530-7)

MUSIC, PHYSICS AND ENGINEERING, HARRY F. OLSON. (0-486-21769-8)

COMPUTERS, PATTERN, CHAOS AND BEAUTY, CLIFFORD A. PICKOVER. (0-486-41709-3)

AN INTRODUCTION TO INFORMATION THEORY, JOHN R. PIERCE. (0-486-24061-4)

INTRODUCTION TO MATHEMATICAL PHILOSOPHY, BERTRAND RUSSELL. (0—486—27724—0)

HELLENISTIC SCIENCE AND CULTURE IN THE LAST THREE CENTURIES B.C., GEORGE SARTON. (0—486—27740—2)

MATHEMATICIAN’S DELIGHT, W. W. SAWYER. (0-486-46240-4)

EINSTEIN’S LEGACY: THE UNITY OF SPACE AND TIME, JULIAN SCHWINGER. (0-486-41974-6) THE UNITY OF THE UNIVERSE, D. W. SCIAMA. (0-486-47205-1)

FROM FALLING BODIES TO RADIO WAVES: CLASSICAL PHYSICISTS AND THEIR DISCOVERIES, EMILIO SEGRÈ. (0-486-45808-3)

FROM X-RAYS TO QUARKS: MODERN PHYSICISTS AND THEIR DISCOVERIES, EMILIO SEGRÈ. (0-486-457834)

GREAT EXPERIMENTS IN PHYSICS: FIRSTHAND ACCOUNTS FROM GALILEO TO EINSTEIN, MORRIS H. SHAMOS. (0—486—25346—5)

SATAN, CANTOR AND INFINITY: MIND-BOGGLING PUZZLES, RAYMOND M. SMULLYAN. (0-48647036-9)

THE LADY OR THE TIGER?: AND OTHER LOGIC PUZZLES, RAYMOND M. SMULLYAN. (0-486-47027-X)

HOW TO CALCULATE QUICKLY: FULL COURSE IN SPEED ARITHMETIC, HENRY STICKER. (0—486—20295—X)

SPEED MATHEMATICS SIMPLIFIED, EDWARD STODDARD. (0-486-27887-5)

A SCIENTIST AT THE SEASHORE, JAMES S. TREFIL. (0-486-44564-X)

INTRODUCTION TO MATHEMATICAL THINKING: THE FORMATION OF CONCEPTS IN MODERN MATHEMATICS, FRIEDRICH WAISMANN. (0—486—42804—4)

SUNDIALS: THEIR THEORY AND CONSTRUCTION, ALBERT E. WAUGH. (0—486—22947—5)

HOW TO SOLVE MATHEMATICAL PROBLEMS, WAYNE A. WICKELGREN. (0—486—28433—6)

THE TRIUMPH OF THE EMBRYO, LEWIS WOLPERT. (0-486-46929-8)

SEE EVERY DOVER BOOK IN PRINT AT

WWW.DOVERPUBLICATIONS.COM

Copyright

Copyright © 1964, 1992 by Luna B. Leopold and M. Gordon Wolman.

All rights reserved.

Bibliographical Note

This Dover edition, first published in 1995, is an unabridged and unaltered republication of the work first published by W. H. Freeman and Company, San Francisco, in 1964.

Library of Congress Cataloging-in- Publication Data

Leopold, Luna Bergere, 1915—

Fluvial processes in geomorphology / Luna B. Leopold, M. Gordon Wolman, John P. Miller.

p. cm.

An unabridged and unaltered republication of the work first published by W.H. Freeman and Company, San Francisco, in 1964—T.p. verso.

Includes index.

9780486139739

1. Physical geography. 2. Rivers. I. Wolman, M. Gordon (Markley

Gordon), 1924— . II. Miller, John P. (John Preston), 1923—1961.

III. Title.

GB55.L4 1995

551.3’5—dc20

95-20080

CIP

Manufactured in the United States by Courier Corporation

68588807

www.doverpublications.com

To Carolyn, Elaine, and Laura

Preface

THIS BOOK deals primarily with landform development under processes associated with running water. The bias of the book is dictated by our experience and interest and also by our belief that there is a great need at the present time for a review of geomorphic processes.

We have emphasized those things which are best known to us (and about which we feel most is known). Many subjects we have included are by no means treated completely, for they are discussed only from one viewpoint. Others that are treated lightly, such as the evolution of slopes, are ones for which little comprehensive quantitative data are available.

Rather than present a mere rehash of published material which we could not adequately discuss, we decided to omit entirely subjects we have not studied ourselves in the field or laboratory. Some summary monographs are available for wind, shore, and glacial processes, and we have not attempted to cover those subjects here. Combining process and stratigraphy for wind, shore, and glacial morphology would only have enlarged this book to unmanageable bulk; and, as Penck argued many years ago, a case can be made for the thesis that river and hillslope processes provide the central theme of geomorphology.

Our emphasis on process is not intended to minimize the importance of the historical aspects of geomorphology. Unfortunately, because of the limited understanding of geomorphic processes and their associated landforms, we ourselves are unable at present to make a truly satisfactory translation from the dynamics of process to historical interpretation. Better future understanding of the relation of process and form will hopefully contribute to, not detract from, historical geomorphology.

Despite its omissions, we hope that our treatment of geomorphology in this book will provide a logical framework for the subject as a whole, within which students and other readers can integrate material appropriate to their own interests or local physiographic environments.

We have sorely missed our compadre and co-author, John Preston Miller, during the last two years when this book was actively being constructed. Those portions which he prepared were perforce revised during that time. We hope that his principal ideas have been retained and that we have not allowed either divergent viewpoints or errors to creep into his work. Though we can put a book together without him, we can not view the high mountains nor can we pitch a camp in just the same spirit as when he was along.

We are indebted to colleagues and friends too numerous to name who helped in a variety of ways—in technical review of portions of the manuscript, in furnishing data and information, in preparation of copy and illustrations, and in our field work. But some should be noted specifically.

First, Mae E. Thiesen, although this is not the first manuscript which she has prepared for us. It is a pleasure to be able here to acknowledge her thoughtful and untiring help in all aspects of manuscript preparation, without which this book would not have been brought to completion.

We are particularly indebted to A. O. Woodford and James Gilluly for their overall review, and to Ralph A. Bagnold, Ivan K. Barnes, John T. Hack, Meyer Rubin, and Estella B. Leopold for their suggestions on portions of the work.

To the other river boys, William W. Emmett and Robert M. Myrick, our thanks not only for help in the field and in preparing the manuscript, but for their company at many delightful campfires beside many distant rivers.

Particular thanks we owe to William H. Freeman, who encouraged us when we most needed such encouragement.

And finally, we wish to mention two men who long have been close friends, admired colleagues, and friendly advisors, Walter B. Langbein and Thomas Maddock, Jr., whose influence on this work has been perhaps deeper and more significant than that of any others.

LUNA B. LEOPOLD

M. GORDON WOLMAN

May 1963

Table of Contents

DOVER SCIENCE BOOKS

Title Page

Copyright Page

Dedication

Preface

Part I - THE EVOLVING LANDSCAPE

Chapter 1 - The Changing Scene

Chapter 2 - Geomorphology and the Field Problem

Part II - PROCESS AND FORM

Chapter 3 - Climate and Denudational Processes

Chapter 4 - Weathering

Chapter 5 - The Drainage Basin as a Geomorphic Unit

Chapter 6 - Water and Sediment in Channels

Chapter 7 - Channel Form and Process

Chapter 8 - Hillslope Characteristics and Processes

Part III - THE EFFECTS OF TIME

Chapter 9 - Geochronology

Chapter 10 - Drainage Pattern Evolution

Chapter 11 - Channel Changes with Time

Chapter 12 - Evolution of Hillslopes

Appendix A - Conversion of Units and Equivalents

Appendix B - Symbols and Nomenclature

Author Index

Subject Index

Part I

THE EVOLVING LANDSCAPE

Chapter 1

The Changing Scene

When a man makes a pilgrimage to the fields and woods of his boyhood, he does not expect to find the hills and mountains dissolved, or the valleys moved. If other men have not torn up the land to build factories and towns, he expects his children to see the hills and swales as his forefathers saw them. And he is almost right. Probably neither he nor the children will ever notice that in fifty years the surface of the ground has been lowered perhaps a fraction of an inch. Why should they? But they might not be surprised to find that the old mill pond behind the dam is now more mud than water.

Under the action of the force of gravity the land surface is sculptured by water, wind, and ice. This sculpturing produces the landforms with which geomorphology is concerned. Some of these forms owe their origins purely to denudational processes; other forms may be depositional; still others owe their existence to combinations of both processes.

A picture of the dynamics of the earth’s surface is by no means complete, however, if only gradation or leveling is considered. Clearly, if there were no counteracting forces we should expect that the land surface, given sufficient time, would be continuously reduced. Eventually, little or no relief would remain. Geologic history demonstrates, however, that the degradational forces acting on the earth’s surface are opposed by constructional forces. These internal, or endogenous, forces cause the land to rise, and as they do so it is subjected to attack by the external, or exogenous, agents. Geomorphology is primarily concerned with the exogenous processes as they mold the surface of the earth, but the internal forces cannot be disregarded when one considers fundamental concepts of the origin and development of landforms.

Ideally, the basic principles underlying the development of landforms can be considered in simple terms. A given land area is composed of a particular set of rocks, which have particular chemical and mineralogic compositions and specific physical properties. Because these rocks were formed at different temperatures and pressures within the earth, when they are exposed at the surface they are no longer in equilibrium with their environment and thus begin to decompose. Where a gradient is created by gravity, the moving water, earth, air, and ice help in the attack upon the rock and remove the products of weathering. In the process, landforms of various aspects are created. In a given environment the physical and chemical constitution of the rocks determines the way in which they will break down and, in turn, the size and quantity of debris made available to the denudational agencies.

Each denudational agent, depending upon its density, gradient, and mass at a particular place, is capable of applying a given stress on the materials available. A certain amount of work may be performed by the application of this stress, and the results of this work are the landforms that we see developed in various parts of the world. In a given climatic and vegetational environment the shape or form of the landscape will vary, depending upon the character of the rock and the type and available stress of the erosional agents. But as the land surface is reduced—so long as the products of weathering and the applied stress remain constant—the form of the land should remain the same.

If one were able to evaluate properly the properties of the rocks and the present and past capabilities of the denudational agencies, he should have no trouble in developing a rational, even mathematical, equation capable of describing the development history and equilibrium form of any landscape. William Morris Davis said essentially the same thing in 1902 when he observed that any landform is a function of the structure of the rocks (including their composition and structural attitude), the processes acting upon them, and the time over which these processes have been active. Only as we study the interrelations of these three factors are we able to discern which combinations produce which particular landforms and how they do so.

Some landforms, such as volcanoes, which may have been unaffected by denudational processes, may be considered purely constructional forms. As soon, however, as they are modified by external agencies, their form begins to represent the resultant of an interaction between the constructional forces, the rock substrate, and the applied stress.

The application of such an ideal concept to any actual landform at the present time is fraught with problems. The natural world is highly variable and the mechanics of uplift, weathering, and erosion are for the most part poorly understood. As will be seen, climate itself is a complex factor, and in most regions of the world inorganic processes are inseparable from the complex organic processes carried on by plants and animals. Although it is frequently convenient and helpful to construct a simplified synthetic picture of the natural environment, we should not lose sight of the fact that a given landscape must be the result of a complex set of factors which encompass the behavior of materials and processes over varying periods of time.

It is important to note that whether one refers to the effect on landforms of different rock types, or to the effect of different rates of uplift, such differences or changes must manifest themselves in the environment of the landform in simple physical terms. A normal fault whose strike is perpendicular to the direction of flow of a river, with downthrown block in the downstream direction, constitutes to the river a merely local increase in gradient. A similar increase in gradient might be effected by local changes in lithology, an abrupt shortening in channel length, or by an abrupt change in discharge downstream. The same physical principles determine the river’s subsequent response in each case. The permanence or impermanence of the change, as well as its possible propagation either upstream or down, will depend upon the type and amount of material available and the distribution and quantity of flow. Any true principle enunciated to explain one of the cases must be applicable to the others as well.

Thus, although the application of the principle to any one example may be fraught with difficulty, an understanding of the principle at least reduces the burden of innumerable unique cases. Geomorphologists have always sought such unifying concepts, and for a proper view of the field as a whole one must turn initially to the classical concepts of landform evolution.

The influence of William Morris Davis on geomorphology was without doubt greater and longer-lasting than that of any other individual. His major contribution was a genetic system of landform description. Beginning in 1899, Davis developed the concept that during erosion of a highland the landscape evolves systematically through distinctive stages, to which he gave the names, youth, maturity, and old age. This entire sequence of stages he called an erosion cycle (or geomorphic cycle), and the end product was supposed to be a surface of low relief, or peneplain. He elaborated the effects of interruptions in the cycle and argued that the principal factors controlling the character of landforms are geologic structure, geomorphic processes, and the stage of development. Davis’ genetic concept of landform development was a brilliant synthesis, which grew directly out of the work by Powell, Gilbert, and Dutton and also from the controversial ideas on organic evolution which were prevalent at the time.

The concept of the erosion cycle was never accepted in Europe to the same degree as in North America. The most serious challenge came during the 1920’s from Walther Penck, who attempted to show a direct causal relation between tectonics and the properties of landforms. Many of his conclusions about the trends and ultimate results of tectonics and erosion processes differed only slightly from those of Davis. Penck, however, emphasized slope development, and his theory of slope development is a major contribution that is still being tested and debated.

The principal alternative to the Davisian conception differs mainly in the view of the effect of time, the third of the three fundamental elements, on landforms. Restating and extending the work of Gilbert, Hack (1960) emphasizes the concept of a dynamic equilibrium in the landscape which is quickly established and which responds to changes that occur during the passage of time. This view postulates that there is at all times an approximate balance between work done and imposed load and that as the landscape is lowered by erosion and solution, or is uplifted, or as processes alter with changing climate, adjustments occur that maintain this approximate balance.

More will be said about these different views in subsequent chapters, as various aspects of the landscape are considered in greater detail. Paralleling developments in other phases of geology, the past decade has witnessed a remarkable increase in the application of analytical and experimental techniques to geomorphic problems. These investigations have taken two principal directions: (1) efforts to describe landforms more precisely through the use of statistics and other analytical techniques, (2) application of physical and chemical principles to field and laboratory studies of geomorphic processes. Although a few geologists—G. K. Gilbert, and later W. W. Rubey—helped to pave the way for this current trend, developments in other fields of science, especially in engineering and physics, were more directly responsible for it. One outstanding example is the field and experimental work on sand transport by R. A. Bagnold during the 1930’s. Another is the contribution of fundamental ideas on the development of stream networks by R. E. Horton. Recently many developments in hydraulics and in the application of soil mechanics have attracted the attention of geomorphologists. At present there is greatly increased interest in the use of more precise tools for studying landforms. The pace of research seems to be quickening and there is reason to hope that a new era of discovery is under way.

Geomorphology in North America has gone through a phase during which extensive description of the landscape in terms of the erosion cycle has been carried out. It was apparently believed that the processes were known or could be inferred, and that form could be assessed by eye.

Similarly, one current earth-history view of geomorphology assumes that enough is now known to interpret landforms and deposits in terms of processes that operated in times past. In the most qualitative way this is probably true. However, we believe that the genetic system breaks down when it is subjected to close scrutiny involving quantitative data. At present deductions are subject to considerable doubt, for the detailed properties of landform have not been studied carefully enough and the fundamental aspects of most geomorphic processes are still poorly understood. So long as this is true, the interpretation of geomorphic history rests on an exceedingly unstable base.

Accordingly, we plan to concentrate on geomorphic processes. The emphasis is primarily upon river and slope processes; river processes will receive greatest attention, since the greatest volume of information available is on rivers. Our objective is to synthesize the material on these subjects in an attempt to assess the current status of knowledge and at the same time to draw attention to its shortcomings.

Process implies mechanics—that is, the explanation of the inner workings of a process through the application of physical and chemical principles. We realize that some readers may be more interested in descriptions of landforms than in the detailed analysis of the processes that formed them. So far as possible, we attempt to relate the processes discussed to specific types of landforms. Unfortunately, the gap between our understanding of specific processes in microcosm and the explanation of major large-scale landforms is still wide. It is interesting to note that geomorphologists seem to have a better understanding of depositional than of erosional forms. This may be because the formation of depositional features such as sand dunes, deltas, and flood plains is more easily seen in the field, or because many erosional features retain less clear evidence of their mode of formation.

Detailed understanding of geomorphic processes is not a substitute for the application of basic geologic and stratigraphic principles. Rather, such understanding should help to narrow the range of possible hypotheses applicable to the explanation of different geomorphic forms and surficial earth processes and deposits.

Our approach involves some use of mathematics. We are aware that the feelings of professional geomorphologists about numbers, graphs, and formulas range from acceptance and enthusiasm to bewilderment and forthright hostility. We have not gone out of our way to be mathematical, but wherever we felt that mathematics contributed either clarity or brevity to the discussion, it has been used. Some fundamental principles of mechanics and statistics are introduced in the text where they are appropriate and necessary to an understanding of the subject at hand. Because fundamental principles of geomorphology are drawn from both mechanics and geology, some readers—depending on their backgrounds—will find specific explanations oversimplified to suit their taste, while others will find the same material wanting in simplicity. Although we have attempted to achieve balance in this regard, the wide spectrum of readers’ interests and background in the subject suggests. that a perfectly happy medium is not likely to be attained at this time.

With those readers who have a conditioned reflex against quantitative geomorphology we agree that numerical descriptions can be used to give misleading and even erroneous impressions of erudition. However, the fact remains that one’s senses, especially sight, when coupled with a conscious or unconscious bias, sometimes play strange tricks. Thus, a property which seems perfectly apparent, or an obvious relation of cause and effect, may upon careful measurement and analysis prove to be exactly the reverse of the apparent or the obvious. Some examples will be cited in the text. From a scientific standpoint, most students agree that numerical data are superior to subjective adjectives—such as big, little, high, low, steep, and gentle—in objective analyses and comparisons.

We recognize that the decision to concern ourselves primarily with the dynamics of processes has some serious pitfalls. The most critical is the fact that field investigations of modern process cannot be segregated completely from historical aspects of landform development. The same statement applies to geologic structure. Each element of the landscape has evolved through a long period to its present configuration, and this heritage doubtless influences the processes now acting upon it. Sequential observations, comparative studies with statistical controls, and perhaps scale models do, however, help to mitigate these problems.

Disclaimers to the contrary, a glance at the chapter headings will show the reader that the book as a whole is arranged according to classical geomorphic principles. Chapters 3 through 7 deal essentially with process, structure, and morphology. The evolutionary or developmental aspect of landforms is treated in Chapters 8 through 11, after the introduction of the concept of time and geochronology in Chapter 8. We hope that this separation will make clearer both the extent and limits of our understanding of surficial processes and landforms.

REFERENCES

Bagnold, R. A., 1941 (reprinted 1954), The physics of blown sand and desert dunes: Methuen Press, London, 256 pp.

Davis, W. M., 1909, Geographical essays: Ginn, New York, 777 pp.

Gilbert, G. K., 1877, Report on the geology of the Henry Mountains: U. S. Geol. Survey, Rocky Mtn. Region, Report, 160 pp.

Hack, J. T., 1960, Interpretation of erosional topography in humid-temperate regions: Am. Journ. Sci., v. 258A, pp. 80-97.

Horton, R. E., 1945, Erosional development of streams and their drainage basins: hydrophysical approach to quantitative morphology: Geol. Soc. Am. Bull., v. 56, no. 3, pp. 275-370.

Penck, W., 1922, Morphological analysis of land forms, ed. H. Czech and K. C. Boswell, 1953, Macmillan, London, 429 pp.

Powell, J. W., 1875, Exploration of the Colorado River of the West and its tributaries: Smithsonian Institution, Washington, D. C., 285 pp.

Rubey, W. W., 1933, Equilibrium conditions in debris-laden streams: 14th Annual Meeting, Am. Geophys. Union Trans., pp. 479-505.

Chapter 2

Geomorphology and the Field Problem

Introduction

As in many of the natural sciences, it is difficult to assess the problems without an initial appreciation of the available field evidence. What does one see in nature? What requires explanation? Even with a good grasp of the tools of mathematics, chemistry, physics, or botany, it is not easy to frame fundamental problems in an understandable context unless one begins with a feeling for relations as observed in the natural setting. So we begin here, not with the tools nor even with the processes, but rather with some field observations.

A Mountain Block in a Semiarid Climate

On Highway 66 one may drive eastward in nearly a straight line from Albuquerque, New Mexico, across the Rio Grande Valley, into Tijeras Canyon of the Sandia Range, and up over the mountain. From the bridge over the Rio Grande a panoramic view reveals a flat trough sloping upward to the Sandia Mountains in the east. In the eight miles between the river and the mountain front is an extensive, treeless, sloping plain nearly smooth in appearance, which abuts sharply against the steep face of the mountain block, as seen in the photograph of Fig. 2-1. On the other side of the river, facing west, buff-colored treeless hills seem to rise in a series of stairsteps, toward a horizon not far distant.

In such an environment and in an arid climate, the landscape is open to view. Geological relations are but little obscured by vegetation, and the arid zones are areas of tension where differential climatic and geologic effects are likely to be prominent rather than subdued.

Figure 2-1.

West front of the Sandia Mountains, New Mexico, looking north across the broad, slightly dissected pediment separating the mountains from the Rio Grande, which is off the picture to the left. [Photo courtesy John Whiteside.]

On closer scrutiny the mountains appear to have a steep, nearly vertical face on the river side, but there is a gentle slope backward toward the east. They resemble somewhat an ordinary brick, half-buried in the dirt but tipped slightly from the horizontal (Fig. 2-2 ) . The gently sloping principal face of the half-buried brick is covered with green—a forest consisting principally of pine, but near the top or northeast corner composed in part of aspen and spruce.

The long and narrow face of the tilted brick is steep, but not vertical, and from our vantage point the corresponding mountain face appears to be quite devoid of trees.

This general view alone is enough to pose several questions. Why does the steep face join so abruptly with a broad sloping plain? Why is the upper part of the steep mountain front devoid of trees, whereas at the same elevation the gentle back-slope is covered by forest?

Figure 2-2.

An ordinary brick, half buried in the dirt but tipped slightly from the horizontal, has the general appearance of the fault block of the Sandia Mountains, New Mexico.

A closer look at the river channel and the adjoining valley floor raises additional questions. The channel is wide, sandy, and shallow. Water is confined to a mere trickle of reddish-brown fluid resembling a chocolate malted milk, not really thick, but giving the impression of being less fluid than water.

The banks of the river channel stand vertically, to a height of 3 to 5 feet. Whereas the channel bed is a mixture of fine sand and silt, with surface ripples and bars indicative of past flows, the banks are dark brown in color and composed principally of silt with some admixture of clay. Why this difference in material of bed and bank?

Stretching out on both sides of the nearly dry channel is a broad flat, covered with cottonwoods and geometrically partitioned into rectangular patches of irrigated farmland. From a more distant view the green of the valley floor is but a ribbon of verdure stretching carelessly down the center of the broad trough between mountains. The general picture is sketched in Fig. 2-3. Details could be supplied if one drove eastward toward and into the mountains.

Leaving the river channel, we can see the dark brown soil which supports the natural cottonwoods and is the basis for the irrigated farming. The flatness of this area makes it easy to till, but shallow depressions are white with efflorescence of salts, or what locally is called alkali. What is the origin and significance of this white powder on the surface?

Through the farmland, unlined irrigation ditches are paralleled by drainage ditches. In some parts of the valley floor slight depressions have an outline which suggests that the river channel formerly flowed there. Some of the depressions are closed lakes and appear to be loops of a former river channel.

Proceeding to the edge of the valley flat, we reach a nearly continuous escarpment over a hundred feet high (E in Fig. 2-3). The face of the escarpment exposes sand containing lenses of rounded gravel, and the same materials are repeated throughout the full height of the cliff except for the very top layer. There a white zone occurs where the cobbles are covered with a cementlike deposit, to a depth of nearly 30 feet. What is the relation of this coarse material bordering the valley and the fine grained material in the vicinity of the river?

The escarpment is carved by rills and V-shaped valleys of varying size that have notched the face a short distance back. All of these rills and valleys are ephemeral; that is, they are dry except during occasional heavy rainstorms. The larger of these minor rills are seen to be continuous with flat or dish-shaped swales on the sloping plain (D in Fig. 2-3). From the top of the escarpment we look down on the green ribbon of the valley floor, and then mountainward up a nearly featureless and treeless plain rising uniformly to the base of the main mass. Even the dish-shaped swales opening into the V notches in the valley-margin escarpment seem to disappear as we look toward the mountains.

Figure 2-3. Block diagram and cross section, showing the main topographic features of the Sandia Mountains, New Mexico, and the Rio Grande.

Figure 2-4.

Topographic relation of pediment and the tributary valleys in the adjacent mountain block.

We move upslope on this plain and approach the mountain front, which at close range thrusts itself skyward out of the plain. Immediately at the base of the bold cliff are a few hillocks standing isolated from the mountain mass, like icebergs broken off the main front of a glacier and surrounded by the ocean in which they float.

The mountain front turns out to be broken by deep notches separated by spurs with knifelike crests. At its base the mountain is granite. Each boulder appears weathered and soft. Indeed, most of them could be broken into pieces by a hammer blow. Only where streams have eroded away the weathered zone is hard rock exposed. Higher on the cliff face, however, are layer upon layer of dolomite, which in many places form vertical cliffs. The rocks in the valley notches and on the streambeds are a mixture of granite and dolomite.

The face of the mountain front generally tends to be a jumble of great and small boulders, perched on a slope so steep that their stability seems doubtful. Boulder-covered spurs look down into the narrow notches of the steep valleys which, hidden from view of the broad plain, contain rivulets of clear water lined with oak and walnut trees, shaded most of the day by the high surrounding slopes. Where does this water come from, in valleys bounded on all sides by dry and rocky cliffs?

Observe the striking difference in the manner in which the sloping plain at the base of the range makes its junction with the mountain spur and with the mountain valley. By foot or by car we pass smoothly from the plain into the valley notch without any perceptible change in slope. The valley gradient smoothly and gradually increases as it penetrates deeper into the mountain mass. It is as if the mountain valleys were extensions of the plain, each extension shaped like an arrowhead bent skyward toward the tip (Fig. 2-4).

In contrast, the spurs and the intervalley scarps meet the plain at a sharp angle, nearly as a wall meets the floor. So also do the side slopes of the isolated, often conical hills, which stand like sentinel towers protecting the approach to the main mass of a castle. Why does the mountain mass appear to thrust boldly out of the plain while the valleys seem to be a smooth headward extension of the nearly flat plain?

As higher elevations are attained, the granite is left behind, the road hugs the dolomite cliffs, and pine is replaced by spruce, with occasional patches of aspen. Where a turn of the valley leads the road in a direction truly perpendicular to the main long axis of the mountain block, rather than horizontal, bedding planes of the dolomite can be seen to be dipping eastward.

The dip of the dolomite strengthens the analogy that the mountain is like a partly buried brick whose broadest face is slightly tipping rather than horizontal. The canyon which the road follows finally tops out on this dipping face of the brick, on a broad sloping land covered with pine forest.

The flat floor of the mountain valley draws attention to still another feature, characteristic also of some of the tributary valleys. The flat floors are trenched by rectangular and chutelike arroyos which, though vertical walled, meander in serpentine curves from side to side down the valley (Fig. 2-5). The 20-foot walls of these trenches expose a reddish silt with lenses of gravel, and this type of material changes but little along the length of the trough. Downstream each of these arroyos decreases imperceptibly in depth. Walls 20 feet in height are characteristic of reaches upstream, but why are they only 2 or 3 feet high near the place where the valley opens out to the promountain plain.

Another characteristic of these arroyo trenches is that where they are deep, well within the mountain mass, the floor of the trench is often composed of bed rock. Thus the arroyos, where developed to maximum depth, apparently have eroded out the full depth of the fine-grained material making up the flat valley floor.

Supplementing the observed field evidence with data from wells drilled for water, let us consider briefly some possible interpretations of the evidence, as well as a few questions that arise from it.

Outcrops on the mountain front indicate that in this area the valley of the Rio Grande must be part of a block downthrown with respect to the mountain mass which was uplifted and tilted slightly to the east. The dolomite beds comprising the upper half of the mountain appear to rest unconformably on the Precambrian granite and metamorphosed basement rocks.

Figure 2-5.

A mountain valley in which the valley alluvium has been trenched by an arroyo.

Slickensides seen on some surfaces of the mountain front and on the triangular faces of the intervalley spurs suggest that the mountain front is part of an eroded fault plane, much of which may be below the plain adjoining the front. The presence of the fault is also inferred from wells drilled in the promountain plain at various distances from the mountain front. Within a distance of about a half mile of the mountain front, wells penetrate 30 to 100 feet of unconsolidated sand and gravel, below which is rock, usually granite (D to C, Fig. 2-6). Water is usually obtained from the basal gravel lying immediately on top of the rock.

Wells drilled only slightly farther from the mountain are known to have penetrated hundreds of feet of sand and gravel and were, nearly without exception, dry holes (C to F, Fig. 2-6). These relations suggest a major fault line, ACB, roughly parallel to the mountain front and on the average about half a mile from it. That the bedrock at shallow depth is granite implies that the mountain front once stood half a mile more toward the valley than its present position.

Since the time of upthrow of the mountain block, the steep face must have progressively been worn back the half mile which now separates it from the major fault. Furthermore, in the back-wearing or retreat of the scarp, although less steep than the dip of the fault the present erosional scarp is still very steep.

Figure 2-6. Diagrammatic cross section through the Sandia Mountains, New Mexico, showing fault plane and erosional scarp retreat.

Does the back-wearing slope attain a particular declivity and retain this angle of slope as it retreats? If so, would this also occur if the environment were a humid rather than an arid climate? Even to begin to answer this kind of question requires some knowledge of the processes involved—weathering, stream action, mass movement, and probably others.

In the retreat of the mountain scarp the thorough weathering of rocks and boulders at the surface prepares them for rapid breakup into small fragments when transported by gravity or water. The steepness of the scarp appears to be associated with the large boulders that weathering produces, and such steep zones have been referred to as boulder-controlled slopes. When these large weathered boulders are sapped or undercut and begin to roll or wash down the steep slope, they smash into small fragments which can be transported even on a relatively flat slope. It has been suggested that this is a principal reason for the marked break in gradient between the mountain front and the alluvial plain.

Even if this were true, there are many other aspects to the process of slope retreat. What are the relative relations of rill development, valley cutting, and unrilled slopes to the problem of mountain-front retreat? What are the roles of mass movement, overland or sheet flow, gully erosion, and solution?

The scarp bordering the valley of the Rio Grande at E in Fig. 2-3 appears to be the result of a lowering of the channel of the river. Presumably at an earlier time the plain joined the river valley in a smooth unbroken curve when the river flowed approximately at the level of E (Fig. 2-3). Subsequently the regimen of the river was so changed that the river incised itself to its present level. This in turn caused tributaries to notch the resulting scarp, as at E. What were the relative roles of land warping, uplifting, and climatic change? What kind of a climatic change is required to alter the relation between rainfall, runoff, and sediment transport?

These effects are known only in a qualitative, not a quantitative sense. For example, it is not known how much the rainfall must change to alter significantly the rainfall-runoff relations. Nor is it clear what factors in climate would have to be changed. Is a certain change in the amount of summer rainfall more, or less, effective than an equal change in spring rainfall? Would a change in rainfall intensity without a change in the seasonal amount have the same effect?

This relatively detailed discussion of a single physiographic environment serves here only as an illustration of a large number of basic problems in geomorphology. Partial explanations of these problems can be offered, but more complete explanations require much more knowledge of processes than is presently available. Restated, a few of the important questions would be these:

What factors govern the longitudinal profile of an alluvial plain? What is the relative importance of discharge, particle size, frequency and magnitude of floods, vegetation, and infiltration losses?

What are the mechanisms by which diastrophic movements result in alteration of river slope and channel elevation? How do the adjustments of channel caused by diastrophism differ from those caused by changes in climatic factors? How will the evidence differ in either case? What climatic factors are controlling, and where and how are their effects felt?

When degradation of a channel occurs, what factors govern the way in which the tributary valleys react? How does the regrading of a rejuvenated tributary progress in time, and what controls the process of regrading? When is a local scarp or knickpoint preserved, and under what conditions is it eliminated?

What factors control the declivity and form of a scarp or hillslope? With the passage of time, through what evolutionary sequence will such a slope pass?

A Meandering River Cut into Bedrock

Any river system or terrain will pose similar problems. A briefer look at a classic physiographic region in the eastern United States will perhaps emphasize this point, while exposing some additional fundamental questions. In the present discussion small details are magnified to emphasize fundamental questions. A road map of Pennsylvania would serve as the basis for these physiographic observations; the basic features of a portion of the area are shown on the topographic map in Fig. 2-7.

The North Branch of the Susquehanna River rises in New York State north of Binghamton. Both it and a tributary to the west, the Chemung River, pass through areas of irregular hummocky topography and through valleys strewn with deposits of sand and gravel. The two join near the New York-Pennsylvania line and, after flowing south for about 20 miles to Towanda, Pennsylvania, the river flows in a southeasterly course to Scranton. In this stretch the course of the river consists of series of huge meandering loops that show clearly on a map. Each meander has a wavelength of about 3 to 4 miles and an amplitude of perhaps 3 miles. A map would not necessarily show that each meander is cut in bedrock. The regular meander pattern of the river traverses the grain or strike of the topography—here formed primarily of sandstone and shale formations. Although gravel deposits occur here and there along the highway, primarily on the insides of the meandering bends, the river throughout much of its course is actually flowing on bedrock.

Figure 2-7.

Portion of Meshoppen Quadrangle, Pennsylvania, showing large meanders of the Susquehanna River, bordered by steep bedrock cliffs.

Comparison with other streams of similar size flowing in alluvium reveals that the meander loops of the North Branch of the Susquehanna are much larger than those for a river of comparable size flowing in alluvium. Some of the gravel deposits contain cobbles and boulders indigenous to northern New England, a region now separated by a drainage divide from the Susquehanna drainage. These deposits, coupled with the lakes shown on the road map and with other deposits found in the area, provide clear evidence of glaciation.

The presence of the large meanders, however, poses several problems. First, how and when were such meanders initiated? Second, why do they appear to be unusually large? Third, how are they maintained as the stream crosses the bedrock? Was the meander form inherited from an earlier time when the river flowed at a higher level, perhaps on a depositional or erosional surface no longer evident in the present topography? Or do the large meanders suggest that at one time much larger volumes of water were carried by the Susquehanna? If so, did these larger rivers provide greater energy with which to mold the large bends in bedrock? Despite our increased understanding of meandering rivers, we can not give unequivocal answers.

To the west, in the vicinity of Williamsport, Pennsylvania, the road map shows the West Branch of the Susquehanna as it passes the city, flowing parallel to the strike of Bald Eagle Mountain, passing east around the mountain in a nose or loop and thence regaining a generally southward course. Elsewhere, however, the river does not appear to pass around the plunging mountain masses, but passes instead directly through the ridges in a series of impressive gorges. Farther south the Susquehanna flows south and then southeast and in the vicinity of Harrisburg, Pennsylvania, passes successively through one resistant bedrock ridge after another. These ridges trend roughly northeast-southwest, or at right angles to the course of the river, and most are capped by resistant quartzite or sandstone. Even the road map shows names classic in the geomorphic literature—Kittatinny Mountain, Blue Mountain, and Tuscarora Mountain. Between the ridge crests, in limestone and shale valleys, the streams tributary to the Susquehanna flow in valleys parallel to the ridge crests.

Field observations and profiles indicate that many of the ridge crests are of relatively uniform elevation. Occasionally the crests of the ridges are broken by shallow notches or depressions, which in some cases appear to be aligned with each other from ridge to ridge and with portions of the courses of the rivers that now flow far below them in the valley. The notches themselves and their alignment suggest the courses of former rivers. Some investigators have reasoned that the aligned notches and the even crests of the ridges represent remnants of a much higher surface over which the ancestral Susquehanna may once have passed. Thus it has been postulated that this ancestral river gradually eroded a course athwart the resistant mountains.

But evidence of this former surface exists only in the resistant ridge crests. A reasonable alternative hypothesis is that the apparent evenness of the crests, as well as their elevation, occurs because they are composed of resistant sandstones and quartzites. Structural weaknesses and progressive downcutting and adjustment to stratigraphic and lithologic controls must then be called upon to explain the unexpected course of the river. Neither hypothesis is as yet fully explained or validated.

Throughout the river system, flat bench lands are evident at irregular intervals along the valleys. In some places these benches or surfaces are underlain by bedrock thinly veneered with gravel. At others the surfaces are quite dissected by tributary streams, and some display characteristic soil profiles. The number of bench levels also varies from place to place. Although it is