RIVER

any natural stream of water that flows in a channel with defined banks (ultimately from Latin ripa, bank). Modern usage includes rivers that are multichanneled, intermittent, or ephemeral in flow and channels that are practically bankless. The concept of channeled surface flow, however, remains central to the definition. The word stream (derived ultimately from the Indo-European root srou-) emphasizes the fact of flow; as a noun it is synonymous with river and is often preferred in technical writing. Small natural watercourses are sometimes called rivulets, but a variety of namesincluding branch, brook, burn, and creekare more common, occurring regionally to nationally in place-names. Arroyo and (dry) wash connote ephemeral streams or their resultant channels. Tiny streams or channels are referred to as rills or runnels. Rivers are nourished by precipitation, by direct overland runoff, through springs and seepages, or from meltwater at the edges of snowfields and glaciers. The contribution of direct precipitation on the water surface is usually minute, except where much of a catchment area is occupied by lakes. River water losses result from seepage and percolation into shallow or deep aquifers (permeable rock layers that readily transmit water) and particularly from evaporation. The difference between the water input and loss sustains surface discharge or streamflow. The amount of water in river systems at any time is but a tiny fraction of the Earth's total water; 97 percent of all water is contained in the oceans and about three-quarters of fresh water is stored as land ice; nearly all the remainder occurs as groundwater. Lakes hold less than 0.5 percent of all fresh water, soil moisture accounts for about 0.05 percent, and water in river channels for roughly half as much, 0.025 percent, which represents only about one four-thousandth of the Earth's total fresh water. Water is constantly cycled through the systems of land ice, soil, lakes, groundwater (in part), and river channels, however. The discharge of rivers to the oceans delivers to these systems the equivalent of the water vapour that is blown overland and then consequently precipitated as rain or snowi.e., some 7 percent of mean annual precipitation on the globe and 30 percent of precipitation on land areas. Rivers are 100 times more effective than coastal erosion in delivering rock debris to the sea. Their rate of sediment delivery is equivalent to an average lowering of the lands by 30 centimetres (12 inches) in 9,000 years, a rate that is sufficient to remove all the existing continental relief in 25,000,000 years. Rock debris enters fluvial systems either as fragments eroded from rocky channels or in dissolved form. During transit downstream, the solid particles undergo systematic changes in size and shape, traveling as bed load or suspension load. Generally speaking, except in high latitudes and on steep coasts, little or no coarse bed load ever reaches the sea. Movement of the solid load down a river valley is irregular, both because the streamflow is irregular and because the transported material is liable to enter temporary storage, forming distinctive river-built features that range through riffles, midstream bars, point bars, floodplains, levees, alluvial fans, and river terraces. In one sense, such geomorphic features belong to the same series as deltas, estuary fills, and the terrestrial sediments of many inland basins. Rates of erosion and transportation, and comparative amounts of solid and dissolved load, vary widely from river to river. Least is known about dissolved load, which at coastal outlets is added to oceanic salt. Its concentration in tropical rivers is not necessarily high, although very high discharges can move large amounts; the dissolved load of the lowermost Amazon averages about 40 parts per million, whereas the Elbe and the Rio Grande, by contrast, average more than 800 parts per million. Suspended load for the world in general perhaps equals two and one-half times dissolved load. Well over half of suspended load is deposited at river mouths as deltaic and estuarine sediment. About one-quarter of all suspended load is estimated to come down the GangesBrahmaputra and the Huang Ho (Yellow River), which together deliver some 4,500,000,000 tons a year; the Yangtze, Indus, Amazon, and Mississippi deliver quantities ranging from about 500,000,000 to approximately 350,000,000 tons a year. Suspended sediment transport on the Huang Ho equals a denudation rate of about 3,090 tons per square kilometre (8,000 tons per square mile) per year; the corresponding rate for the GangesBrahmaputra is almost half as great. Extraordinarily high rates have been recorded for some lesser rivers: for instance, 1,060 tons per square kilometre per year on the Ching and 1,080 tons per square kilometre per year on the Lo, both of which are loess-plateau tributaries of the Huang Ho. This article concentrates on the distribution, drainage patterns, and geometry of river systems; its coverage of the latter includes a discussion of channel patterns and such related features as waterfalls. Considerable attention is also given to fluvial landforms and to the processes involved in their formation. Additional information about the action of flowing water on the Earth's surface is provided in the article structural landform: Stream valleys and canyons. Certain aspects of the changes in rivers through time are described in climate: Atmospheric humidity and precipitation: Effects of precipitation, and the general interrelationship of river systems to other components of the Earth's hydrosphere is treated in hydrosphere: Biochemical properties of the hydrosphere: Rivers and ocean waters. For information concerning the plant and animal forms that inhabit the riverine environment, see inland water ecosystem: Aquatic ecosystems: Riverine ecosystems. any natural stream of water that flows in a channel with more or less defined banks. Modern usage of the term includes multichanneled, intermittent, and ephemeral streams, as well as those in bankless channels; the concept of channeled surface flow, however, remains central. The word stream is frequently preferred in technical writing, where it is used as an inclusive term that refers to all natural waterways, from the largest rivers to the smallest creeks. Rivers constitute a fundamental link in the hydrologic cycle, the circulation of water from the oceans through the atmosphere (evaporation) to the land via precipitation and back to the oceans by way of rivers and, to a lesser degree, groundwater discharge. As one of the most important geomorphic agents, rivers play a major role in shaping the surface features of the Earth. Even apparently arid desert regions are greatly influenced by fluvial action when periodic floodwaters surge down usually dry watercourses. Throughout history, rivers have provided one of the easiest and, in some areas of the world, the only means of entry and passage for explorers, traders, conquerors, and settlers. Rivers assumed great importance, for example, in Europe after the fall of the Roman Empire and the decay of its road system. Rivers in medieval Europe supplied the water that sustained the growth of cities and were widely used as sources of power. Western European history records the rise of more than a dozen national capitals on sizable rivers, exclusive of seawater inlets. In modern history, both in North America and northern Asia, natural waterways dictated in large part the lines of exploration, conquest, and settlement. River flow is sustained by the difference between water input and output. Rivers are nourished by overland runoff, by groundwater seepages, and by meltwater released along the edges of snowfields and glaciers. Direct precipitation also contributes to river flow, but generally in very small amounts. Losses of river water result from percolation into layers of porous and permeable rock, gravel, or sand, evaporation, and ultimately outflow into the ocean. The shape of a river system provides evidence of the geologic and topographical factors that have been involved in its formation and development. In a dendritic river system, for example, the absence of a particular pattern of tributaries indicates that the system has had undisturbed growth, with the composition and structure of the underlying rock exerting no discernible influence on the stream course. A river pattern characterized by sharp right-angled bends, however, is one that is following the joints in the underlying bedrock. A river system may be trellised, displaying the influence of alternately hard and soft bands of bedrock. Sometimes there may be few tributary streams, strongly suggesting that, as in the case of limestone, the bedrock is permeable and drainage is going underground. Rivers, especially those in humid regions, can be considered as branching systems of often innumerable tributaries. Seasonal variations in discharge define the regime of a river. Variations in the amount of precipitation can cause significant changes in discharge. The annual floods on the Nile, for example, reflect high summer rainfall in the highlands of Ethiopia. Rivers that rise in mountainous areas with perennial snow cover will also show peak flows relating to seasonal snow and glacier melt. Some rivers in the high northern latitudes have diminished flows in winter as their catchments freeze over. Rivers in humid equatorial areas, however, vary only slightly throughout the year. The ability of a river to alter the landscape is demonstrated most clearly in the channel it cuts. A river may erode its channel by dissolving the rock over which it flows. Most riverine erosion is caused by abrasion. Particles transported by the flowing water, which range in size from silt to small boulders, collide with the riverbed and break off other pieces; the cumulative action of these particles can be likened to a file gradually smoothing a surface. The effectiveness of abrasion depends on the rate of water flow and the amount of the load that a river carries. The net effect of channel erosion is to carve a valley. Channel erosion tends to cut downward, but few river valleys remain steep-sided, V-shaped ravines. The form of a river valley is significantly affected over time by other processes such as weathering and mass-wasting (movement of mantle rock) as well as by the cumulative effect of erosion by tributary channels and the lateral movement of the main channel. Small steep-sided valleys, commonly called canyons and gorges, usually occur in upland areas. In such cases, the valley sides and riverbanks are for the most part coextensive. In river valleys in lowland areas, where the channel extends across a floodplain, no direct contact generally occurs between the channel and the valley sides. The tendency of a river to curve and to develop broad looping curves is called meandering. A river heavily loaded with sediment may deposit the material at a rate that causes the channel to break up into several smaller channels. These channels typically diverge and converge, forming a braided pattern. Deposition from rivers occurs when the velocity of flow is checked. This may occur when a river enters the ocean or a lake and creates a delta (q.v.) or when a sudden change in gradient at the mouth of a canyon or valley causes the river to form an alluvial fan (q.v.). In time of flood, river water spreads out from its overflowing channel and deposits its load over a wide area. This relatively flat surface constitutes the floodplain and is perhaps the most common of fluvial features. Additional reading General works Discussions of all aspects of rivers are found in Luna B. Leopold, M. Gordon Wolman, and John P. Miller, Fluvial Processes in Geomorphology (1964); G.H. Dury (ed.), Rivers and River Terraces (1970); Richard J. Chorley, Stanley A. Schumm, and David E. Sugden, Geomorphology (1984); Arthur L. Bloom, Geomorphology: A Systematic Analysis of Late Cenozoic Landforms (1978); Dale F. Ritter, Process Geomorphology, 2nd ed. (1986); and Marie Morisawa, Rivers: Form and Process (1985). See also Laurence Pringle, Rivers and Lakes (1985); and the Rand McNally Encyclopedia of World Rivers (1980).Environmental problems attendant on river use are discussed in M.J. Stiff (ed.), River Pollution Control (1980); Environmental Effects of Cooling Systems (1980), a report from the International Atomic Energy Agency on cooling systems and thermal discharges from nuclear power stations; Cooling Water Discharges from Coal Fired Power Plants: Water Pollution Problems (1983), proceedings of an international conference; R.G. Toms, River Pollution-Control Since 1974, Water Pollution Control, 84(2):178186 (1985); and M. Chevreuil, A. Chesterikoff, and R. Letolle, PCB Pollution Behavior in the River Seine, Water Research, 21(4): 427434 (April 1987). George Harry Dury Dale F. Ritter The Editors of the Encyclopdia Britannica The Editors of the Encyclopdia Britannica River channels and waterfalls Works on the formation and change of river channels include Walter B. Langbein and Luna B. Leopold, River Meanders, Theory of Minimum Variance (1966), U.S. Geological Survey professional paper no. 422-H; Mark A. Melton, Methods for Measuring the Effect of Environmental Factors on Channel Properties, Journal of Geophysical Research, 67(4):148590 (April 1962); and N.A. Rzhanitsyn, Morphological and Hydrological Regularities of the River Net (1964; originally published in Russian, 1960).A dated but still useful source on waterfalls is the article by Theodore W. Noyes, The World's Greatest Waterfalls, National Geographic Magazine, 50:2959 (July 1926), on the Niagara, Victoria, and Iguau falls. Modern treatments of waterfalls are rare; the interested reader might best consult the following references: H.F. Garner, Derangement of the Rio Caroni, Venezuela, Revue de Gomorphologie Dynamique, 16:5483 (1966), describing the occurrence of Angel Falls; Martin von Schwarzbach, Islndische Wasserflle und eine genetische Systematik der Wasserflle berhaupt, Zeitschrift fr Geomorphologie, 11:377417 (Dec. 1967), one of the best general surveys of the several kinds and occurrences of waterfalls, with specific reference to Icelandic examples; Shailer S. Philbrick, Horizontal Configuration and the Rate of Erosion of Niagara Falls, Geological Society of America Bulletin, 81(2):372331 (Dec. 1970), providing a summary of information on the history of Horseshoe Falls and on the general recession of cap-rock-type falls; Eberhard Czaya, Waterfalls and Rapids, ch. 4 in his Rivers of the World (1981, reprinted 1983; originally published in German, 1981), pp. 121137, which includes a list of famous waterfalls classified by location and height; and R.W. Young, Waterfalls: Form and Process, Zeitschrift fr Geomorphologie Supplementband, 55:8195 (1981). George Harry Dury Lawrence K. Lustig Rivers as agents of landscape evolution Literature concerning the evolution of valleys and the origin of transverse canyons is found mostly in older classic treatments of the topic, such as William Morris Davis, Geographical Essays (1909, reprinted 1954); and two articles from the Bulletin of the Geological Society of America: Arthur N. Strahler, Hypotheses of Stream Development in the Folded Appalachians of Pennsylvania, 56 (1):4587 (Jan. 1945); and J. Hoover Mackin, Erosional History of the Big Horn Basin, Wyoming, 48(6):813893 (June 1, 1937). An excellent discussion of the initial development of valleys and canyons in an area of recent tectonism is given in Theodore Oberlander, The Zagros Streams: A New Interpretation of Transverse Drainage in an Orogenic Zone (1965). Discussions and additional references about valley morphology in natural and experimental settings can be found in Stanley A. Schumm, The Fluvial System (1977); and Stanley A. Schumm, M. Paul Mosley, and William E. Weaver, Experimental Fluvial Geomorphology (1987).Detailed analyses of floodplains are provided by Edward J. Hickin and Gerald C. Nanson, The Character of Channel Migration on the Beatton River, Northeast British Columbia, Geological Society of America Bulletin, 86(4):487494 (April 1975); and by two U.S. Geological Survey professional papers: M. Gordon Wolman and Luna B. Leopold, River Flood Plains: Some Observations on Their Formation (1957), no. 282-C; and Stanley A. Schumm and R.W. Lichty, Channel Widening and Flood-Plain Construction Along Cimarron River in Southwestern Kansas (1963), no. 352-D.Detailed discussions of terrace formation can be found in Luna B. Leopold and John P. Miller, A Postglacial Chronology for Some Alluvial Valleys in Wyoming (1954), U.S. Geological Survey water-supply paper no. 1261; and two papers in the Geological Society of America Bulletin: John H. Moss and William Bonini, Seismic Evidence Supporting a New Interpretation of the Cody Terrace near Cody, Wyoming, 72(4):547555 (April 1961); and Dale F. Ritter, Complex River Terrace Development in the Nenana Valley near Healy, Alaska, 93(4):346356 (April 1982).Important papers treating processes and characteristics of alluvial fans in detail include two U.S. Geological Survey professional papers: William B. Bull, Geomorphology of Segmented Alluvial Fans in Western Fresno County, California (1964), no. 352-E; and Charles S. Denny, Alluvial Fans in the Death Valley Region, California and Nevada (1965), no. 466. See also R. Craig Kochel and Robert A. Johnson, Geomorphology and Sedimentology of Humid-Temperate Alluvial Fans, Central Virginia, in Emlyn H. Koster and Ron J. Steel (eds.), Sedimentology of Gravels and Conglomerates (1984), pp. 109122; Neil A. Wells and John A. Dorr, Jr., Shifting of the Kosi River, Northern India, Geology, 15(3):204207 (March 1987); and Richard H. Kesel, Alluvial Fan Systems in a Wet-Tropical Environment, Costa Rica, National Geographic Resources, 1(4):450469 (Autumn 1985). More general reviews and discussions of experimental work are Roger LeB. Hooke, Processes on Arid-Region Alluvial Fans, The Journal of Geology, 75(4):438460 (July 1967); and William B. Bull, Alluvial Fans, Journal of Geological Education, 16(3):101106 (June 1968).Extensive reviews of delta formation can be found in Martha L. Shirley (ed.), Deltas in Their Geologic Framework (1966); W. Fisher et al., Delta Systems in the Exploration for Oil and Gas (1969, reprinted 1974); and James P. Morgan, DeltasA Rsum, Journal of Geological Education, 18(3):107117 (May 1970). Detailed analyses of specific deltas, including the Mississippi River delta, can be found in Francis P. Shepard, Fred B. Phleger, and Tjeerd H. Van Andel (eds.), Recent Sediments, Northwest Gulf of Mexico (1960); and Richard J. Russell, Geomorphology of the Rhne Delta, Annals of the Association of American Geographers, 32(2):149254 (June 1942), and River and Delta Morphology (1967).The characteristics and formative processes of estuaries are discussed in Andr Guilcher, Coastal and Submarine Morphology (1958); Maurice L. Schwartz (ed.), The Encyclopedia of Beaches and Coastal Environments (1982), with illustrated entries on estuaries and on estuarine coasts, deltas, habitats, and sedimentation; Russell Sackett, Edge of the Sea, rev. ed. (1985); and Eric C.F. Bird, Coasts: An Introduction to Coastal Geomorphology, 3rd ed. (1984). More technical treatment is presented in George H. Lauff (ed.), Estuaries (1967); Bruce W. Nelson (ed.), Environmental Framework of Coastal Plain Estuaries (1973); and L. Eugene Cronin (ed.), Estuarine Research, vol. l, Chemistry, Biology, and the Estuarine System (1975). Estuarine sediment is described for 45 estuarine zones of the United States in David W. Folger, Characteristics of Estuarine Sediments of the United States (1972), U.S. Geological Survey professional paper no. 742. Dale F. Ritter