to 46 m/yr (10 to 150 ft/yr); Mississippi River near Rosedale, Mississippi, 48 to 192 m/yr (158 to
630 ft/yr).
Archaeologists have also provided clear evidence of channel changes that are completely
natural and to be expected. For example, the number of archaeologic sites of the floodplains
decreases significantly with age because the earliest sites are destroyed as floodplains are
modified by river migration. Lathrop (1968), working on the Rio Ycayali in the Amazon
headwaters of Peru, estimates that on the average a meander loop on this river begins to form
and cuts off in 5,000 years. These loops have an amplitude of 3.2 to 9.7 km (2 to 6 mi) and an
average rate of meander growth of approximately 12 m/yr (40 ft/yr).
A study by Schmuddle (1963) shows that about one-third of the floodplain of the Missouri River
over the 274 km (170 mi) reach between Glasgow and St. Charles, Missouri, was reworked by
the river between 1879 and 1930. On the Lower Mississippi River, bend migration was on the
order of 0.6 m/yr (2 ft/yr), whereas in the central and upper parts of the river, below Cairo, it
was at times 305 m/yr (1,000 ft/yr) (Kolb 1963). On the other hand, a meander loop pattern of
the lower Ohio River has altered very little during the past thousand years (Alexander and
Nunnally 1972).
Although the dynamic behavior of perennial streams is impressive, the modification of rivers in
arid and semi-arid regions and especially of ephemeral (flowing occasionally) stream channels
is startling. A study of floodplain vegetation and the distribution of trees in different age groups
led Everitt (1968) to the conclusion that about half of the Little Missouri River floodplain in
western North Dakota was reworked in 69 years.
Historical and field studies by Smith (1940) show that floodplain destruction occurred during
major floods on rivers of the Great Plains. As exceptional example of this is the Cimarron
River of Southwestern Kansas, which was 15 m (50 ft) wide during the latter part of the 19th
and first part of the 20th centuries (Schumm and Lichty 1957). Following a series of major
floods during the 1930s it widened to 366 m (1,200 ft) and the channel occupied essentially the
entire valley floor. During the decade of the 1940s a new floodplain was constructed, and the
river width was reduced to about 152 m (500 ft) in 1960. Equally dramatic changes of channel
dimensions have occurred along the North and South Platte Rivers in Nebraska and Colorado
as a result of control of flood peaks by reservoir construction, decrease in annual flow by
irrigation and restriction of channel width by bridges. In their natural state the rivers were 600
to 1,500 m (2000 to 5,000 ft) wide but now are less than 300 m (1,000 ft). Changes of this
magnitude due to changes in flow are perhaps exceptional, but emphasize the mobility of
rivers and their ability to adapt to changing conditions.
Another somewhat different type of channel modification, which testifies to the rapidity of fluvial
processes, is described by Shull (1922, 1944). During a major flood in 1913, a barge became
stranded in a chute of the Mississippi River near Columbus, Kentucky. The barge induced
deposition in the chute and an island formed. In 1919, the island was sufficiently large to be
homesteaded, and a few acres were cleared for agricultural purposes. By 1933, the side
channel separating the island from the mainland had filled to the extent that the island became
part of Missouri. The island formed in a location protected from the erosive effects of floods
but susceptible to deposition of sediment during floods. For these reasons the channel filling
was rapid and progressive. It cannot be concluded that islands will always form and side
channels fill at such rapid rates, but island formation and side-channel filling appear to be the
normal course of events in any river transporting moderate or high sediment loads regardless
of the river size. These topics are discussed in detail in Chapter 5.
1.6
Previous Page
