At the beginning of Tertiary time an extensive land surface existed. While streams were laying down widespread sheets of sands in the plains area to the north the land surface in western Kansas was being subjected to erosion largely effected by through-flowing streams. As a result great thicknesses of Upper Cretaceous sediments were removed, so that in Ford County all of the strata above the Greenhorn limestone are missing. In parts of Ford County the Greenhorn and Graneros have been removed entirely by erosion, and in the southeastern part of the county the Dakota formation is absent and the Tertiary formations rest upon the Kiowa shale. The surface of the Dakota formation was also subjected to erosion in the areas in which the overlying Cretaceous rocks were completely removed. The bedrock surface thus had a topography of considerable diversity and relief prior to the deposition of the Ogallala formation.
Factors other than erosion also contributed to the configuration of the bedrock floor during Tertiary time. Crustal deformation is believed to have taken place either before or during the deposition of the Ogallala formation. In late Tertiary time erosion was followed by an epoch of deposition during which heavily-laden streams from the Rocky Mountains traversed western Kansas and deposited the sediments of the Ogallala formation in a broad alluvial plain. Through-flowing streams occupied the main valleys, and deposition continued after the valleys were filled until the interstream divides were covered with a nearly continuous mantle of sediments.
Several causes have been postulated by various authors for the change in behavior of the traversing streams. Interruptions in the gradation cycle from a period of erosion to one of deposition and reversal again finally to erosion have been attributed by Haworth (1897, p. 14) to the instability of the structural slope. Haworth correlated deposition with different periods of rapid elevation intervening between long periods of low upraising with repeated increases in stream gradients. Johnson (1901, p. 628) suggested that the effect of climatic oscillation rather than earth movement was responsible for the conditions whereby streams in Tertiary time were alternately degrading and aggrading. Darton (1920, p. 8) emphasized the effect of differential uplift upon Tertiary streams, but admitted climatic conditions as a contributing factor.
Near the close of middle Pliocene time there may have been some upwarping of the surface to the west of this area (Smith, 1940, p. 94) resulting in the rejuvenation of the master streams and in bringing to a close a period of widespread deposition of the Ogallala. The upper Pliocene part of the Ogallala, the Rexroad member, represents deposits that were trapped in local areas of downwarping and down-faulting. The principal area in which deposits were trapped during this time was the downfaulted area in Meade County, which probably extended into Ford and Clark counties. Smith (1940) suggests that:
"The Rexroad was laid down on the deformed surface of the Ogallala, in a structural and topographic expression progressively modified by marginal erosion concurrent with deposition."
On the basis of test holes in southern Ford County, it is known that deposits of gravel and interbedded limy clay assigned to the Rexroad member rest unconformably on the Cretaceous, the middle Pliocene being absent locally. The Rexroad seems to contain more or less reworked material from the Ogallala, but a large part of it may have been derived from the same source area as the Ogallala.
Crustal deformation is believed to have taken place after the deposition of the Ogallala. In the southwestern part of the county the anomalous reversal in the trend of Crooked Creek was brought about by the combined factors of faulting, downwarping, stream erosion and deposition, and solution collapse in the underlying Permian beds. Haworth (1897, pp. 22, 23) emphasized the importance of tectonic movement in explaining this change in the direction of Crooked Creek in the vicinity of Wilburn, a former post office in the SW sec. 35, T. 29 S., R. 26 W. He writes as follows:
"The sharp angle at Wilburn and the southwestern direction for nearly 20 miles across a plain sloping to the southeast are certainly very remarkable, and probably have a cause different from that which ordinarily determines the location and direction of streams. But if in post-Tertiary times a triangular area, equaling in size and position the present artesian area, could have dropped 100 feet or more, with a single fault line extending southward to beyond the limits of Kansas, thereby changing the direction of Crooked Creek into the present channel below Wilburn, the general physiographic conditions could easily be accounted for."
A geologic section across the southern border of Ford County, based on test holes (pl. 5, section C-C.), shows the position of the inferred fault in the vicinity of Crooked Creek a. well as the stratigraphy of that part of the county.
A diversified topography comprising a series of basins along the structural trough was developed in late Pliocene time by the interaction of faulting and solution. In these basins were trapped the thick deposits of upper Pliocene sediments comprising the Rexroad member of the Ogallala formation (Frye and Hibbard, 1941, pp. 394, 395).
A disconformity exists between the Rexroad member of the Ogallala and the overlying Pleistocene and Recent Kingsdown silt in Ford County. During this interval deposits comprising the Pleistocene sands and gravels of the Meade formation were being laid down in Meade and Clark counties to the south and may have been deposited locally in Ford County.
Frye and Hibbard (1941, p. 397) suggest that the deposits of fine sand and silt that constitute the lower part of the Kingsdown were laid down during late Pleistocene time by streams that must have shifted laterally at frequent intervals. They also point out that stream deposition probably was accompanied and followed by eolian activity, because the fluvial deposits grade upward into loess, and that the deposition of the stream-laid deposits of the Kingsdown marked the close of the Pleistocene. They add that the upper eolian part of the Kingsdown may have been deposited in the Recent epoch.
According to Smith (1940, p. 115), a part of the Kingsdown was derived from the Ogallala formation. This would indicate that there was an interval of erosion in adjacent areas concurrent with the deposition of the Kingsdown silt in Ford County. He points out that
"The abundant silt characterizing the greater part of the formation, however, seems adequately explained only by derivation from eolian loess. The even bedding and fine lamination indicate subaqueous deposition, either by flood waters or under lacustrine conditions.... The thickness of the formation, however, is too great to be explained wholly on this basis, and fits better with the postulate of material supplied by erosion of loess deposits, either contemporaneous and, or, older, to the west."
He further suggests that deposition must have been due to flattening or actual reversal of stream gradients as a result of crustal warping, with climatic changes as a minor contributing factor.
The transition from the deposition of the water-laid sediments of the Kingsdown to the deposition of the overlying loess mantle which forms the upper part of the Kingsdown is indistinguishable. The loess is possibly contemporaneous with and in part grades laterally into the water-laid deposits. This seems to indicate that during the latter part of the Pleistocene there was a change in climate from humid to relatively arid conditions with considerable wind movement. Loess deposition has continued intermittently until recent time. The erosional development of the area at the time of loess deposition did not differ greatly from that of the present only in that the valleys were not yet quite so deep.
At the beginning of the Recent epoch the major streams began the down-cutting that has produced the present topography. In Ford County there are two recognizable terraces along the Arkansas valley. The terrace deposits are best displayed on the south side of the river, the lower terrace lying at about 5 to 8 feet above flood-plain level and the higher terrace occurring at from 15 to 25 feet above the flood plain. Smith (1940, p. 150) suggests that the terraces are of cut-and-fill origin and of late-glacial or post-glacial age. The terrace deposits are younger than the Kings-down silt, as evidenced by their lower topographic position. It is probable that much of the material in the terrace gravels was derived from source areas to the west and was deposited during late Pleistocene or Recent time when the river was at a higher elevation. Dune sand is widely distributed along the south side of the Arkansas valley in Ford County. Smaller areas of dune sand occur also in other parts of the county, notably north of the river in the vicinity of Ford and in the southwestern part of the county (pl. 1). The age of the dune sand is not definitely known, but it is probable that accumulation of some of the sand started in late Pleistocene time and continued at different times and in different places up until recent times. Darton (1920, p. 3) believed that the present river flood plains constituted the source of the material. Smith (1940, pp. 165-167) disagrees with this mode of origin and suggests that the sand was derived from the terrace deposits, and that toward the southern part of the dune belt the sand originated possibly from sands of the Rexroad member of the Ogallala formation, or the Kingsdown silt, or both, or simply from the denuded slopes cut in the Ogallala. He also suggests that the presence of a continental ice sheet during one or more of the Pleistocene glacial stages could have been responsible for the northerly dune-building winds of the past.
Since the close of the Pleistocene epoch the general aspect of the area has not changed materially. Adjustments in the courses and gradients of streams undoubtedly took place. Some of the streams have deepened their channels, whereas other streams have built up their flood plains. When a stream has its volume diminished or has its efficiency lowered by a decrease in its velocity, it may become overburdened with sediment, in which case it will deposit part of its load. This process is known as aggradation or alluviation. Flood plains are alluvial flats that may be covered by water at times of flood. During floods new layers of sediment are added to the flood plain.
The Arkansas River has no large tributaries in Kansas above the mouth of Pawnee River, which joins the Arkansas at Larned. According to George S. Knapp (personal communication, April 30, 1942), who kindly supplied the figures given in the preceding table, large floods that originate in Colorado proceed down the valley with constantly diminishing volume.
|Flood Peaks on the Arkansas River in 1923|
|Date||Maximum flow at Syracuse (second feet)|
|Date||Maximum flow at Larned (second feet)|
As indicated in the same table, during the flood of May, 1923, the maximum discharge of 11,200 cubic feet per second at Syracuse gradually diminished downstream until at Larned, three days later, the maximum discharge was only 2,990 cubic feet per second, a reduction in volume of 8,210 cubic feet per second. A large part of the water thus "lost" from the flooded stream must have percolated into the porous alluvium, thus recharging the ground water reservoir; and a large part of the load of sediment carried by the floodwaters must have been deposited on the flood plain of the river. Thus, Arkansas River is an aggrading stream, at least in the part of its course between Syracuse and Larned.
Several changes in stream courses took place during Pleistocene time. Streams were rejuvenated and cut deeply into the plains surface following uplift to the west, and the original courses of these streams were later altered by piracy to form the present drainage pattern. Arkansas River throughout most of its course in Ford County has not cut its channel as deeply as have the smaller streams in the northern part of the county. At a point due north of Dodge City the stream bed of Sawlog Creek is approximately 50 feet lower than the stream bed of Arkansas River at Dodge City. South of the river, however, Mulberry Creek has not cut its channel as deeply into the plains surface as have the streams north of the river, the creek bed at a point due south of Dodge City being about 55 feet higher than that of Arkansas River at Dodge City. The "perched" position of Arkansas River has been discussed by Smith (1940, pp. 146-149), who points out that crustal warping within the Kansas area probably played an important part in raising, or in preventing the lowering of, the stream's gradient along this part of its course, and that this condition was closely related to the events responsible for the great bend of the river just to the east.
Haworth (1897b, p. 30) postulated an altogether different cause for the great and unusual bend that the river makes in passing from eastern Ford County to the vicinity of Great Bend. He pointed out that the river encountered the easily-eroded Dakota formation where it rises above the surface in eastern Ford County and attacked it with great vigor, following it as far north as Great Bend. As the general inclination of the Dakota formation is to the northeast, the river throughout this interval would follow the path of least resistance, and would therefore migrate northward, downdip with the strata. This hypothesis hardly seems tenable in the light of present knowledge.
It is quite significant that west of the great bend in Ford County no tributaries enter Arkansas River from the north or the south, the river appearing to flow along a ridge trending east and west. Mulberry Creek joins Arkansas River at Ford, within the bend area. Below the bend the drainage becomes more natural; tributaries enter the Arkansas from both sides, and the river no longer flows along a ridge. The absence of bluffs north of the river near the eastern edge of Ford County is also a noteworthy feature.
Kansas Geological Survey, Ford County Geohydrology
Web version April 2002. Original publication date Dec. 1942.
Comments to email@example.com
The URL for this page is http://www.kgs.ku.edu/General/Geology/Ford/04_geol4.html