Geomorphology of Cheyenne Bottoms

Blake Smotherman

ES 546 Field Geomorphology

Abstract Introduction Previous Research Geologic Analysis
Soil Analysis Discussion References


Cheyenne Bottoms is an approximately 166 square kilometer topographic low in central Barton County, Kansas. This low is presently occupied by a large playa and wetland system with two streams, Blood Creek and Confusion Creek, feeding into it from the northwest and one, Cheyenne Creek, leaving in the southeast. During a recent field expedition, geologic and sedimentologic evidence was gathered to attempt to explain the existence of the feature. This information, coupled with an examination of previous investigations, was synthesized in order to better explain the geologic and geomorphologic processes that have created this feature by observing land forms and rock outcrops at the current land surface, and visually analyzing soil and rock samples. 


Over the course of three days in October, soil and rock samples were gathered from selected stops in the area in and around Cheyenne Bottoms. The soil samples collected were samples of sand and clay from soil exposures, 'fossil' sand dunes at Camp Aldrich, and the marsh lands near Confusion Creek and Blood Creek. Geologic samples were collected from outcrops along the rim of Cheyenne Bottoms and from outcrops of similar lithology from nearby sites. These samples were mostly sandstones and limestones.
  Shaded relief map showing location of Cheyenne Bottoms, Deception Creek and Blood Creek (Aber and Pavri, 2004).

Previous Research

There have been multiple theories about the geologic origin of the topographic low that Cheyenne Bottoms occupies. Haworth (1897) attributed the feature to stream erosion. Johnson (1901) proposed the removal of soluble masses of salt in the underlying bedrock. Bass (1926) prepared a map of salt thickness in western Kansas that shows thinning of the salt beds beneath Cheyenne Bottoms. Latta (1950) indicated that salt solution and subsidence may have been in part responsible for the feature, but also indicated that stream erosion during the Pleistocene played an important part in its origin and present configuration. Bayne's (1977) analysis of the geology and structure of Cheyenne Bottoms concluded that fault movement into the Heebner Shale, and possibly starting at the Precambrian basement rock, could be the structural cause of Cheyenne Bottoms.


Bayne's (1977) isopach maps of the bedrock-alluvium contact(left) and the top of the Stone Corral Formation (right). Based on well logs used by Latta (1950).

A bedrock ridge extends southward beneath the sand dunes on the east side of the basin. The Blood Creek-Cheyenne Bottoms channel slopes toward the southeast to Ellinwood, Kansas where it turns northeast (Bayne, 1977). A bedrock high in the southeast corner of the study area probably caused this deflection (Bayne, 1977). The angle of deflection is the same as many of the northeast-southwest trending faults seen in central Kansas.

When analyzing Bayne's map of the Stone Corral Formation, it appears that there is one northwest-southeast trending structure and two northeast-southwest trending structures. This map was imported into Arcmap and hypothetical faults were assumed to be in the synclinal structures. These areas were lined with red trend lines. These structures are roughly parallel with the direction of geologic lineaments in the Precambrian basement rocks that underly the area. (see KGS gravitational magnetometry map). The pattern and trend of these lineaments are oriented in a different configuration and direction from the overlying pattern of north-south trending, solution-widened fractures observed in the Cretaceous bedrock at the surface.

Map adapted from Bayne (1977) showing hypothetical faults trending at northwest-southeast and northeast-southwest orientations.

A gravitational magnetometry map with Precambrian structural faults. Note the concentration of perpendicular faulting in the Cheyenne Bottoms area. These faults may propagate through the overlying bedrock, contributing to the creation of a structural and topographic low that Cheyenne Bottoms occupies (Xia et. al, 1995).

A gravitational magnetometry map produced by the Kansas Geological Survey shows the regional structure dominated by the Central Kansas Uplift. Newer faults generated by the upwarping of this feature may contribute to its expression as a topographic low and to the occupation of a playa and alluvial van within its boundaries (Xia et. al, 1995).


In conjunction with the faulting detailed above, there could be karst solution weathering in the subsurface. This weathering could be accentuated by faults under Cheyenne Bottoms. These faults could act as preferential pathways for the movement of water into the subsurface.

Diagram demonstrating how paleokarst can form along preferential pathways of groundwater flow associated with fault zones and bedding planes (Walters, 1978). Note some similarities with Gentile's (1981) interpretation of the Belton Ring Fault Structure detailed below.

Cross section from Walters (1978) demonstrating depth from the surface to the Hutchison Salt Member of the Wellington Formation. If an unconformity exists above the Hutchinson Salt, indicating a prolonged period of erosion at or near the contact, solution weathering of the salt could have occurred, then begun collapsing at a later time due to the overlying material. Descriptions from well logs (Latta, 1950) and the 'Stratigraphic Succession in Kansas' (Zeller et. al, 1968) seem to indicate that the cyclothemic nature of the Wellington Formation could have allowed such dissolution to happen. Some salt solution weathering, along with Bayne's (1977) deep fault movement, aeolian erosion and lacustrine compaction and soil formation during the Pleistocene could have all played a part in the formation of Cheyenne Bottoms.

Cross section from Latta (1950) showing underlying stratigraphy of Cheyenne Bottoms according to well logs.


Shaw (1915) showed that lakes can impound behind streams that are rapidly receiving large amounts of alluvium. He was looking at large Pleistocene lakes that formed in Illinois as glacial alluvium was transported down the Mississippi River. Perhaps similar factors could have been at work at Cheyenne Bottoms as Pliocene-age alluvial material was deposited across Kansas to the Flint Hills, concurrently building impoundments that formed Pleistocene lakes such as the one that once occupied Cheyenne Bottoms (Aber, 2009, personal communication).


A comparably sized structure has been identified in the Kansas City, Missouri area by Gentile (1983). The Belton Ring Fault Complex has been interpreted as a collapse feature caused by karsted Mississippian-age Burlington Limestone that has propogated through overlying Pennsylvanian-age strata. It measures approximately eight kilometers across, and is located under Belton, Missouri, south of Kansas City. It is easily identified as it has pirated streams in the area and routed them through the circular faults in the uppermost bedrock. The karst formation is interpreted to have formed during an episode of uplift and erosion in which the Burlington Limestone was exposed to the surface, formed karst features such as sinkholes and macropore conduits, then was overlain by Pennsylvanian-age Lansing Group and Kansas City Group cyclothem deposits composed of thin beds of shale, limestone, sandstone and coal.

The Belton Ring Fault Complex. The vegetation demonstrates how stream channels in the area have been pirated to follow the faults in the underlying bedrock. Image generated in ArcMap 9.3.

Faults begin to propagate from the Mississippian-age strata through the overlying Pennsylvanian-age strata. (Gentile,1983)

Differential subsidence occurs as collapse in the Mississippian strata continues, producing an uneven horst and graben expression on the surface. (Gentile, 1983)

Flow of the West Fork of East Creek downcuts and intercepts the lystric curvature produced by the fault complex, causing the creek to adjust its flow to the fault expressions in the Pennsylvanian rocks. (Gentile, 1983)

Cross section from Gentile (1983) that shows faulted structure in the Pennsylvanian-age strata.

(Left) GIS map of the hydrographic profile at the Belton Ring Fault Structure (Missouri Geological Survey, 2006). The arrow represents general groundwater gradient. Black lines represent the groundwater gradient, black dots represent wells that the gradient is derived from. Yellow, green and tan lines represent faults, synclines and anticlines in the Pennsylvanian-age bedrock. Though paleokarst at the Mississippian-Pennsylvanian unconformity is thought to have formed the structure, upper groundwater movement is isolated at or near the surface due to the low permeability of the shale and shale residuum derived from the breakdown of the Pleasanton Group, generating a relatively flat groundwater surface over the feature. (Right) Cheyenne Bottoms also has a flat groundwater gradient relative to the regional groundwater hydrography. This would also seem to indicate that, even if there is subsurface faulting and karst present, it does not effect the uppermost groundwater gradient.

Unlike Cheyenne Bottoms and similar features that show evidence of wetter conditions in the past, the Belton structure does not appear to be associated with any large fluvial feature such as a relict channel or evidence of having lacustrine features associated with it in the Holocene or Pleistocene. The Belton Structure is today a topographic and structural high between several rivers in the area. Therefore, this feature may only serve as a possible analog for the subsurface bedrock beneath Cheyenne Bottoms. Any solution weathering under Cheyenne Bottoms would seem to have to be the result of paleokarst in the Permian or Ordovician bedrock. Today, there is no evidence of substantial amounts of surface water having the hydraulic head or the preferential pathways needed to form karst in the Cretaceous-age Dakota Formation or to penetrate over 1500 feet of bedrock to the Permian-age Hutchinson Salt (Fetter, 2001).

Geologic Analysis

General Geologic Map of the area around Cheyenne Bottoms (Kansas Surficial Geology Map, 2008)

This sample of limestone was taken from an outcrop on the north rim of Cheyenne Bottoms, at a small scenic stop on the south side of Highway 4 near Redwing, Kansas. Note the pelycepod shell fragments in a shaly limestone matrix, possibly belonging to Ostrea or Inoceramus. Some sand was noted in the lime matrix. The rock indicates deposition in a high-energy tidal zone. Though initially identified the Dakota Sandstone, this may indicate a gradational change from the upper Dakota Sandstone to the lower Graneros shale (Zeller et. al, 1968)


The Dakota Formation in this area is most often a brown and tan, interbedded and lenticular sandstone (Franks, 1966). It contains sandstone weakly cemented with calcite or strongly cemented with iron oxide. Granular aggregates of iron oxide are seen on weathered surfaces.

The following samples of the Dakota sandstone was taken from Pawnee Rock, north of Pawnee, Kansas. Note several voids in the sample, indicating some development of macropore permeability in the formation. The upper strata of the outcrop is well cemented with iron oxide, almost an orthoquartzite. Below, it is a friable sandstone. Some conduit development was observed along bedding planes in the lower part of the outcrop (see image below), indicating heterogeneous permeabilities in the outcrop. This means a head of pressure on groundwater, and the chemical/ mechanical ability to mobilize certain minerals.

For example, manganese is transported in solution as groundwater passes through the sandstone, depositing skins and cements of manganese oxide in bedding planes,fracture zones and between sand grains (Ritter, 2002). Similar coatings are seen on some sand grains from the Camp Aldritch dune samples, possibly indicating that some of the sand in those dunes has been recycled from the Dakota Sandstone.

Bedding planes in the Dakota Sandstone. Macropore conduits are to the left of the rock hammer. Iron oxide skins forming on the surface of the outcrop.

Iron oxide cemented layer of sandstone in the upper part of Pawnee Rock. These samples of the Dakota Sandstone show angular, arkosic sand grains and what appear to be cements of iron oxide and silica. The material appears to be poorly to moderately-cemented, and prone to dissolution at the surface, meaning that this material could breakdown and contribute sand grains to the dune fields located at Camp Aldrich to the east.


In its outcrop area the Wellington is predominantly shale with minor amounts of limestone and dolomite, siltstone, and gypsum and anhydrite (Swineford, 1955). Thick beds of highly soluble salt, the Hutchinson Salt Member, are present in the subsurface. No outcrops of the Wellington Formation or the Hutchinson Salt occur in the study area. In places where the formation is at the surface, the salt is generally not present due to its solubility.

Soil Analysis
Previous reports from the National Resource Conservation Service (1981) and other ESU students of this class (Laird, 2004; Salley, 2004) have provided adequate morphologic descriptions of soils observed around the Cheyenne Bottoms area. This analysis focuses on the interpretation of how the soils present condition and position reflect past and present geomorphologic processes in the area.

Soils map of Cheyenne Bottoms (NRCS, 1981)

During the October field study, soil samples were collected from selected soil profiles and the marshes within Cheyenne Bottoms. These samples were analyzed under 50x and 200x magnification to attempt to determine particulate composition, angularity, presence of organic detritus and basic mineralogic constituents. The soils were tested for the presence of calcium carbonate using a 10% HCL solution. No fizzing indicative of carbon dioxide release was observed with a 10x loupe. The soils are made up of #200 or smaller fines, primarily clay, that are primarily deposited by flow from Blood and Deception Creeks.

Soil samples from the marsh in Cheyenne Bottoms were heated to 215 degrees Celsius for 10 hours in order to out-gas water and to examine the organic and inorganic constituents in the sample. The sample was approximately 90 percent fines (smaller than a #200 sieve), with a slight inclusion of fine sands in the samples. Theoretically, the smaller percentage of silts and sands could be due to removal of the material during dry periods when this material may become airborne and transported elsewhere, while the clay remains saturated longer and compacts more readily (Aber, 2009). While larger particles of plant debris were observed in examination with 10x and 15x loupes, smaller particles were not recognizable at 50x and 250x examinations. The lack of small organic particles is probably due to anoxic conditions in the marsh that retard the microbial breakdown of organic material. A clay content greater than 50% was inferred by using the qualitative 'ribbon' testing to determine cohesion and plasticity.

Blocky to prismatic structure in residuum from road cut near Deception Creek in northwest part of Cheyenne Bottoms. High sodium content (NRCS, 1981) could account for prismatic structure in some soil outcrops. Fragments of "bog iron", iron-rich clays concentrated in marshland deposits and compressed into claystone, are plentiful in the soil outcrop. Some Pliocene-age alluvial cobbles were also observed in the road cut. The light color is probably due to high silica and sodium content in the soil (Ritter, et. al 2002).

Sample of soil from marsh in Cheyenne Bottoms at 50x magnification. Sample is mostly high-plasticity clays, probably from the smectite group, with slight inclusions of sand and organic material. These types of clays are expansive, meaning that they swell to about 30 percent their normal size when wet and form an effective seal to water attempting to move into the subsurface (Miles, 2007). When these clays dry, they contract, forming systems of cracks on the surface that can penetrate a meter or more into the underlying soil.

Sand samples from Camp Aldrich, located in the sand dunes east of Cheyenne Bottoms. The left sample is magnified at 50x, the right at 250x. The samples show the sub-rounded nature of the sand, indicating some mechanical rounding from breakdown during transport, most likely aeolian. Clay skins shown in the right sample may indicate coating and oxidation during wetter climatic periods in the past 6,000 years.

Sample of sand from the Nebraska Sand Hills.
Image used by permission (Elder 2006).

Like the samples from Cheyenne Bottoms, the Sand Hills material contains grains of silica as well as lithics, probably chert, quartzite or other resistant materials. The degree of mechanical rounding is similar, as well as grains exhibiting differential coloring from 'solarization', discoloring from exposure to ultraviolet sunlight, strong chemical reactions, or from exposure to heat at the surface, such as from prairie fires. Both samples of sand are arkosic enough to appear to be derived from the Pliocene alluvium that blankets much of the Great Plains. And the well sorted nature of the material indicates that it could have been sifted and transported by Holocene winds. The sub-angular nature of some of the grains hints at an earlier depositional period when this material was being fluvially transported from source areas in the Rocky Mountains to the west.


Cheyenne Bottoms appears to be a polygenetic feature, with many forces at work through time. From examining the isopach maps made by Bayne (1977), there appears to be a bedrock subsidence beneath Cheyenne Bottoms that was the initial setting of the playa's formation. This topographic low may be the result of localized faulting, perhaps similar to Gentile's (1983) Belton Ring Fault Complex, or a deeper fault system as proposed by Bayne (1977).

The entire High Plains experienced uplift and tilting from the Mio-Pliocene into the Pleistocene (Aber, 2009). Kansas was covered with Pliocene-age sediment to the western border of the Flint Hills. Older landscapes were buried by alluvium, but still impact surface features, as evidenced by the bends seen in modern stream flows that parallel lineaments in Cretaceous-age bedrock (Bayne, 1977).

During the Pleistocene, a wetter climate and possibly a dam of alluvium (Shaw, 1915) formed a lake in Cheyenne Bottoms. According to drilling performed by the Kansas Geological Survey, the basal lacustrine sediments were being deposited approximately 100,000 years old (Aber, 2009). Wave action on the lake could have enlarged the depression in the bedrock, while the weight of the lake compacted the underlying, unconsolidated alluvium.

In the Altithermal of the Holocene (approximately 7500-5000 B.P.), sands that were deposited as alluvium in the Pliocene and Pleistocene, as well as Dakota Sandstone residuum, were mobilized and built up the dune fields to the east of Cheyenne Bottoms. The alluvial dam dried and blew away as the lake disappeared. As the climate became wetter after Altithermal, a thin mantle of vegetation fixed the dunes in place, except for geologically brief periods of reactivation, such as the "Dust Bowl" period of the 1920's.

Many playa features may be the product of both dissolution of carbonates and salts by surface and groundwater, as well as strong desert winds mobilizing vast amounts of sand to further enlarge the area of the feature. It may be that solution-weathered formations in arid environments tend to form large sink structures.

The marsh playa in Cheyenne Bottoms can be viewed as the modern environment. Work performed by Pfaff (2007) demonstrated that the lake that formerly occupied Cheyenne Bottoms was a shrinking body of water with increased salinity reflecting the concentrations of minerals by higher rates of evapotranspiration. Therefore, part of the alluvium beneath Cheyenne Bottoms and the dune field to the east of it can be thought of as the 'relict' landscape of the Pleistocene, with a Holocene marsh imposed on the old lake bed and a mantle of Holocene soil and vegetation development covering the older dune field.

Many playa features may be the product of both dissolution of carbonates and salts by surface and groundwater, as well as strong desert winds mobilizing vast amounts of sand to further enlarge the area of the feature. It may be that solution-weathered formations in arid environments tend to form large sink structures.

Sand field in southwestern Libya, filling what appears to be ancient inland sea with sand-filled paleo-channel. Imaged from Google Earth.

Alluvial fan in northwestern China. Imaged from Google Earth.


Aber, J.S., 2004 Course Webpages for GO 546 : Field Geomorphology

Aber, J.S., 2009, Personal communication about basal lacustrine deposits at Cheyenne Bottoms.

Aber, J.S., 2009, Personal communication about ideas on aeolian sediment deposition during arid climatic episodes.

Aber, J.S. and Pavri, F., 2006a, Cheyenne Bottoms, Kansas, Hyspire- Hyper Spatial Imagery of Rural Environments,


Bayne, C. K.  1977.  Geology and Structure of Cheyenne Bottoms: Barton County, Kansas.  Kansas Geologic Society.  Bulletin 211, Part 2.  p 1-11

Brady, N. C. and Weil, Ray R.  2004.  Elements of the Nature and Properties of Soils, 2nd ed.  Prentice Hall, New Jersey. p 63, p 590

Elder, J., 2006, Sand Micrograph Gallery. URL:

Fetter, C., 2001, Applied Hydrogeology, Fourth Edition; Prentice Hall, New Jersey, pgs. 311-319

Fredlund, G., 1991, Analysis of Quaternary pollen from Cheyenne Bottoms, Kansas: Evidence for late Quaternary vegetation and climates in the Central Great Plains, [Ph.D. thesis]: University of Kansas published by the Kansas Geological Survey, Open-file Report 91-43, 186 p.

Fredlund, G., 1995, Late Quaternary Pollen Record from Cheyenne Bottoms, Kansas, Quaternary Research 43, p. 67-79.

Frye, J.C., Leonard, A.B., Pleistocene Geology of Kansas,


Gentile, Richard J., Geology of the Belton Quadrangle, Missouri Geological Survey Report of Investigations Number 69, pgs. 29-35

Kansas Geological Survey, 2008, Surficial Geology of Kansas: Kansas Geological Survey, Map M-118, scale 500,000

Kruger, J. M., 1997, Gravity and Magnetic Maps of Kansas, Open-file Rept. 96-51


Laird, P., 2004, Geomorphology of the Suburface and Surface for Cheyenne Bottoms


Latta, B. F., 1950, Geology and Groundwater Resources of Barton and Stafford Counties, Kansas, Kansas Geological Survey Bulletin 88, Sections: Geology, Geologic History, Groundwater


Meltzer, D. J., 1999, Human Responses to Middle Holocene (Altithermal) Climates on the North American Great Plains, Quaternary Research 52, 404-416

Miles, R., 2009, The Cooperative Soil Survey, Center for Applied Research and Environmental Systems (CARES), University of Missouri, Columbia


Missouri Geological Survey, 2006, Missouri Environmental Geology Atlas (MEGA)

Moore, R.C. et. al, The Kansas Rock Column, 1951,


Pfaff, L. J., 2006, Late Quaternary Ostracodes as Indicators of Salinity Conditions and Paleohydrology in a Mid-continent Wetland, Cheyenne Bottoms, Kansas, M.S. Thesis: Emporia State University

Ritter, D., Kochel, R., Miller, J., 2002, Process Geomorphology, pgs. 52-63

Soil Survey of Barton County, Kansas.  United Sates Department of Agriculture, Soil Conservation Service.  1981.  p 63


Salley, S.W., 2004, The Geomorphology and Geologic History of Cheyenne Bottoms Central Kansas,


Schmidt, Andy, 2004, Digital Elevation Model for Cheyenne Bottoms, Kansas Geological Survey data

Shaw, E.W., 1915. Newly Discovered Beds of Extinct Lakes in Southern and Western Illinois and Adjacent States. Illinois Geological Survey Bulletin 20, 139-157

Walters, Robert F., 1978, Land Subsidence in Central Kansas Related to Salt Dissolution, Kansas Geological Survey Bulletin 214, Pt. III

J. Xia, R. Miller, D. W. Steeples, and D. Adkins-Heljeson, 1995, M-41E Residual Bouguer Gravity Map of Kansas, the Second-order Regional Trend Removed, scale 1:1,000,000, 1995