BRICE 1966 Rill Gully

Erosion and Deposition in
the Loess-Mantled Great

Plains, Medicine Creek
Drainage Basin, Nebraska
GEOLOGICAL SURVEY PROFESSIONAL PAPER 352-H
Prepared as part of a program of the

Department of the Interior for development
of the Missouri River basin and part of the
soil and moisture program
Erosion and Deposition in
the Loess-Mantled Great
Plains, Medicine Creek
Drainage Basin, Nebraska
By JAMES C. BRICE
EROSION AND SEDIMENTATION IN A SEMIARID ENVIRONMENT
GEOLOGICAL SURVEY PROFESSIONAL PAPER 352-H
Prepared as part of a program of the

Department of the Interior for development
of the Missouri River basin and part of the
soil and moisture program
UNITED STATES GOVERNMENT PRINTING OFFICE, WASHINGTON : 1966

UNITED STATES DEPARTMENT OF THE INTERIOR
STEW ART L. UDALL, Secretary
GEOLOGICAL SURVEY
William T. Pecora, Director
For sale by the Superintendent of Documents, U.S. Government Printing Office

Washington, D.C. 20402 Price 55 cents (paper cover)
CONTENTS
Page -i .'. Page

Abstract ______.___ _ __ _____..____ _____. ,___-_-____ ___ 255 Drainage system ______.L__£ L_____ - _ - 282
Introduction __ __________..___._____- .__.___.___..___-__. ____ -__ 255 Evolution of the drainage system ___._______-_ - 282
Climate, by Cloyd H. Scott _.__._.._._._.____ _....._.. 258 Morphometry ___ ___ _ __ _ _.__ - ~ 284
Drainage transformation and variations in drain-
Vegetation and paleoclimate ___.._______ ___ ___ ___ __ ____-____ 262
age texture _______________. __ 288
Climax plant communities ______________-__ ___-__-____ _ 262
Characteristics of native grasses _____________________________ 262 Gully erosion __._._ __.._.__..________-____ 290
Effects of drought ________.._.._______________________________.... 263 Definition and classification of gullies __________- _- 290
Effects of land use ______________________________________________________ 263 Age and activity of erosional scarps __ _____ 291
Native trees and dendrochronology _ __ _._-__ Channel scarps and valley-bottom gullies .__ ___ 291
264
Pre-Pleistocene rocks ______ _________ _ __________________________ Development of valley-bottom gullies ____________ 295
264
Unexposed rocks _____-__________.________________________________ ______ History of late Recent gullying on Dry Creek ______ 300
264
Exposed rocks - ___ _______________________________________________ Valley-head and valley-side gullies ________ __ 301
264
Niobrara Formation __________________________________________ Measurement of valley-head and valley-side gullies
264
Ogallala Formation ________________________________________ 265 on Dry Greek ____.__ __ 304
Pleistocene and Recent deposits ________________________________ Areal distribution of gullies _.._-__ - 307
265
Terraces -___ ____--________________ __ ____________________________ Formation of valley-head and valley-side gullies ___ 309
265
General stratigraphy of the Pleistocene and Gullying and land use ______-__ __ 313
Recent _ _ ._ ____-__ ____________ _______________ ____ Control of gullies __________ ___ -_-_--- _ __- 314
268
Stratigraphy on Cedar Creek near Stockville ________ 270 Suspended sediment and the hydraulic geometry of
Stratigraphy on Cut Canyon near Curtis ______________ 271 channels, by Cloyd H. Scott __ __ _ 315
Stratigraphy on Elkhorn Canyon near Maywood _ 272 Hydraulic relations at a section and in a down-
Loveland Formation and Sangamon soil ________________ 274 stream direction ___________________ 315
Peorian Loess, Brady soil, Bignell Loess, and Suspended-sediment discharge relations ___ 321
modern soil ______________________________ 275 Particle-size distributions of suspended sediment
Stockville terrace deposits ________________________________ 278 and bed material _._____.__- _ __ . - 322
Mousel terrace deposits ________________________________________ 278 Relative sediment contributions of the major tribu-
Recent alluvium ________________ .________________________ ______ 278 taries _________________ -___ _ _ -_ _ 325
Significance of carbonized fragments of grass ________ 280 Geomorphic properties in relation to water and sedi-
Sand dunes _______________.______________________________________.__ 280 ment discharge _ _____________________ _ ^_ __ _ 328
Archeology and human occupation ________________________ 280 Summary and conclusions ____-___ __ 332
Results of archeologic investigations ________________________ 280 References __-_-____-_____ __ 333
Settlement and land use __________________________________ 281 Index .. _____._ ..__._______ __ 337
ILLUSTRATIONS
Page
FIGURE 175. Regional physiographic map _ ________-._______-______ _ _ - __ __ __ - 256
176. Location map _____________________________ _______________-,,___-__ ____-_ - _ __- 257
177. Photograph of typical relief in southern part of basin _ __________ _ _ - 258
178. Graph of average monthly temperature and maximum, average, and minimum
monthly precipitation at Curtis, Nebr., 1931-55 ____ ________ _ __ 259
179-180. Graph of annual precipitation, 1895-1960
179. Curtis, Nebr. _ ___ _._____ __ _ _________________________________________ 260
180. McCook, Nebr. .. _._._._. __ _. ____.._.______.__ ________.___._ 261
181. Graph of rainfall intensity-duration-frequency curves for Curtis, Nebr., 1951-58 ____ 262
182. Photographs showing effect of land use on native vegetation _______________________ 264
183. Geologic section across lower tip of basin ___________________________________________________._._.._ 266
184. Geologic section across the basin _______________________________________________________________________________ 267
185. Section showing general topographic and stratigraphic relations of Pleistocene and
Recent deposits .__________ ____________ _______________________________________________________________ 268
186. Channel profile, valley flat profile, and terrace profiles along the main course of
Medicine Creek ________ _________________ ___.-________ ___________________.________________________.___.... 270
ill
IV CONTENTS
Page
FIGURES
187-189. Profile and geologic section
187. On Cedar Creek _______________________________________________________ 271
188. Along road entering Cut Canyon _ __ __ ____________ ________ _...__ 272
189. On Elkhorn Canyon ______ __________._ ___ __ ___ __ _ ______________.___ 273
190-193. Photographs showing
190. Terrace sequence and valley fills along Elkhorn Canyon ____ _.____________ 273
191. Valley fill of the Peorian Loess set unconformably against Sangamon soil
and Peorian Loess _ _-___ __ __ _-_ _________._.__-______-____ 274
192. Peorian Loess overlying A horizon of Sangamon soil ___ _ _ ___________ _ 276
193. Banding in Recent alluvium, Dry Creek ____________________________.___ __ 279
194. Block diagram showing evolution of the relief since deposition of the Peorian Loess... - 283
195. Photographs of remnants of the side slopes and valley flats of former drainage
systems ____________________ ____-______ ___________.______________ 284
196. Map showing area of capture of North Plum Creek by Deer Creek _________________. 285
197. Maps showing variations in drainage texture ________________- ____________ __ __ 287
198. Graph showing mean channel length in relation to channel order ____________________ 289
199. Graph showing number of channels of each order in relation to channel order __ __ _ 290
200. Composite sketch showing varieties of valley-bottom gullies ______________________ 292
201. Map of head scarps in major valley-bottom gullies ____________________________________ 293
202. Profiles of valley and channel for some major tributaries __ ____ ___ _____ . 295
203. Graph showing local valley slope in relation to drainage area ___ __ 296
204. Photographs of typical valley-bottom gullies ________________________ ________________ 297
205. Profile showing changes in a large valley-bottom gully on Dry Creek ________ __ __ 298
206. Profiles of a large valley-bottom gully in a tributary of Curtis Creek Canyon ___ . 299
207. Profile showing changes in a gullied tributary of Dry Creek ___ _. _____ __ ____ - 300
208. Profiles of channel, valley flat, and terraces on Dry Creek ________________ ___ 301
209. Cross profiles of Dry Creek ___________ _________________________ __________________________-_ 302
210. Photographs showing downstream changes in channel of Dry Creek _____ __ ______ 303
211. Composite sketch showing varieties of valley-head and valley-side gullies ________ 304
212. Photographs of valley-head and valley-side gullies __ __ _.___________________ ____-__- 305
213. Graph of cumulative frequency distribution of enlargements of valley-head and valley-
side gullies on Dry Creek _ ._ _ ________ _ _ __-___ ___.__ __ _ ___ __ 306
214. Aerial photograph of a severely gullied area on upper Dry Creek ___________________ 308
215. Map showing areal frequency distribution of active valley-head and valley-side
gullies _ _____._-_______ ______ ______________________ __-__-________________ 310
216. Graph showing frequency of active valley-head and valley-side gullies in relation to
area of upland and adjusted frequency of first-order channels _____ _ ______. 311
217. Profiles and plan view of a large valley-head gully ______________________________ _ _ 312
218. Profile of a large valley-head gully _______________________________________________________ 313
219. Photograph of the head scarp of a small valley-head gully on Dry Creek ________________ 314
220. Hydrographs of daily mean flows equaled or exceeded about 1 percent of time _________ 316
221. Graph showing change of width, depth, and velocity with increasing discharge at a
section, Dry Creek near Curtis _- _ -_-__ .__________ ___________ _ ___ __ -__ 317
222-223. Graph showing change of width, depth, and velocity in a downstream direction for
channels of
222. Perennial streams ____-_-_______ -______ ________-__ _____________..._ 319
223. Ephemeral streams _____-____._________________________________^______________ 320
224-225. Graph showing change of suspended-sediment load with increasing discharge at a
section
224. Perennial streams ___ _ ______-________.______________________________________ 323
225. Ephemeral streams ____ _____________________________________________________________ 324
226-231. Graph showing
226. Size distribution of suspended sediment ______________________________________ 325
227. Size distribution of bed material ______ ________ _ __ _ _____________... 326
228. Cumulative water discharge and sediment discharge per square mile _________ 327
229. Mean channel slope in relation to channel order ______________________________ 328
230. Adjusted frequency of first-order channels in relation to mean valley-side
slope, mean slope of first-order channels, and mean length of first-
order channels _-_ _ ________ ___ ___ ____________ ______________ 329
231. Slope of first-order channels in lower Dry Creek subbasin in relation to
length ____ _-___ . ______ ________ _____________________________ 330
232. Trial graph showing multiple linear regression of sediment discharge in relation to
relief ratio, percentage of upland, and adjusted frequency of first-order chan-
nels ______ _______________ __._________________________________________ __ 331
CONTENTS
TABLES
Page
TABLE 1. Thickness and percentage of each rock type in seven exposed sections of the Ogallala
Formation .______...._..___._.._.._......_......._......_ __ _ _____-______-_ 265
2. Correlation of stratigraphic units and terraces of the Pleistocene and Recent in Ne-
braska and Wyoming. ._ .____._ __.........................._ _._._..... _ _ _.__....._.. 269
3. Measured and derived properties for the Medicine Creek basin and its major subbasins...... 288
4. Drainage system properties, listed according to channel order, for major subbasins of
the Medicine Creek basin ._.___..._......_._....._....... ._.._. ._...__- __ .-..__... .____ 294
5. Measurements relating to the size and topographic setting of major valley-bottom gullies ___ 299
6. Number of gullies according to type and size in the Medicine Creek basin and in major
subbasins ___________________-_________-_. .______________ 307
7. Gully frequencies and related data for subbasins of the Medicine Creek basin . _______ 309
8. Values of 6, /, and m for streams in Medicine Creek basin ..._._.._ .._.... ...-.__...._.._. 321
9. Areas in upland and valley flat in comparison with areas in two categories of land use 330
EROSION AND DEPOSITION IN THE LOESS-MANTLED GREAT PLAINS
MEDICINE GREEK DRAINAGE BASIN, NEBRASKA
By JAMES C. BRICE
ABSTRACT change in width. Curves relating suspended-sediment load

The Medicine Creek basin is representative of the loess- to water discharge for the perennial streams are characterized
mantled Great Plains in climate, surface materials, and by a break in slope at water discharges above normal but
relief forms, but its relief, degree of dissection, and rate below the bankfull stage.
of erosion are higher than average. In operation, if not in In general, the concentration of measured suspended sedi-
rate, processes of erosion and deposition in the basin are ment is higher in wet years than in dry, and the ephemeral
considered to be representative of loess-mantled regions in a streams have higher concentrations than the perennial
semiarid or subhumid climate. streams. Differences in runoff and sediment discharge among
About 12,000 years ago, the valley fill of the Peorian five subbasins are attributed mainly to differences in relief
Loess was incised and the drainage system was ramified to ratio, adjusted frequency of first-order channels, and per-
approximately its present extent. During the subsequent centage of area in upland.
episode of deposition (about 1,000-5,000 yr ago), the deposits Active valley-head and valley-side gullies, and all but a
of the Stockville terrace accumulated and the valley sides few of the active valley-bottom gullies, are attributed to
were graded. The formation of the Stockville terrace was land use since settlement rather than to climatic change.
followed by a minor episode of deposition, another of valley Restoration of native vegetation to the heads of valleys,
incision, and finally the accumulation of the late Recent together with conservation measures on the upland, would be
alluvium, which has been intermittently incised during the effective in the control and prevention of valley-head gullies,
past 500 years. which are the most numerous in the basin.
The property of the drainage system that is most useful
in accounting for the areal distribution of gullies is called INTRODUCTION
adjusted channel frequency, in the derivation of which the
drainage basin area is adjusted by subtracting the area of The purpose of this report is to give an integrated
upland. The slope of the exponential curves, relating mean account of the geologic and hydrologic factors that
channel length to channel order, changes at the lower channel
orders because of drainage transformations, in late Recent
are pertinent to erosion and deposition in a part of
time, that affected only channels of low order. the loess-mantled Great Plains and to evaluate the
Gullies in the basin are widened, lengthened, and deepened major factors involved in gully erosion, channel
mainly by scarp erosion. They are classified on the basis of deposition, and the discharge of water and sediment
topographic location as valley-bottom, valley-side, and valley- through channels. Climatic change as a cause of
head gullies. Trenching of a valley reach to bedrock takes modern erosion is evaluated in the light of ecologic,
place not only by coalescence of discontinuous gullies but also
by the successive upstream migration of several channel stratigraphic, and archeologic evidence from terrace
scarps. Many large valley-bottom gullies have evidently deposits of late Pleistocene and Recent age. Proc-
originated on locally steepened valley reaches, but the ratio esses of scarp erosion are analyzed, gullies are
of local slope to drainage area that is critical for the initia- classified, and the areal distribution of gullies is
tion of a gulley is not sharply denned. Valley-bottom gullies correlated with morphologic properties of the basin.
advance mainly because of plunge-pool action, but this
mechanism is less important in the advance of valley-side Generalizations are made as to the hydraulic geome-
and valley-head gullies. The areal frequency distribution of try of the channels and the relations between dis-
valley-head and valley-side gullies is correlated with two charge and suspended-sediment load. Differences in
geomorphic properties: percentage of area in upland and runoff and sediment discharge among five subbasins
adjusted frequency of first-order channels. are attributed mainly to differences in specific mor-
Hydraulic geometry of the channels is expressed by the
slope of curves relating width, depth, and velocity to dis-
phologic properties among the subbasins.
charges equaled or exceeded 1, 2, and 25 percent of the time. The Medicine Creek basin was selected by several
At the gaging stations on Dry, Mitchell, and Brushy Creeks, cooperating agencies (U.S. Bur. of Reclamation, U.S.
the flow is categorized as ephemeral; at the others, as peren- Agr. Research Service, and the Univ. of Nebraska)
nial. In a downstream direction, an increase in discharge
of ephemeral streams is accommodated by relatively large
as a suitable area in which to investigate erosion
changes in depth and velocity, but the increase in discharge and runoff. Collection of data, which was begun in
of perennial streams is accommodated by a relatively large 1951 and ended in 1958, was directed toward an
255
256 EROSION AND SEDIMENTATION IN A SEMIARID ENVIRONMENT
evaluation of soil and water conservation practices of all types and sizes occur, and the frequency of
and also toward an understanding of the processes gullies ranges from high in some places to low in
and causes of rapid erosion in this region, which others.
is the major subject of the present report. Medicine Creek, which is a tributary to the
The regional setting of the Medicine Creek basin, Republican River, has a drainage area of about
which is in the loess-mantled part of the Great 690 square miles that lies between the Platte and
Plains, is shown in figure 175. The processes of the Republican Rivers. Major subbasins and gen-
erosion and deposition in the basin are characteristic eral geographical features are shown in figure 176.
of loess-mantled regions in a semiarid or subhumid Remnants of the upland surface reach a maximum
climate, and they apply specifically to an area of altitude of about 3,150 feet in the northern part
about 115,000 square miles that is within the Mis- of the basin and descend gradually to an altitude of
souri River basin and is mantled with 8 feet or about 2,400 feet at the bluffs of the Republican
more of loess. Most of this area is within the Great River. The maximum local relief, about 200 feet,
Plains, although part is in the Interior Plains. The is in the northern part of the basin. In general,
Medicine Creek basin is dissected to a much greater the upland is more sharply dissected in the northern
degree than many parts of the Great Plains, and part, and the divides tend to be flat but narrow;
the local relief is greater than average. Neverthe- in the central and southern parts, the divides are
less, nearly all the relief forms typical of the Great broader and slope gently toward the valleys. The
Plains occur within the basin, including areas of general aspect of the relief in the southern part of
nearly flat upland and sharply dissected areas of the basin is shown in figure 177, which is an oblique
narrow divides and flat-bottomed valleys. Gullies aerial view southwest from a point above Medicine
EXPLANATION
Sand dunes Mantled by 8-16 feet Mantled by 16-64 feet Mantled by more than Medicine Creek basin
of loess of loess 64 feet of loess
FIGURE 175. Map showing the regional physiographic setting of Medicine Creek basin. Distribution and thickness of loess generalized from Thorpe
and others (1952).
LOESS-MANTLED GREAT PLAINS, MEDICINE CREEK BASIN, NEBRASKA 257
L__
] i^
L=rCÎ\«^-
\Area. of this report
R. 31 W
40°30 r
EXPLANATION
Boundary of report area
Boundary of subbasin
Stream-gaging station
FIGURE 176. Location map of Medicine Creek basin.

CLIMATE
By CLOYD H. SCOTT
The climate of the Medicine Creek basin is typical
of the interior of continents of the middle latitudes
with rather light rainfall, hot summers, cold winters,
large variations in precipitation and temperature
from year to year, and frequent changes in weather
from day to day.
The mean annual temperature at Curtis, near the
center of the basin, is about 52°F. Maximum and
minimum temperatures above 100° and below 0°F
occur occasionally. The coldest month is January
when the average temperature is about 27° F, and
the warmest month is July when the average tem-
FIGURE 177. Oblique aerial photograph of typical relief in southern part of perature is about 78°. The monthly mean tempera-
Medicine Creek basin. Medicine Creek at lower left. Photograph by River ture at the U.S. Weather Bureau Station at Curtis
Basin Surveys, Smithsonian Institution, 1948.
for the years 1931-55 is shown in figure 178.
Creek and about 7 miles north of the town of Much of the precipitation occurs during April
Cambridge. through September, generally as thunderstorms. On
Fieldwork for geomorphic purposes was begun in the average, more than half the total annual pre-
the summer of 1953 and continued during the sum- cipitation occurs from May through August. Pre-
mers of 1954-56. Special hand-level and stadia-rod cipitation varies widely from the average (fig. 178);
equipment was devised to permit the measurement 7 months had no precipitation at least once during
of slopes without an assistant. Field studies, as well the period 1931-55, and some months had no pre-
as office work, were facilitated by excellent topo- cipitation several times during that period. Maxi-
graphic map and aerial photographic coverage of mum amounts exceeded the averages by about two
the entire basin. On the 1 :24,000-scale topographic to seven times.
maps, prepared by photogrammetric methods and The annual precipitation also varies widely from
published by the jGeological Survey in 1957-58, even year to year (fig. 179). The minimum annual
minor topographic features are accurately repre- amount of slightly less than 11 inches was recorded
sented. The Munsell color system was used for field in 1934, and the maximum of about 38.2 inches was
description of loess and soil. recorded in 1915. The 66-year average (1895-60)
Data contributed by the cooperating agencies have was about 21.5 inches, but the 25-year average
been used in the preparation of this report. The (1931-55) was only 19.1 inches.
detailed surveys necessary to measure rates of The 5-year moving averages (figs. 179 and 180)
channel erosion on Dry Creek were made by the show the general trends of precipitation at Curtis
Bureau of Reclamation, who also contributed data and McCook from 1895 to 1960. The annual totals
on precipitation. Land use was tabulated by the are highly variable from one year to another, but the
Agricultural Research Service and the Soil Conser- 5-year moving averages show a somewhat cyclic
vation Service. Runoff and suspended sediment were trend. Even though the high- and low-precipitation
measured by the Geological Survey. years tend to be grouped somewhat, about half of
Many persons contributed to this report, which the years have above-average precipitation amounts.
was prepared under the successive supervision of However, during 1951-58 at Curtis, 5 years had
P. C. Benedict, regional engineer, and D. M. Cul- precipitation below average; 4 of the years had
bertson, district engineer, U.S. Geological Survey. more than 6 inches below average.
The section on native vegetation was read by E. J. The 5-year moving average for Curtis indicates
Dyksterhuis, range conservationist for the Soil Con- a long-term trend toward decreasing annual pre-
servation Service. Fossil snails were identified and cipitation. The consistency of the precipitation data
information on snail ecology was given by W. J. at Curtis was checked by plotting a double-mass
Wayne of the Indiana Geological Survey. One car- curve of data at Curtis and at McCook. A pattern
bon-14 age determination was made by Meyer Rubin, composed of data for McCook, Cambridge, and
geochemist of the U.S. Geological Survey, and an- North Platte was first used; but the data for North
other by J. L. Kulp of Columbia University. Platte and Cambridge were inconsistent with the
100
90
Average monthly
I 80 temperature
LU
CC.
I
<
^ 70
CO
LU
LU
o:
O
g 60
Annual average
temperature
LU
o: 50
40
30
20 L
FIGURE 178. Average monthly temperature and maximum, average, and minimum monthly precipitation at Curtis, Nebr., 1931-55
pattern, whereas the data for McCook seemed to be crease in annual precipitation at Curtis is probably
consistent. The double-mass curve showed a trend a result of uneven areal distribution and is not
toward more precipitation at Curtis beginning about considered to be representative of the region. As
1908 but returned to about the original slope in shown by the double-mass curve, the data at Curtis
1914. There is no record that the gage was moved may have been somewhat biased for the years
in 1908, although the location was changed in 1914 1908-14.
(U.S. Weather Bureau, 1955a). Precipitation at Curtis averaged about 18.77
The pattern of the 5-year moving average for inches for the years 1951-58, or somewhat below
Curtis (fig. 179) is generally similar to the pattern the 66-year average of about 21.5 inches. Extremes
for McCook (fig. 180) except that the precipitation of annual precipitation for this period ranged from
for the period 1908-18 was higher at Curtis than at 31.61 inches in 1951 to 12.36 inches in 1952. The
McCook and for the period 1940-52 was lower at occurrence of a high and a low in consecutive years
Curtis than at McCook. The apparent secular de- is not unusual owing to the large variations in rain-
40
38
36
34
32
30
28
5-year moving average (plotted at

£ 26 middle of 5-year period)
z
g /66-year average
5,4
22
20
18
16
14
12
10
1895 1900 05 10 15 20 25 30 35 40 45 50 55 1960
FIGURE 179. Annual precipitation at Curtis, Nebr., 1895-1960.
fall over much of the central part of the continent. of rainfall to maximum 60-minute values. An in-
At Curtis, only 2 of 6 years since 1895 that had in spection of the 6- and 24-hour amounts indicated
excess of 30 inches of precipitation, were followed that conversion for these periods was unnecessary.
by years that had more than the long-term average The plot of rainfall intensity versus return period,
amount. An extreme occurred in 1915-16 when the which was used to define the curves of figure 181,
difference between precipitation amounts exceeded showed little scatter from the average curve, which
20 inches. indicated that no extreme storms had occurred dur-
Although averages and extremes give some indi- ing the period.
cation of precipitation, a more meaningful descrip- The U.S. Weather Bureau (1955b) gives rainfall
tion of rainfall can be obtained through the use of intensity-duration-frequency curves for about 200
rainfall intensity-duration-frequency curves. Such stations, including North Platte, for the years
curves (fig. 181) were developed for Curtis for 1906-51. Comparison of the curves for Curtis with
return periods of 2 and 5 years based on the annual those for North Platte shows that differences are
series of 1-, 6-, and 24-hour amounts. The California minor between the two stations for both the 2- and
method of plotting positions, Tr = n/m, was used; 5-year return periods, even though only 8 years of
a log normal distribution was assumed. Tr is the record was used to define the curves for Curtis.
return period in years of item having order number The similarity between the short- and long-term
m in, a decreasing series, and n is the period of curves and the fact that no extreme amounts of
record in years. precipitation fell during the short period indicate
An empirical factor of 1.13 (U.S. Weather Bureau, that the rainfall regimen was well represented at
1957) was used to convert the clock-hour amounts Curtis during the years 1951-58, even though the
40
38
36
32
30 5-year moving average

(plotted at middle
of 5-year average)
co 28
LJ
I
o
^ 26
66-year average.
22
20
18
10
1895 1900 05 10 15 20 25 30 35 40 45 50 1960
FIGURE 180. Annual precipitation at McCook, Nebr., 1895-1960.
average precipitation for the period was somewhat North Platte is in the valley of the Platte River,
lower than the long-term average. the wind speeds and directions may not be entirely
Evaporation is also variable from year to year, representative of conditions in the Medicine Creek
although the variations probably are not as extreme basin. The average wind speeds at North Platte
as those of rainfall. The only evaporation data range from about 12 to 13 miles per hour March
available are for a few years at Medicine Creek through July and from a little less than 10 to
Dam, and these data were used to check evaporation slightly more than 11 miles per hour for the re-
maps in a technical paper by the U.S. Weather mainder of the year. December and January have
Bureau (1959a). According to the maps the average the lowest average speeds with 9.8 and 9.7 miles
annual evaporation from a class A pan in Medicine per hour, respectively, and April has the highest
Creek basin is about 75 inches, or about S1/^ times average with 13.1 miles per hour. The prevailing'
the average annual precipitation; and lake evapora- direction is from the southeast April through Au-
tion is about 50 inches, or nearly 21/2 times the aver- gust; south-southeast in September and January;
age annual precipitation. Evaporation increases northwest in February, October, and November;'
from north to south in the basin, but the average north in March; and west-northwest in December.
difference is only about 3 to 4 inches per year. The annual prevailing direction at North Platte is
Data on wind speed and direction nearest Medi- considered to be from the southeast (U.S. Weather
cine Creek basin are those at North Platte. Because Bureau, 1959b).
the notes of Deputy Surveyor D. V. Stephenson,

made in the Medicine Creek basin during the summer
of 1872: "land broken by steep ravines running in
almost every direction, in bottoms of some of which
there is some blue grass. The upland is about three-
fourths covered with short buffalo grass. * * * grass
almost entirely short buffalo, growing in bunches."
Deputy Surveyor J. L. Slocum made the following
notes, also in the summer of 1872: "Soil is dry and
sandy, produces only short buffalo grass in scatter-
ing bunches unfit for cultivation or grazing. * * *
Considerable good grass in bottom of ravines, upland
is sparsely covered with short buffalo grass." Dep-
uty Surveyor Charles Wimpf wrote as follows, in
August of 1869: "Bottoms of ravines filled with a
thick very good grass which gives an excellent hay.
The hills are covered with a short sweet grass."
Although the term "buffalograss" may have been
used loosely by the early surveyors, their comments
do indicate that during 1869-72 the uplands and
valley-side slopes were covered with a rather sparse
growth of short grass, and the valley bottoms with
a thicker growth of tall grass.
E. J. Dyksterhuis( writen commun. 1958), range
conservationist for the Soil Conservation Service,
considers that the climax community on the uplands
in regions of climate and soils like those of the Medi-
cine Creek basin contains a dominance of midgrasses
with an understory of short grasses. He also says
that the dominant midgrasses include western wheat-
grass, needlegrasses, and little bluestem; little blue-
stem, side-oats grama, and blue grama dominate on
0.01 the valley sides, and big bluestem and switchgrass on
RAINFALL DURATION, IN HOURS the valley bottoms.
FIGURE 181. Rainfall intensity-duration-frequency curves for Curtis, Nebr., CHARACTERISTICS OF NATIVE GRASSES
1951-58.
Characteristics of native grasses are summarized
VEGETATION AND PALEOCLIMATE in this section from Weaver and Albertson (1943,
CLIMAX PLANT COMMUNITIES 1944), U.S. Department of Agriculture (1948), and
The Medicine Creek basin lies in a transitional Phillips Petroleum Co. (1956).
zone between the tall-grass region to the east (prairie Big bluestem is a coarse perennial native bunch-
grassland) and the short-grass region to the west grass that reaches a height of 6 feet under favorable
(plains grassland). Trees are restricted to valley conditions. The root system is extensive; it may
bottoms, except where they have been artificially penetrate to a depth of about 8 feet and form a
planted. dense mat to a depth of 1 foot. Under favorable
Some information regarding the history of plant conditions it, together with other species of its
communities appears in the notes of surveyors for community, will form a continuous sod; under less
the General Land Office who made brief reference favorable conditions it forms separate, scattered
to the grasses they saw while laying out the land- bunches. Both rainfall and dust are largely inter-
division grid in 1869-72.1 Their comments also cepted by the foliage before reaching the ground.
indicate the condition of the vegetal cover before According to Clark (1937), a prairie covered by big
white settlement. The following excerpts are from bluestem may intercept as much as 53 tons of water
per acre during a rainfall of 1 inch per hour,
1 Field notes of survey are on file at the Dept. of Educational Lands and
Funds, State Capitol Bldg., Lincoln, Nebr. whereas a prairie covered by buffalograss will inter-
cept about 28 tons per acre. Such intercepted rain- area; and in the most severely affected localities,
fall commonly evaporates from the leaves of the the grass was so decimated that nearly all the ground
plant without reaching the ground. was bare. These bare patches, described by Weaver
Buffalograss is native, sod forming, fine leaved, and Albertson and noted by the writer (Brice, 1958)
and perennial and typically grows to a height of throughout the Medicine Creek basin in 1953-56, are
4 to 6 inches. The roots of well-developed buffalo- important in slope erosion because erosion scarps
grass extend to a depth of 4 to 5 feet and form a commonly form at their edges. ,
very thick mat to a depth of 1 foot. Blue grama is The influence of ancient droughts and ancient
a low native perennial grass whose flowering stems episodes of aridity on the grasses of the Medicine
reach a height of about 18 inches. It is the most Creek basin can be inferred from Weaver and
drought resistant of all the native prairie grasses, Albertson's study. Short-term droughts would have
and it spreads, even during drought, mainly by means reduced the cover of climax grasses, but erosion
of seedlings. The root system is similar to that of would not have been severe because these grasses
buffalograss. are replaced by weeds during drought and recover
All these native grasses may form a dense thick, rapidly after drought. Sheet-wash and raindrop
continuous sod, although the bluestems and blue impact would have removed soil from the irregular
grama are more likely to form separate bunches than patches of bare ground between sod-covered areas,
buffalograss. The taller grasses intercept a greater and low sod scarps would have formed on the slopes.
amount of rainfall, which is usually lost to the soil; During ancient episodes of aridity that lasted for
but a corresponding decrease in erosion by raindrop hundreds or thousands of years, a climax community
impact is effected. Windblown dust is retained more of grasses adapted to prevailing climate and rainfall
effectively by the taller grasses. distribution would have become established. Such
a community would probably have consisted of an
EFFECTS OF DROUGHT association of short grasses similar to the association
Some insight into the influence of ancient arid now in eastern Colorado, where the average annual
climatic regimes on native grasses and on the precipitation is about 15 inches. If this short-grass
ability of these grasses to prevent erosion may be association were significantly different from that
gained by study of the effects of the drought of established before the arid episode, stream profiles
1933-40 in Kansas and Nebraska. Such study was and slopes would have been regarded by gullying and
made by Weaver and Albertson (1943, 1944). These other erosional processes.
authors took for a base station Hays, Kans., which
has an average annual precipitation of about 22.9 EFFECTS OF LAND USE
inches and which is located about 120 miles south The pronounced effect of land use is strikingly
of the Medicine Creek basin. Total precipitation shown by the contrast in vegetation on either side
at Hays for the 6 years preceding 1933 was below of a fence that follows the north-south section line
normal. Precipitation during each of the 4 driest between sections 15 and 16, T. 9 N., R. 27 W.
years was about 16 inches. ("See-fig. 182, upper.) The heads of two tributaries
According to Weaver and Alberston, decrease in to" East Fork of Dry Creek are isolated by this fence,
grass cover as a result of drought varied with the and the aspect of the vegetation in these tributary
intensity of grazing and of burial by dust and also heads is unlike that observed at any other place in
with the type of grass cover, as grasses with shorter the Medicine Creek basin. Although no information
roots underwent greater destruction. The basal on the land-use history of the tributary heads was
cover of short grasses in Kansas ranged between obtained, the contrast in vegetal cover at the fence
80 and 95 percent in 1932, and this cover was com- is evident on both the 1937 and the 1952 aerial
monly reduced to 20 percent or less during the worst photographs. Therefore, the fence had divided areas
years of the drought. In some localities, particularly of contrasting land use for at least 15 years, and
those that had been overgrazed or covered by blown the aspect of the vegetation in the tributary heads
dust, the native plant population was reduced almost in 1953 suggested that it may have been disturbed
to zero. Where the effects of drought were moderate, in no important way since white settlement. Besides
the sod was interrupted by patches of bare ground, the much greater thickness of the sod and the domi-
a square foot or less in area, that formed an irregular nance of tall grasses in the tributary heads, the
but continuous network. Where the effects of thickness of brush such as wild plum, sumac, buck-
drought were more severe, the patches of bare brush, and wild rose is exceptional (fig. 182,
ground were larger, up to several square yards in middle). A few ash trees were observed on the
264 , EROSION AND SEDIMENTATION IN A SEMIARID ENVIRONMENT
valley bottoms. The termination of the small valley-

bottom gullies (fig. 182,,,upper) before reaching the
fence may be fortuitous, but no gullies were observed
in the tributary heads. By contrast, gullies are in
most tributary heads in this part of. Dry Creek (fig.
182, lower). Step scarps occur on the steep slopes
of the heavily vegetated area, but they are much
subdued and seem to be inactive. The upland draineci,
by the tributary heads is planted in row crqps orr
small grain, as elsewhere in the basin. t .,
NATIVE TREES AND DENDROCHRONOLOGY '
During the period 1869-72 General Land Office,
surveyors reported that trees in the Medicine Creek,
basin were confined to stream valleys, in which they
identified cottonwood, ash, boxelder, elm, hackberry,
and cedar. These same kinds of trees now grow
from place to place in the valleys, but the wider
valley flats have been mostly cleared of trees.
A tree-ring indication of precipitation during the.
past 400 years in western Nebraska, including the
vicinity of North Platte, was reported by Weakly
(1940, 1943). According to his interpretation, the
climate has been characterized by frequent dry years
and by less frequent droughts that lasted 5 years
or more. The 13 droughts (period of five or more
successive dry years) had an average duration of
13 years, and the average interval between droughts
was about 20 years. The longest drought lasted
from 1539 to 1564 (26 yr), but other lengthy'
droughts were from 1587 to 1605 (19 yr) and from
1688 to 1707 (20 yr).
PRE-PLEISTOCENE ROCKS
UNEXPOSED ROCKS
Neither the lithology nor the structure of unex-
posed rocks in the Medicine Creek basin has any
apparent direct influence on the present relief. The
Paleozoic and Mesozoic formations that overlie thei
Precambrian basement dip gently to the west and
are overlain unconformably by Tertiary formations
that dip gently to the east.
EXPOSED ROCKS
NIOBRARA FORMATION
The Niobrara Formation of Late Cretaceous age
FIGURE 182. Photographs showing effect of land use on riative vegetation, crops out along the banks of Harry Strunk Lake and
East "Fort Dry Creek, August '1953. Upper, Areas separated by a fence
and contrasting sharply in vegetal cover. Denser vegetation is at right of. from place to place along Medicine Creek down-
fence. Although sparse, the vegetation at left of fence was better than stream from the lake, but it is not exposed elsewhere
average for the Medicine i Creek basin in 1953. Middle, Dense cover of
grass and brush in ungrazed valley head. View looking west. Vegetation in the basin. Where exposed, the Niobrara consists
more dense on south side of valley. Lower, Sparse cover of short grass,
yucca, and Weeds in heavily grazed 5 valley head, 1.5 miles southeast of
mainly of orange or white chalk interbedded with
valley head shown, in middle. ,Valley is oriented in north-south direction. thin layers of altered volcanic ash. ,
O GALLALA FORMATION though Wellfleet is not built on the terrace surface,
Outcrops of the Ogallala Formation of Tertiary conspicuous flat-topped remnants of the terrace
(Pliocene) age are mainly restricted to valley sides stand about 125 feet above the valley flat at intervals
and channels of larger streams in the Medicine of a mile or more between Wellfleet and Maywood,
Creek basin, and the best exposures are along Medi- Nebr. For example, the remnant in the SE*4 sec. 8,
cine Creek and Cedar Creek in the vicinity of Stock- T. 8 N., R. 29 W., about 1 mile north of Maywood,
vine. No outcrops were observed in the northern is well defined both in the field and on the Curtis
parts of the basin where the loess cover is thickest. NW quadrangle sheet. There are similar flat-topped
Available well logs indicate that the Ogallala under- remnants along Well Canyon and other major tribu-
lies most of the basin and that its thickness increases taries, but the originally flat surface of the terrace
from about 200 feet in the southern part to about has been dissected in most places.
400 feet in the northern part. (See figs. 183 and 184.) The best preserved and most continuous terrace
Exposed sections of Ogallala at seven localities in the Medicine Creek basin is named the Stockville
were measured and described in detail. Thicknesses terrace after the town of Stockville, Nebr., which
of described sections ranged from 6 feet at one lo- is built on a remnant of the terrace, along the west
cality on Cedar Creek to 73 feet at another locality side of Medicine Creek. The Stockville terrace is
downstream, and the total thickness of described represented along Medicine Creek by broad rem-
sections was 290 feet. In general, the Ogallala con- nants that stand above the valley flat at a height
sists of clay, silt, volcanic ash, sand, and gravel, of about 50 feet in the lower course of the creek
poorly sorted into different beds that are cemented and about 30 feet in its upper course. The Stockville
with (and partly replaced by) different amounts terrace can be traced almost continuously along
of carbonate. Five rock types were distinguished, most tributary channels in the basin. Moreover, the
and the total thickness of each type as represented sides and heads of valleys are broadly graded to the
in the seven measured sections is given in table 1. level of the Stockville terrace.
The lowest terrace of general importance in the
TABLE 1. Thickness and percentage of each rock type in Medicine Creek basin is named the Mousel terrace
seven exposed sections of the Ogallala Formation after the Mousel School, which is in sec. 25, T. 5 N.,
R. 26 W., on the east side of the valley of Medicine
Thick- Percent
ness of total Creek.' (See fig. 176.) The Mousel terrace is repre-
(ft) thickness
sented'along Medicine Creek by widely scattered
22 7.5 remnants that stand 15 to 20 feet above the valley
Sand, pebbly, and silt; loosely cemented, locally flat. Similar remnants are along most major tribu-
156 54.0
Limestone, sandy, rather uniformly cemented . _ _ 62 21.4 taries, such as Well Canyon, Cedar Creek, and Fox
26 9.0 Creek; but the Mousel terrace cannot be traced along
Clay; alternating beds of fine-grained limestone _._. __ 24 8.1
most minor tributaries because its surface has been
buried by late Recent alluvium. Even where the
terrace; is well defined, there has been little grading
With regard to the origin of the Ogallala, Frye of the ivalley sides to the level of the terrace; the
and others (1956) have presented convincing eyi- slope that forms the riser of the Stockville terrace
dence that it was deposited by streams flowing east- rises abruptly above the tread of the Mousel terrace.
ward from the Rocky Mountain region. In the
earlier part of their history these streams occupied A sequence of well-defined terraces appears along
broad, relatively shallow valleys eroded into Cre- most streams in Nebraska, but general agreement
taceous bedrock. However, as alluviation proceeded, has not been reached as to terrace nomenclature.
the deposits progressively overlapped the gentle Condra and others (1950) describe a sequence of
valley sides, most divides were buried, and eventually six terraces in central Nebraska; and Schultz and
a coalescent alluvial plain was formed. others (1951) describe a sequence of five terraces
that applies to Nebraska generally. Neither sequence
PLEISTOCENE AND RECENT DEPOSITS would be expected to apply to all Nebraska streams
TERRACES because of complications, such as overlap of an older
Three terraces were identified along Medicine terrace deposit by a younger terrace deposit, that
Creek and its tributaries. The highest of these is may be peculiar to a single stream.
named the Wellfleet terrace after the village of Schultz and others (1948, 1951) have specifically
Wellfleet, Nebr., which is on Medicine Creek. Al- applied their terrace sequence to alluvial terraces
Z
o
H
CL
X AdVNd3_LVnO
u
UB!OUHII
Pleistocene
Illinoian Wisconsin
.
fi S 5.
Kg
*£&
«?§
QUATERNARY m
X
D
>
Z
>
H
o
z
Creiaceoits
1
CRETA-
CEOUS
Indianola, Nebr
Republican River
2,93 'Nisva naaao aMioiaan 's^[Iv^J xvano aaTLNvw-ssaoi
268
in the southern part of the Medicine Creek basin, present in the Medicine Creek basin, but the physio-
in particular along Lime Creek. According to their graphic expression has been obliterated by deposi-
nomenclature, alluvial terraces in Nebraska are num- tion of the upper part of the Peorian Loess and by
bered upward from the youngest and lowest, which grading of the valley sides during the time that the
is called Terrace-0, to the highest and oldest, which Stockville terrace deposits were accumulating.
is called Terrace-5. Although their numerical sys-
GENERAL STRATIGRAPHY OF THE
tem of terrace nomenclature is not used, the terrace PLEISTOCENE AND RECENT
sequence in this report is based on, and in general
agreement with, the work of Schultz and others. Stratigraphy of the Pleistocene deposits in Ne-
Correlation of terrace deposits with Pleistocene time braska is well established, and satisfactory correla-
units is uncertain, particularly in view of the present tions have been made with the deposits of Kansas
controversy about the units. A tentative correlation and other adjoining States, although nomenclature
of the Medicine Creek terrace sequence with other differs somewhat from State to State. For regional
terrace sequences and with the Pleistocene time descriptions of the stratigraphic units the reader is
units is given in table 2. referred to publications of the Nebraska Geological
The Stockville terrace of this report corresponds Survey (Condra and others, 1950; Lugn, 1935) and
physiographically to Terrace-2 of Schultz and others of the Kansas Geological Survey (Frye and Leonard,
(1948, 1951), and the Stockville terrace deposits 1952). Leonard (1950, 1952) has established the
correspond to fill A of Terrace-2. No terrace de- ranges of land snails that are generally abundant
posits corresponding to fill B of Terrace-2 were in the Pleistocene deposits of Kansas.
observed in the Medicine Creek basin by the writer. The general relations of Pleistocene and Recent
The Wellfleet terrace, which is underlain by the stratigraphic units in the Medicine Creek basin are
upper part of the valley fill of the Peorian Loess, shown in figure 185, and in table 2 correlation is
corresponds physiographically with the Terrace-3 tentatively made with formally named units in Ne-
of Schultz and others and with the Terrace no. 3 braska and Wyoming. Long profiles of terrace de-
of Condra and others (1950, p. 35). No physio- posits along the course of Medicine Creek are shown
graphic expression of the Terrace-4 of Schultz and in figure 186. Correlation of the upper Wisconsin
others nor of the Terrace no. 4 of Condra and others and Recent units is based on carbon-14 age deter-
was recognized in the Medicine Creek basin. At a minations and on physiographic expression as ter-
locality on Medicine Creek (sec. 11, T. 5 N., R. 26 races. Correlation of the older units is based on
W.), Schultz and others (1948, p. 36) have identi- fossil evidence, on soil markers such as the Sangamon
fied a surface as the tread of Terrace-3. This surface soil, and on lithology. Pleistocene and Recent de-
is interpreted by the writer as a long valley-side posits in the Medicine Creek basin consist of dune
slope graded to the level of the Stockville terrace. sand, loess, and alluvium, overlying a pre-Pleistocene
At this same locality the surface identified as Ter- (or lower Pleistocene) surface cut into the Ogallala
race-4 by Schultz and others is interpreted as the Formation. The deposits of Nebraskan or Kansan
tread of Terrace-3, or the Wellfleet terrace. A ter- age are mainly of alluvial silt and sand; gravel, in
race equivalent to Terrace-4 was doubtless once lesser amounts, is confined to the bottoms of former
! 25
Q
50 -
75 -
Sappa Formation
100 -
jpgallala Formation
Mouse! terrace .
deposits
VERTICAL EXAGGERATION X8
FIGURE 185. General topographic and stratigraphic relations of Pleistocene and Recent deposits in the Medicine Creek Basin. Differences in strati-
graphy and relief on either side of the figure represent the range of differences among valleys in the basin.
TABLE 2. Correlation of stratiyraphic units and terraces of the Pleistocene and Recent in Nebraska and Wyoming
Generalized alluyial sequence
Central Nebraska Nebraska Wyoming Medicine Creek, Nebraska
Age western United States
(Condra and others, 1950) (Schultz and others, 1951) (Leopold and Miller, 1954) (This report) (Miller, 1958)
WWWWWWWVWWWW -i wvwvwwwwwvw -i WWWVWWVWWWWWVWVWWV -i - TJ wwvvvwwwvwwwwwwvvwwvvvw
Wisconsin "VI" silt \ , . Recent valley
Soil Z \Terrace-0 Lightning fm lightning Recent alluvium \ fl te d Deposition 3
Late Cochrane silt \ \ terrace (420±160)^ , \ terrace Began A.D. 1200-*600^ ended 1880
wvwwwwwwwww -\ >- wwwwwwwww _-\ wwwwwvvvvvvwvwvvvwvww i L vwwwwvwwvww^vwvwvwvvwww r
Early Cochrane silt \ Moorcroft-deposition Moor- Deposition 2 o
Mousel terrace deposits \ Mousel
S0) (2,140±150^ 1 or stability \ crof t Upper part-^no younger
wfwwww* -, terrace (U43±240)^ \ terrace than A.D. UOO-1200^
Kaycee fm I \ (2,200±200)^ I
Lower part dates 2,200-2,400^
Soil Y I Terrace-1
Altithermal ~\
wwwwwwww \ wwvvvwwvwvvwwwwwvwv -\ wwwvwwwvvwwvvvyvvwvwwvwww
Paleosol \ Soil development \
\ I Kaycee
\ Late Mankato silt \ TT ,. \ terrace \ Stockville Deposition 1
-?- ? ? WWWVWWVWWWWV -i *- Ucross fm I \ terrace (Probably correlates with a
Mankato Stockville terrace deposits \
\ Terrace late Wisconsin glacial sub-
Early Mankato silt \ stage. Available dates O
Bignellsilt \ na2
suggest an «ge of 7,200-
WWWWWVWWWVW i *- vvvwwwwwwwwvwwwvww -\ 7,800^)
Soil YY \ Terrace-2
Brady soil \ (9,880+670)^ \ fill A Brady soil \
v \ Wellfleet
Wiscon- Late Gary silt \ \ terrace
_ . .,1 Terrace
sin Gary Upper Peorian silt I no 3 WWWWIAWWWVWWV -i \r
JXJ
Early Gary silt P '2
Soil development Peorian Loess \
Brady soil \ K
Late Tazewell silt \
Tazewell Middle Peorian silt \ WWWWWVWWWWWW -i L O
Early Tazewell silt I
V
C Soil W \Terrace-3
V Soil development \
lowati Lower Peorian silt \ no. 4 Todd Valley fm \ Peorian Loess o
01 wwwvwwwwwwv . L
1
£ Todd Valley sand \ Early lowan silt I
Sangamon
- vwwwwvwvwvwwww L
Loveland soil Sangamon soil \ Sangamon soil
Loveland fm Loveland fm Terrace-4
w
Illinoian Loveland Formation CO
Crete fm Crete fm Ii
Yarmouth soil \ 2
Yarmouth Sappa soil Yarmouth soil
Sappa f m \
Upper Sappa fm
Pearlette ash lenticle Grand Island fm \ W
Kansan Sappa Formation
Lower Sappa fm wwwwwwwwvwwvww L
Grand Island fm Red Cloud fm
Aftonian Soil development Fullerton f m
- Fullerton fm
Nebraskan Holdrege fm
Holdrege fm "' : , . r,<
Years B.P., C dates reported by Schultz and others (1951) MWVWW Denotes major periods of valley incision
Years B.P., C 14 dates given in this report
Dates from pottery and dendrochronology Encloses the units underlying a physiographically
Years B.P., C 14 dates from various localities expressed terrace. Wind-laid silt deposited
Years B.P., C 14 date reported by Wedel and Kivett (.1956) \ subsequently on terrace surface is not shown
to
05
2800
2700
Z 2600
2500
2400
2300
2200
70 60 50 40 30 20 10
DISTANCE FROM CONFLUENCE, IN MILES
FIGURE 186. Channel profile, valley flat profile, and terrace profiles along the main course of Medicine Creek.
valleys (figs. 183 and 184). Although loess may Nebraska Conservation and Survey Division in co-
have been deposited during either Kansan or Ne- operation with the U.S. Geological Survey. Figure
braskan time, no loess of Kansan or Nebraskan age 183 is simplified from an unpublished section by
has been identified with certainty either on Medicine J. L. Deffenbaugh, based on his interpretation of
Creek or elsewhere in Nebraska. At two localities logs of test holes drilled in 1948. On his original
in the Medicine Creek basin, a paleosol was identi- section, Deffenbaugh distinguishes the Loveland,
fied as the Yarmouth soil on the basis of snail fauna Sappa, and Grand Island Formations. Figure 184
and high content of volcanic ash. Silts beneath the represents an interpretation, made by the writer,
Yarmouth soil are correlated with the Sappa For- of logs of test holes drilled in 1960. Although the
mation. No deposits of ash of the Pearlette Ash logs include full descriptions of the rocks penetrated,
Member of the Sappa Formation were seen in the the writer was not able to distinguish with confi-
Medicine Creek basin, but deposits up to 30 feet dence any formational boundaries beneath the San-
in thickness crop out in areas adjacent to the basin. gamon soil and above the Ogallala Formation. To
No Pleistocene deposits older than the Sappa were facilitate comparison of figures 183 and 184, some
recognized with certainty in the Medicine Creek of the stratigraphic units distinguished by Deffen-
basin. baugh were combined in order that they correspond
The thickness and areal distribution of pre-Wis- with the units shown on figure 184.
consin stratigraphic units cannot be determined
without data from drilling. Moreover, the logs must STRATIGRAPHY ON CEDAR CREEK NEAR
STOCK VILLE
include full and accurate descriptions of the rocks
that are penetrated, because the different Pleistocene Along Cedar Creek, about 3 miles southwest of
stratigraphic units are similar in rock type to one the village of Stockville (Bartley NW quad., NE^4
another and to the Ogallala Formation. Two geo- sec. 18, T. 6 N., R. 27 W.), several sections have been
logic sections (figs. 183 and 184) have been compiled exposed by lateral cutting of the creek, and the
from logs of test holes that were drilled by the terrace sequence is more complete and distinct than
SOUTH
Mousel /*' st°c kvil| e

terrace
2500'
DATUM IS MEAN SEA LEVEL
FIGURE 187. Profile and geologic section on Cedar Creek.
usual for the Medicine Creek basin. The section Retinella electrina (Gould) 3
described below applies to the right-hand (south) Lymnaea [Fossaria] parva ________ 2
side of figure 187: Depth Vallonia gracilicosta Reinhardt 15
(ft) Valvata tricarinata (Say) 4
Peorian Loess: According to Leonard (1950), G. proarmifera is
Silt, pale-yellow (2.5F 7/3 dry, 2.5F 6/3 moist),
calcareous. Surface soil not present _. .. .. . 0-15 restricted to the Yarmouth, and none of the species
Loveland Formation: in the collection is restricted to deposits younger
Silt, very pale brown (WYR 7/4 dry, 10F# 5/4 than the Yarmouth. Correlation of the soil with the
moist); a few streamers of fine sand near base 15-20
Gravel, sand, and silt, crossbedded, calcareous, Yarmouth and of the underlying silt with the Sappa
locally cemented with carbonate ________________________ 20-22 is, therefore, supported by the faunal evidence.
Silt, pale-brown (WYR 6/3 dry, WYR 4/2 V. tricarinata is aquatic, and the fauna as a whole
moist), noncalcareous; indistinct sedimentary
lamination .. -..--...._.__....__.. ......_._._.._..._.... 22-26
indicates a moist environment such as would be
Yarmouth soil: afforded by a^valley.
Silt, clayey, gray-brown (WYR 5/2 dry, 10F.R The stratigraphic and topographic relations shown
4/2 moist). Weak columnar structure, col- in figure 187 indicate that the valley of Cedar Creek
umns irregular, 2 to 5 in. wide; noncalcareous.
Upper contact gradational ______^__-__^_ _____________ 26-28.9
was cut not later than Kansan time and that at
Sappa Formation: least five episodes of valley alluviation, each fol-
Sand, fine, and ashy light-gray (2.5F 7/2 moist) lowed by cutting, have taken place since Yarmouth
silt; a few thin white irregular seams of vol- time.
canic ash. Upper 0.5 ft mottled with darker
gray, gradational with unit above. Upper 2 STRATIGRAPHY ON CUT CANYON NEAR CURTIS
ft is Cca horizon of Yarmouth soil; contains Near the confluence of Cut Canyon with Fox
meshwork of fine carbonate veinlets ____ ___________- 28.9-32 Creek, a significant section of Pleistocene deposits
Sand, fine, and ashy silt; a few streamers of
sand and fine gravel near base; noncalcareous 32-35
is exposed along the county road that descends
Gravel and other debris from Ogallala Forma- eastward from the upland and crosses Cut Canyon
tion ; crossbedded; a few streamers of fine sand 35-37 in the NWi/4 sec. 29, T. 9 N., R. 28 W. The presence
Ogallala Formation: 37-53 of deposits of the Sappa indicates that the canyon
A snail fauna was collected from the lower part of was cut not later than Kansan time, and the Pleisto-
the soil identified as Yarmouth and from the upper cene sequence of pre-Wisconsin age seems to be
part of the unit identified as Sappa, and identifi- complete. Both the Yarmouth and the Sangamon
cation was made by Dr. W. J. Wayne of the Indiana soils are well developed. Of the terrace deposits of
Geological Survey. In the following table, the num- late Wisconsin and Recent age, however, only the
ber of individuals of each species is given in the Stockville is exposed (fig. 188).
right-hand column. A stratigraphic section along the road, beginning
Gastrocopta tappaniana (C. B. Adams) _^_______- 2 1,500 feet west of Cut Canyon bridge and ending
proarmifera Leonard _._._. ._..-..._-_.__..__ 3 1,150 feet west of the bridge, is as follows:
Gyraulus tircumstriatus (Tryon) ___________________ 21
_ . T : , Depth
Helicodiscus parallelus (Say _- _ _ . 3 Peorian Loess: , ~> - (ft)
Physa anatina Lea ~ _ _._ ~_ _______ 2 Silt, yellowish-gray, massive, homogeneous ________ 0-30
WEST EAST
2800' -i
2600'
FIOUBE 188. Profile and geologic section along road entering Cut Canyon.
Depth
(ft)
checked and corrected by Dr. W. J. Wayne of the
Sangamon soil and Loveland Formation: Indiana Geological Survey. The number of each
Silt, yellowish-brown (10YR 5/4 when moist), species found is indicated in column at right in the
friable; no distinct structure ._____________________ 30-32
Silt, light-yellowish-brown (10YR 6/4 when
following list:
moist); weak prismatic structure, noncalcare- Carychium exile canadense Clapp ____._ ____ 2
ous -_-_I____________________________________________ 32-38 perexiguum Baker ______ _ ___________________ 2
Silt, clayey, approximately same color as unit Gastrocopta pentodon (Say) ______________________ 6
above, but has pinkish cast ___._-______.______ 38-38.5 proarmifera Leonard ______-____ __- _ -_ 2
Silt, streaked and veined with abundant soft Gyraulus circumstriatus (Tryon) _ ______________ 19
carbonate ____ ____________________________________ 38.5-43 Pupilla muscorum (Linne) __________________ 1
muscorum sinistra Franzen ______-______-_ _ 1
Yarmouth soil: Retinella electrina (Gould) ___.___ ___._ _.__-___-_. 1
Silt, light-brownish-gray (5Y 7/2 when dry, Vallonia gracilicosta Eeinhardt _______-_-_ ___ 10
WYR 6/2 when moist), clayey; streaked with Valvata tricarinata (Say) ________ ___ __ _ _ 24
carbonate from unit above, but otherwise non- Vertigo nylanderi Sterki __________ ______-_____ _ 3
calcareous. Breaks into irregular small, poly- According to Leonard (1950), G. proarmifera and
hedrons (5 to 10 mm in diameter) that fit to-
gether. Contact with unit above sharp but
P. muscorum sinistra are restricted to the Yarmouth
irregular ___J_____-___i____.._______ _____________ 43-46.5 and (7. perexiguum is restricted to deposits of Ne-
Silt, yellowish-brown (10YR 5/4 when moist); braskan and Yarmouth age; none of the species
contains abundant small iron concretions ________ 46.5-47 found is restricted to deposits younger than Yar-
A stratigraphic section 800 feet west of the bridge mouth age. Therefore, correlation of the ashy silt
is similar to that described above, except the Sappa with the Sappa Formation is reasonable.
Formation is exposed and the Yarmouth soil is STRATIGRAPHY ON ELKHORN CANYON NEAR
somewhat different, as shown below: MAYWOOD
Depth The Peorian Loess, Stockville terrace deposits,
(ft)
Yarmouth soil: and late Recent alluvium are exposed along Elkhorn
Silt, olive-gray (SYR 4/2 when moist) ____________ 0-1 Canyon about 2 miles southwest of the village of
Silt, light-gray (WYR 7/2 when moist) ; crumb Maywood, near a bridge where the county road
structure __________________________________________________________ 1-2.5
Silt, light-gray, mottled with rust color; struc-
crosses the canyon (Curtis SW quad., center of
ture weakly prismatic, prisms have irregular section line between sees. 31 and 32, T. 8 N., R. 29
sides, 1 to 3 cm wide _________________________,_: 2.5-4.1 W.). The exposure was studied after it had been
Silt, light-gray; no distinct structure _._...__v_ 4.1-7.5 thoroughly moistened by rainfall and runoff, because
Sappa Formation: moistening greatly increases visibility of the sedi-
Sand, fine-grained, and silt; cross-laminated ___ 7.5-9.5 mentary structures and textures. The relations be-
< i
No fossils were found at this section, but fossils tween the different units are shown .in figure 189,
were found in gray ashy silt, correlated with the and a general view of the locality is shown in figure
Sappa, about 600 feet to the east. The faunas were 190. Evidence was found in the late Recent alluvium
îdentified by the writer, and the identification was for two episodes of cutting and filling, which took
SOUTH NORTH
2730' -i
Stockville terrace
2720' -
2710' -
Numbers refer to parts of
Peorian Recent deposits
Loess
2700' -
2690' - \,
Layer from which

2680' - C u sample W-754 \
was collected \
Flow line of Elkhorn Canyon Recent deposits
Ogallala Formation
2670'
VERTICAL EXAGGERATION X2V2
FIGURE 189. Profile and geologic section on Elkhorn Canyon.
place (according to a carbon-14 age determination) higher content of organic matter. Part 1 is also
within the past 300 to 400 years. banded, but the banding is inconspicuous because
The late Recent alluvium consists of four parts, of the relatively lighter color of the clay-rich layers.
which are designated by the numbers 1 through 4 Part 2 is riddled with worm burrows, which are
on figure 189. Part 1 is oldest, lightest in color, and partly filled with castings and dark silt, and many
most compacted, whereas the younger parts are of which terminate in a round cavity partly filled
successively darker in color and less compacted. with castings. The darker clay-rich bands seem to
The contact between part 1 and part 2, which lies have been preferred by the worms, for these bands
at a depth of about 12 feet below the ground surface, contain castings in greatest abundance. Worm cast-
is conformable and nearly horizontal. Beneath the ings are much less apparent in part 1, although
contact, the deposits of part 1 show no evidence of dark vertical streaks, which probably represent the
erosion or of leaching. Part 2 is characterized by filled burrows of worms, were seen from place to
light brownish gray bands of fine sand and silt place, and the bedding is apparently riddled by worm
(2.5Y 6/2 moist) alternating with darker grayish burrows. A lens of charcoal and wood ashes, mixed
brown bands of clayey silt (2.5Y 4/2 moist), whose with silt and containing a single large fragment of
darker colors result from a finer texture and a charred bone, was found interbedded in part 2 at
a depth of about 7.5 feet below the ground surface
and 3 to 4 feet above the contact with part 1.
Charcoal (Geol. Survey sample W-754; analysis by
Meyer Rubin) from the lens yielded a carbon-14
date of 420±160 years B.P. (before present).
Parts 1 and 2 of the late Recent alluvium are de-
scribed here because the contact between these parts
is considered to mark a distinct change in the rate
of accumulation of the late Recent alluvium; the-rate
of accumulation of part 2 is considered to be much
the more rapid. Furthermore,-the more rapid accu-
mulation of part 2 probably reflects an increase in
rate of erosion on the uplands and valley1 sides.'
Part 1 is lighter in color because its slower rate of
accumulation permitted a greater degree of oxida-'
tion of finely divifled organic matter, which was 5
FIGURE 190. View of terrace sequence and valley fills along Elkhorn Canyon.
Prominent scarp marks front of Stockville terrace; lower scarp marks con-
originally intermixed with the silt and clay. A
tact between parts 3 and 4 of late Recent alluvium. slower rate of accumulation of part 1 is also indi-
cated by the greater destruction of its sedimentary LOVELAND FORMATION AND SANGAMON SOIL
structures by earth worms. If the described changes The most useful and distinctive soil marker within
from part 1 to part 2 do in fact reflect a change in the Pleistocene deposits is the Sangamon soil, which
rate of accumulation, this change took place about is developed on the Loveland Formation. No other
500 years ago, as indicated by the position of the soil in the region, including the modern soil, ap-
radiometrically dated charcoal. According to arche- proaches the Sangamon in depth of profile or in
ologic evidence, a period of drought may have begun degree of development of the Cca horizon. On the
about 500 years ago, when the region was apparently uplands the Loveland consists of loess, which mantles
evacuated by people of the Upper Republican the valley sides and merges into contemporaneous
Culture. valley-fill deposits of silt and sand. The maximum
Parts 3 and 4 of the late Recent alluvium are dark observed thickness of Loveland in the Medicine Creek
gray brown (WYR 5/2 moist) and show rather dis- basin is 35 feet (along Cedar Creek), and the aver-
tinct banding. Carbonized fragments of grass stems, age thickness is about 20 feet. At some exposures
having an average length of about 0.5 mm, are scat- in valleys, the Loveland is missing and the Sangamon
tered throughout, but they are more abundant in soil is developed on the Ogallala Formation.
the darker, clay-rich bands. Dark laminae formed Moist loess of the Loveland is typically light yel-
of ashes and charcoal, which probably represent the lowish brown (WYR 6/4), but darker colors (WYR
debris from grass fires, were seen from place to 5/3 and WYR 5/4) were observed at some localities.
place. Much of the alluvium seems to have passed The Sangamon soil, as observed from a distance, is
through the digestive tracts of earth worms, whose a distinct dark-brown humus-rich band that sepa-
castings dominate the texture and account for the rates loess of the Loveland from the distinctly lighter
loose consolidation. colored loess of the Peorian. (See fig. 191.) This
The steeply dipping erosional contact that sepa- band ranges from about 4.5 to about 2 feet in thick-
rates parts 1 and 2 from part 3 is attributed to the ness and from yellowish brown (WYR 5/4) where
passage of a major gully head scarp along the valley the soil is weakly developed to very dark gray brown
of Elkhorn Canyon. Part 3 of the late Recent allu- (WYR 4/3) where it is strongly developed. A Cca
vium, which was deposited in the gully left by this horizon is nearly everywhere present in the soil,
head scarp, was in turn trenched by a second head but its depth below the dark band ranges from
scarp following consecutively behind the first. The about 2 feet in some localities to about 25 feet in
former sides of the trench made by the second head others, and its thickness is also variable. The varia-
scarp are represented by a pair of low terraces that tions indicate that development of the Sangamon
were traced (on aerial photographs) to their ter- in some localities was nearly continuous, whereas
minus about 3 miles upvalley. Deposits formed in other localities the developing soil was intermit-
behind the second head scarp (part 4) were in turn tently buried by fresh accumulations of silt.
trenched by a third head scarp following consecu- In spite of the conspicuous development of the
tively behind the second. In 1952 the third head humus-rich part of the Sangamon soil profile, it
scarp was 13,300 feet upvalley from this locality, contains an abundance of fresh ferromagnesian min-
and it advanced about 500 feet during the period
1937-52.
Relatively little time is required for the formation
of a series of low terraces by the passage of consecu-
tive gullies, because deposition is rapid on the floor
of a gully and the head scarps travel rapidly as a
result of concentration of flow on the gully bottom.
The average rate of advance of the third head scarp
during the period 1937-52 is probably conservative,
partly because the head scarp has reached bedrock.
An average advance of about 200 feet a year, at-
tained by some head scarps in the Medicine Creek
basin during the period 1937-52, would be adequate
to account for passage of the events described during
a period of about 350 years. On the other hand, FIGURE 191. Valley fill of the Peorian Loess set unconformably against
these events would not likely be compressed into a Sangamon soil and Peorian Loess. Steeply dipping unconformity, indi-
cated by arrows, is exposed on either side of the small valley, and the
period of much less than 350 years. valley fill of the Peorian is between the arrows.
erals and hence has undergone little chemical weath- silt or of some other sediment type, are called the
ering. About 20 species of heavy minerals were valley-fill deposits of the Peorian.
identified under the petrographic microscope; of Leonard (1951) recognizes three biostratigraphic
these, clinozoisite, green hornblende, apatite, and zones in the Peorian Loess of Kansas, but these
brown hornblende are by far the most abundant. In zones are not marked by any physical discontinuities
addition, calcic feldspar (labradorite) was identified in the loess. The basal zone, which is devoid of
among the light minerals. molluscan fossils, is overlain by a lower molluscan
faunal zone and an upper molluscan faunal zone.
PEORIAN LOESS, BRADY SOIL, BIGNELL LOESS, Between the upper and lower faunal zones is a
AND MODERN SOIL transitional faunal zone, which bears elements of
An account of the involved history of the terms both the lower and upper faunal assemblages.
"Peoria" and Peorian has been given by Leonard A snail fauna was collected about 25 feet strati-
(1951, p. 323). In this report, the term Peorian graphically above the Sangamon soil at a locality
Loess is applied to deposits that are stratigraphically on Dry Creek to ascertain the approximate position
above the Sangamon soil and terminate in the Brady of these faunal zones relative to the total thickness
soil. The absolute time at which the Sangamon of the loess of the Peorian in the Medicine Creek
Interglaciation ended has not been determined, but basin. The locality is in the NW^4 sec. 16, T. 9 N.,
a reasonable estimate can be made by extrapolating R. 27 W. The stratigraphic position of the fauna
from existing carbon-14 dates; the estimate of Frye is about midway of the total thickness of the loess
and Willman (1960, p. 2) is between 50 and 70 on Dry Creek. A sample of loess weighing about
thousand years B.P., as measured by carbon-14. 25 pounds was collected, and the contained snails
The Brady soil has been dated, by means of a carbon- were obtained by washing the loess through a sieve.
14 age determination on organic carbon from the Identification of the snails was made by Dr. W. J.
A horizon of the soil, at 9,160±250 years B.P. Wayne of the Indiana Geological Survey, and the
(Rubin and Suess, 1956, p. 443). This date is prob- faunal list is as follows (the number of individuals
ably too young because of root hairs in the sample. of each species is given in right-hand column) :
In general, the Peorian Loess mantles the whole Columella alticola (Ingersoll) ___ _ - 2
Discus cronkhitei (Newcomb) ___. __ ._-._ 2
of the Medicine Creek basin except the modern shimeki (Pilsbry) ._.____-__________ __ _- 24
valley flats and the Stockville and Mousel terraces. Euconulus fulvus (Miiller) __ ____ -_ 2
On the uplands the Peorian consists of massive silt, Pupilla vmiscorum (Linne) _ ______ 4
which grades laterally (at least at the surface) to Succinea grosvenori Lea _._-_ _-_ - 16
fine sand in the vicinity of dune areas. This massive Vallonia gracilicosta Reinhardt ._ - ___ 2
Vertigo modesta (Say) __________________________ 3
silt is called loess because its topographic position tridentata Wolf .._____________ 10
indicates that it is eolian and because the snails that
it contains are not aquatic. In valley locations also, Of the nine species listed, five belong to the upper
the greater part of the thickness of the Peorian and transitional zones of Leonard, and four occur
consists of massive silt, except in the valleys in the in all three zones. Because of the absence of S. avara
vicinity of dune areas, where it consists of silt and and the numerical dominance of D. shimeki and
fine sand. The lowermost 10 to 20 feet of the Peorian S. grosvenori, which are upper zone species, the
in the valleys commonly consists of silt interbedded assemblage is assigned by the writer to the upper
with fine gravel, silty clay, or sand; and this lower- zone. The fauna in this single locality indicates that
most part contains, in some localities, aquatic snails at least half of the total thickness of the Peorian
or pelecypods. The massive silt of the valleys is not in the Medicine Creek basin was deposited during
called loess because its agent of deposition is not the time represented by the upper faunal zone of
Leonard. Leonard correlated his upper zone with
apparent. Where a vertical face cut in this silt is
exposed to wind and rainwash, it commonly shows the Tazewell Stade of the Wisconsin Glaciation.
During accumulation of the Peorian on the up-
well-defined lamination that could be attributed
lands, a period of incision to bedrock took place in
either to wind or to water. Eolian deposition of silt the valleys. Valley fill of the Peorian that accumu-
on the uplands was probably accompanied by an lated after this incision is set unconformably against
equal or greater amount of eolian deposition in the the Loveland Formation, the Sangamon soil, and
valleys, but the silt in the valleys was probably the Jower part of the Peorian. (See figs. 185 and
reworked by water before final deposition. Deposits 191, both of which show valley fill of the Peorian
of the Peorian in valley locations, whether of massive set unconformably against the Sangamon soil and
the lower part of the Peorian.) From a distance the from the logs of test wells drilled by the Nebraska
valley fill of the Peorian looks almost white, whereas Conservation and Survey Division. (See figs. 183
the loess of the Peorian that overlies the Sangamon and 184.) The thickness of the loess reaches a maxi-
soil has a yellowish cast. On the uplands no physical mum in the vicinity of the Platte River, gradually
discontinuity within the Peorian was discerned. decreases southward to a minimum in a belt which
Molluscan fauna collected from the base of the is about 30 miles south of the Platte River and 10
valley fill of the Peorian at two localities indicates miles north of the Republican River, then again in-
that deposition of the valley fill began in Tazewell creases in thickness in the direction of the Republi-
time. The first locality is on the bank of Dry Creek, can. A well near the north side of the Republican
about 700 feet upstream from the gaging station, River valley penetrated a thickness of 60 feet of
in the SE^4 sec. 24, T. 8 N., R. 28 W. Here the loess, whereas a well near the south side of the
valley fill rests unconformably on the Ogallala For- valley penetrated a thickness of only 30 feet. About
mation. Identification of the fauna was made by 50 feet of loess of the Peorian is exposed in the
the writer, with the assistance of Dr. W. J. Wayne. vertical wall of a deep pit, 1.5 miles west of Eustice,
The faunal list is as follows: Nebr. (fig. 192).
Columella alticola (Ingersoll) The valley fill of the Peorian is thicker than the
Discus cronkhitei (Newcomb) loess. The exposed thickness of valley fill of the
shimeki (Pilsbry) Peorian on the main stem of Medicine Creek, as
Gyraulus circumstriatus (Tryon) measured from the surface of the Wellfleet terrace
Physa anatina Lea to the modern valley flat, is about 125 feet. On the
Pupilla muscorum (Lirme)
Retinella electrina (Gould) assumption that the valley fill extends to bedrock,
Succinea grosvenori Lea as it does on Dry Creek, the total thickness on
Vallonia gracilicosta Reinhardt Medicine Creek is about 200 feet. On the same
Vertigo gouldi coloradensis Cockerell assumption, the total thickness on Well Canyon,
modesta (Say) at a locality about 7 miles north of Curtis, is 180
Zonitoides arboreus (Say)
feet. The exposed thickness in the upper reaches
According to Leonard (1952, p. 19), D. shimeki of Dry Creek is 42 feet, and the total thickness is
and D. cronkhitei are reliable indexes to the upper about 65 feet.
(Tazewell) faunal zone of the Peorian. None of the
Loess of the Peorian in the Medicine Creek basin
other species in the assemblage is restricted to de-
is a massive friable light-colored silt, similar to
posits either older or younger than this upper faunal
loess that has been described in other regions. The
zone.
loess when dry is typically light gray (WYR 7/2)
A second fauna from the valley fill of the Peorian and when moist is light brownish gray or pale
was collected at the side of Mitchell Creek, near a
county road in the NW1/^ sec. 9, T. 6 N., R. 26 W.
Identification was made by the writer, with the as-
sistance of Dr. W. J. Wayne. The faunal list is as f^^^, ~+
follows:
Discus cronkhitei (Newcomb)
'Euconulus fulvus (Muller)
'Gcfstrocopta armifera (Say)
Lymnaea (Stagnicola) pahistris (Muller)
, Retinella electrina (Gould)
., Succinea grosvenori Lea
Vallonia gracilicosta Reinhardt
Vertigo tridentata Wolf
Although this Assemblage is different from the as-
semblage collected from the valley fill of the Peorian
on. Dry 'Creek, a correlation with the upper mollus-
can faunal zbne of Leonard (1951, 1952) is indicated
by the presence of D. cronkhitei and £. grosvenori.,
The,average tnickness of the loess of the Peorian, FIGURE 192. Peorian Loess (50 ft thick) overlying a horizon of Sangamon
soil (upper dark band, marked S). Units beneath Peoriah Loess are
in the Medicine Creek basin is about 50 feet, and the, Lpveland Formation (20 ft), Yarmouth soil (lowermost 'dark band,,
maximujp thickness is,about 80 feet. Thicknesses' marked Y), Sappa Formation (33 ft), and Pearlette A§h Member of
S'aJppa' Formation'(20 ft): Excavation, made for removal of Pearlette,
were measured at'exposures in upland'localities and. is 1.5 miles west of,'Eustice, Nebr. Photographed by C. H. Hembree, 1956.,
'brown (10YR 6/2 or 6/3). Although the loess ap- stances that were peptized or dissolved in the soil
pears to be homogeneous when viewed from a dis- solution are left as coatings on grains. Although
tance, closer inspection reveals abundant hollow the grain coatings described by Kubiena 1 were of
'tubules made by rootlets, worm burrows, and small humus, coatings of clay also probably form by
botryoidal structures which are worm castings. 'wetting and drying. ' '
Within 5 feet of the ground surface, the loess may In this region loess of the Peorian is generally
show such an abundance of these botryoidal struc- "calcareous, but the calcium carbonate is not uiii-
tures as to indicate that all of it has passed through formly disseminated. Calcium carbonate in the loess
the digestive tracts of earth worms. Snail shells occurs in the form of silt-sized grains, which are
are generally present and are locally abundant. doubtless primary; as powdery streaks and vein
Lamination, although not usually observable, may fillings; as fine-grained aggregates that line or fill
appear where erosion by wind and raindrops has rootlet tubules, worm burrows, or other openings;
been sufficiently delicate to etch the laminae into and as concretions. Many loess outcrops show zone's
relief. up to several feet thick that are noncalcareous'.
Thin sections of seven samples of loess of the Frankel (1957) has reported vertical variations in
.Peorian, collected at different localities in the Medi- the distribution and concentration of secondary
cine Creek basin, were studied under the petro- calcareous concretions and in the abundariee "and
graphic microscope. The most distinctive textural state of preservation of fossil mollusks in the loess
features of loess, when viewed at high magnification, in Nebraska. He attributes the"s£ variations 'to an
are the rather close packing arrangement and the intermittent rate of loess depositibn. S16w: deposi-
presence of biref ringent clay coatings around detrital tion of loess is accompanied by solution of snail
silicate grains. Grains larger than 125 microns in shells and by accumulation of secondary carbonate
maximum diameter constitute less than 1 percent by at depth. Rapid accumulation of loess is _ indicated
volume of the samples, and grains larger than 31 by unaltered snail shells. Frankel suggests that
microns constitute 55 to 75 percent by volume. many phantom soils are present. The writer has
Particles larger than 31 microns may be regarded observed that massive silt, accumulated as1 Valley
as the framework or skeleton of the loess, and par- fill and indistinguishable from upland wind-laid
ticles smaller than 31 microns may be regarded as loess, is more commonly noncalcareous than upland
the matrix. The clay grain coatings, which are evi- loess.
dently authigenic, may be regarded as cement. The The Brady soil' was named and described by
framework grains are mainly separated by their Schultz and Stout (1948) from a locality on the
clay coatings and by the matrix. Contacts between steep south side of the Platte Valley near Bignell,
framework grains are mainly tangential or long, Nebr. This locality is also the type section for the
and an average of about two contacts per grain was overlying Bignell Loess (Schultz and Stout, 1945).
observed. According to the writer's observations, the huinic
The birefringent clay coatings around detrital zone of the type Brady soil is about 1.3' feet thick,
silicate grains consist of crystalline flakes oriented and the soil is weakly calcareous throughout; soft
parallel with the surface of the coated grain. These streaks and films of carbonate appear to a depth of
coatings merge with the matrix and probably account 2.2 feet from the top of the soil, and these are con-
in considerable part for the coherence of the loess spicuous from 1.3 to 2.2 feet. Overlying the Bratjy
and for the ability of loess to maintain a vertical soil is a thickness of 8 to 10 feet of Bignell Loe'sS,
face. Clay coatings on flat grains, such as mica on which the modern soil is developed. This modern
flakes and glass shards, are in general thicker than soil is noncalcareous to a depth of 3.2 feet and shows
coatings on equant grains. Clay coatings on grains a distinctly columnar structure to a depth of 1.3 feet.
are not characteristic of siltstones deposited under Above the modern soil is about 1 foot of] lighter silt,
water. In loess the coatings are probably formed evidently colluvial and related to cultivation; of fields
by wetting and drying qf th'q surface layer during upslope from this locality. The modern sorf appeal's
deposition. Kubiena (1938, p. 134) has attributed to be more strongly developed than the Bra&y soil.
the formation of grain coatings in soils to such 1 In spite of the fact that Medicine^ Creek* basin is
wetting and drying. After rainfall, the soil solution adjacent tb the type locality 6f both the 'Brady soil
may-fill'all pore spaces in the soil, but evaporation and the overlying Bignell'Loess, the Brady soil, as
at the ground surface causes the soil solution to separately developed from the riioderii soil, wa4 op-
retreat to the angles of the intergranular spaces and served at only three exposures, all in the northern
to the surfaces of grains. As the soil dries, sub- part of the basin. " ' -
< General deposition of a thin layer of Bignell Loess the fauna, none of the species is extinct, and all are
over the uplands of the Medicine Creek basin is fresh-water species that now inhabit parts of central
indicated by abnormally thick A horizons that com- United States. Leonard (1952, p. 16) concludes,
monly appear in the modern soil. The modern soils from his study of Bignellian molluscan faunas in
of Lincoln County, Nebr., have been mapped and Kansas, that the environment of Bignell Loess depo-
described by Goke and others (1926); and the soils sition was very much like present conditions in the
of Frontier County, Nebr., by Bacon and others Great Plains. The Bignell Loess is considered by the
(1939). The Medicine Creek basin lies within these writer to be the upland counterpart of the Stockville
two counties. Principal soils of the loess-mantled
areas are the Holdredge very fine sandy loam and terrace deposits.
the broken phase of the Colby very fine sandy loam. MOUSEL TERRACE DEPOSITS
The Holdredge soil, which is developed mainly on The Mousel terrace deposits consist mainly of silt
areas of flat or rolling upland, is mature. The B and differ from the Stockville terrace deposits only
horizon of the Holdredge lies between 10 and 24 by being somewhat darker in color and more com-
inches in depth and shows an imperfectly developed pact. In some localities the Mousel terrace deposits
columnar structure; and the Cea horizon, which contain fragments of carbonized grass stems, which
contains a concentration of carbonate in the form are rare in the Stockville terrace deposits. Along
of streaks, splotches, or threads, lies between 3 the main course of Medicine Creek, two buried
and 4 feet in depth. A dark-brown zone, probably humus-rich bands, which are interpreted to be
representing the Brady soil, lies between the B immature soils, were observed in deposits of the
horizon and the Cca horizon at many localities.
Mousel terrace.
STOGKVILLE TERRACE DEPOSITS Charcoal suitable for carbon-14 age determination
Typical relations of the Stockville terrace deposits was collected from the Mousel terrace deposits on
to other stratigraphic units are shown, in figure 185. Dry Creek in the NW^NEi/4 sec. 1, T. 7 N.,' R.
Along major valleys, the unconformity between the 28 W. The charcoal was in a lens of wood ashes
Stockville terrace deposits and older units is rarely and silt, buried at a depth of 14.5 feet below the
marked by a scarp; if one were originally present, ground surface. The age was reported as 2,200±200
it has been obliterated by erosion on the valley-side years (J. L. Kulp, Lamont No. 239C).
slopes. Along minor valleys, the unconformity is
RECENT ALLUVIUM
usually marked by a distinct scarp.
The Stockville terrace deposits consist mainly of Where valley trenching has exposed the full thick-
calcareous silt. In minor valleys the silt is less ness of the late Recent alluvium as at localities on
compact, the color is darker, and the bedding is less Dry Creek, Cedar Creek, Elkhorn Canyon, and Lime
well defined than in major valleys. Dark humus-rich Creek the alluvium rests on the bedrock floor of the
bands, beneath which the silt is leached to a depth valley. Therefore, at these localities the valleys were
of a foot or two, are observable at most exposures incised to bedrock during the episode of incision that
of the Stockville terrace deposits. The depth of followed accumulation of the Mousel terrace deposits.
burial of the uppermost band ranges, in different Incision to bedrock may have been confined tq
parts of the basin, from about 2 to 20 feet; and localities where the bedrock is high beneath the
the number of bands varies from one exposure to valley floor. In some valleys, such as Dry Creek and
the next. These bands, together with the underlying Elkhorn Canyon, accumulation of the late Recent
leached zone, .probably represent immature soils that alluvium has been interrupted by one or more
formed during periods of slow deposition. The Stock- episodes of incision during which the valley was
ville terrace deposits occur in nearly all valleys of trenched for part of its length by the upstream,
the present drainage system, including valleys of migration of intermittently spaced channel scarps.
first order.* The thickness, which depends on the In other valleys, such as Well Canyon and Fox Creek,,
size of the valley, is about 100 feet along Medicine accumulation of the late Recent alluvium has evi-
Creek and Well Canyon, about 60 feet along Lime dently proceeded without the interruption of valley
Creek, and about 30 feet along Dry Creejc. incision.
. A fauna of t snails and pelecypods was collected In the upper reaches of Dry Creek, where the
from the StocJkyiHe terrace deposits exposed along previously untrenched valley flat is being trenched
Medicine Cre^k ija the Np cor. sec, 13, T, 6 N., R. 27 by head ward migration of a major channel scarp,
W. According to Dr. W. J. Wayne, who identified the silty late Recent alluvium consists of a banded
alluvium is exposed in the walls of rapidly advancing

trenches on Dry Creek and Curtis Creek Canyon.
The present rate of accumulation of alluvium on
valley flats must be known in order to relate sedi-
ment discharge by streams from the drainage basin
(sediment yield) to the amount of sediment that is
eroded from uplands and valley-side slopes (gross
erosion). In spite of a thorough search, exposures
of the late Recent alluvium yielded no artifacts
nor other materials that would give an accurate
indication of the rate of accumulation. Valley flats
in many localities were examined after runoff events,
and the depth of accumulation of freshly deposited
sediment was measured. The thickness of the freshly
deposited silt layer ranges widely (from a fraction
of an inch to about 6 in.) from one locality to an-
other, and almost as great a range was observed
from place to place at a single locality. Moreover,
FIGURE 193. Banding in late Recent alluvium, Dry Creek. freshly deposited alluvium may either remain at its
(SW%NE% Sec. 29, T. 9 N., R. 27 W.)
place of deposition or be removed during the next
runoff event. Farmers were asked about the time
upper part, about 14 feet thick, and a lower part required for fenceposts on the valley flat to become
that is indistinctly banded and lighter in color. wholly or partly buried. In the upper part of Dry
(See fig. 193.) The dark bands of the upper part Creek, two farmers estimated independently that
are somewhat more clayey than the light bands, 5 or 6 feet of sediment had accumulated on the valley
and the dark color is a result of a finer texture and flat between 1920 and 1953. On Well Canyon one
a higher content of organic matter. Both light and farmer estimated an accumulation of 2 feet on the
dark bands are composed mostly of loosely consoli- valley flat in the past 40 years (1916-56) and an-
dated silt, through which abundant fragments of other estimated an accumulation of 2 feet in the
carbonized grass stems are scattered. Earth-worm past 30 years (1926-56). On Medicine Creek, about
castings are a conspicuous feature of the upper 4.5 miles north of Cambridge, a farmer reported
banded part, but they are less apparent in the more that the wire of a hogpen, last used before the 1935
compact silt of the lower part. flood on Medicine Creek, was buried to a depth of
A similar division of the late Recent alluvium of 6 feet by 1957. An average rate of accumulation
Elkhorn Canyon into an upper part that is banded of about 1 inch per year on the valley flats of the
and a lower part that is indistinctly banded has drainage basin is probably of the right order of
already been described. The banded fill is attributed magnitude. A range between ^ inch and 3 inches is
to a rate of accumulation so rapid that time has not estimated, and this range would be found not only
been available for oxidation of finely divided organic from valley to valley but also from place to place
matter in the clayey bands. If the accumulation of within the same valley.
banded fill began at the same time on Dry Creek as The rate of accumulation of alluvium within a
on Elkhorn Canyon, the banded fill shown in figure trenched channel, downstream from an actively ad-
193 has accumulated during the past 500 years, vancing channel scarp, must be considered separately
approximately. This corresponds to an average rate from the rate of accumulation on an untrenched
of accumulation of about 0.34 inch per year, al-
valley flat. Examination of freshly deposited allu-
though the appearance of the banding indicates that
vium on the floor of trenched channels, several hun-
the rate of accumulation was not uniform.
The banded alluvium seems to be characteristic of dred feet downstream from an actively advancing
valleys that are being actively trenched, whereas channel scarp, indicates that 0.5 foot or more of
the more compact unbanded alluvium is character- alluvium can be deposited during a single runoff
istic of more stable valleys. For example, the allu- event. At several localities on Dry Creek, tin cans
vium beneath the valley flats of Well Canyon and and wire were found buried to a depth of several
Fox Creek, which are not being actively trenched, feet in alluvium that had accumulated on the floor of
is unbanded. On the other hand, distinctly banded the trench. Evidently, alluvium accumulates rapidly
in trenched channels but is susceptible to removal have a maximum length of about one-fourth of a
by a second episode of trenching. mile, an average width of about 200 feet, and heights
ranging from 10 to 30 feet; they show the furrowed
SIGNIFICANCE OF CARBONIZED FRAGMENTS crest that is typical of longitudinal dunes in the
OF GRASS
Sandhills. They are probably relicts of a more ex-
In this region the younger a fluvial deposit is, tensive dune system that has in most places been
the darker is its color. The dark color is due in destroyed by blowouts but has in some places been
part to finely divided organic matter and in part modified into U-shaped dunes. Although the longi-
to fragments of carbonized grass stems, which are tudinal dunes are oriented about N. 65° W., the ends
more abundant in the younger deposits. The car- of most of the U-shaped dunes are oriented more
bonized fragments are attributed to grass fires, in- northerly. Elongated blowouts tend to be oriented
asmuch as uncarbonized plant fragments would at about N. 20°-30°W. Evidently, the prevailing
decay in this environment. The scarcity of carbon- wind direction has shifted since the formation of
ized stems in the older deposits might mean that the longitudinal dunes. Most of the blowouts are
grass fires were uncommon during the deposition clearly man induced, for they are associated with
of these deposits, or it might mean that any frag- areas that formerly were or currently are cultivated.
ments originally present have been destroyed by The stream-dissected relief of the loess-mantled
oxidation or by the activities of earth worms. In an part of the basin is separated from the sand-dune
effort to decide between these alternatives, five thin relief by a transitional belt of ground that is inter-
sections of the Stockville terrace deposits, one thin mediate both in relief and in particle size of under-
section of the Mousel terrace deposits, and two thin lying materials. The relief of this transitional belt
sections of the late Recent alluvium were examined. is characterized by broad undrained depressions,
In mineralogy, degree of development of clay coat- generally elongated northward and separated by
ings around grains, and packing of particles, all nearly level ground or by broad, rounded elevations.
these thin sections were similar to thin sections of The underlying material, which is intermediate in
loess. However, the upper Recent alluvium shows particle size between dune sand and loess, is re-
many open spaces, which are attributed mainly to stricted to upland locations. The mixture of dune
the activity of earth worms. Although carbonized sand and loess, on which the Anselmo fine sandy
fragments in the Stockville terrace deposits are loam and the Colby fine sandy loam are developed,
small, they show no signs of partial oxidation. The is generally less than 3 feet thick and is underlain
tentative conclusion is that grass fires were more either by dune sand or by loess (Goke and others,
common during deposition of the Mousel terrace 1926; Bacon and others, 1939). Because of seasonal
deposits and the upper Recent alluvium than during shifts in wind direction, silt from the loess-mantled
deposition of the Stockville terrace deposits. A areas has become mixed with sand from the dune
greater incidence of grass fires might reflect a drier areas.
climate, or it might reflect a greater use of fire
In general, the dune sand does not seem to have
drives as a method of hunting game. The extent to
transgressed very much over the loess-mantled areas
which fire drives were used by prehistoric man on
since the end of deposition of the Peorian that is,
the Great Plains is largely speculative (Wedel,
1961, p. 76). within the past 12,000 years. In areas bordering
the Sandhills, the surface of the Wellfleet terrace is
SAND DUNES generally free of windblown sand.
The sand dunes within the Medicine Creek basin ARCHEOLOGY AND HUMAN OCCUPATION
are part of a much larger dune area that is an outlier RESULTS OF ARCHEOLOGIC INVESTIGATIONS
of the Sandhills of Nebraska. The geomorphology of
the Sandhills has been briefly described by Smith Archeologic investigations have been made in the
(1955). In general, the dune relief in the Medicine Medicine Creek basin by the Nebraska State His-
Creek basin is irregular and hummocky, dominated torical Society, by the Nebraska State Museum, and
by innumerable blowouts, most of which are stabil- by the River Basin Surveys of the Smithsonian
ized by grass. However, an indistinct linear pattern Institution. Three cultural aspects have been dis-
was observed in aerial photographs of areas about tinguished. These are represented by the Early
7 miles north and 10 miles northwest of Wellfleet. Lithic sites, which are buried about 35 feet beneath
The pattern is made by low, discontinuous longi- the surface of the Stockville terrace (Terrace-2) ;
tudinal dunes in subparallel arrangement. The dunes by Woodland sites on the Mousel terrace (Terrace-
1) ; and by Upper Republican sites on the Stockville adjacent to upland areas from which colluvium
terrace. might be readily derived, and no fluvial sedimentary
Personnel of the University of Nebraska State structures were observed. Probably most of the silt
Museum have excavated two sites on Lime Creek, was deposited on both cultural sites by the wind
designated Ft-41 and Ft-42, and one site on Medi- after the passing of the Upper Republican Aspect
cine Creek about a half a mile downstream from the that is, during the past 500 years. Conditions under
mouth of Lime Creek (Ft-50). All these sites are which the silt was deposited were of more than local
in the lower part of the Stockville terrace (Terrace- magnitude, as the silt mantle appears on Upper
2) at occupation levels marked by abundant artifacts Republican sites throughout the Republican River
and many hearths. No human remains were found. basin in southern Nebraska from Frontier County
Schultz and others (1951), who investigated both eastward to Webster County (Wedel, 1941). Prob-
the geologic and archeologic aspects of the sites, ably the Upper Republican people inhabited the
report dates of 9,167±600 years B.P. and 9,880± region during a wet period and emigrated about
670 years B.P. from two charcoal samples taken 500 years ago at the onset of a dry period. During
from the bone-artifact zone of site Ft-41. A char- the wet period a humic zone developed, which was
coal sample from the lower occupation zone of Ft-50 buried by windblown silt accumulated during one
yielded a date of 10,493±1,500 years B.P. or more later dry periods.
Two Woodland sites in the Medicine Creek Dam
area are described by Kivett (1949). Habitation SETTLEMENT AND LAND USE
areas (marked by shallow basins, post molds, shallow One of the first settlers of the basin came to the
pits, flint tools, and scant pottery fragments) were Stockville area in 1860. His dwelling was used for
on the Mousel terrace (Terrace-1) and were covered the formal organization of Frontier County in 1872,
by about 18 inches of silt. The remains give Wedel when the population of the county consisted of a
(1949) the impression of a rather uncertain hold few stockraisers and only two permanent settlers.
on the region by a culturally simple group. Accord- Through the later 1870's, settlers gradually entered
ing to Kivett, calcite-tempered ware found at the Frontier County, and a little village arose at Stock-
site is a marker for the Keith focus, which is one ville. By 1880 farmers had begun to claim; land and
of several western Woodland cultural complexes. to settle on the divides. Until the Free Range Law
Although no radiocarbon dates were obtained at the was repealed in 1885, farming on the uplands was
Woodland sites on Medicine Creek, a radiocarbon of little consequence. By 1900 most of the desirable
date of 1,343±240 years B.P. has been obtained land had been taken under the Homestead, Timber
from a Keith-focus site on Prairie Dog Creek in Claim, and Preemption Acts.
northern Kansas, about 60 miles southeast of the Bacon and others (1939) note that corn has beeri
Medicine Creek sites (Wedel and Kivett, 1956). the main crop since farming began. The first'settlers
The Upper Republican Aspect represents the latest did not understand how to adjust their farming
wholly prehistoric culture on Medicine Creek and is methods to the highly variable precipitation, and
the most abundantly represented by archeologic re- their practices were crude and wasteful. A series
mains, which are on the Stockville terrace and are of dry years, accompanied by plagues of grasshop-
buried under a silt mantle 6 to 18 inches thick pers, culminated in the disastrous droughts of 1893
(Kivett, 1949). The date of the culture is estimated and 1894. Many of the early farmers were forced
to be about 500 to 600 years B.P. (Kivett, oral to leave the region, and agricultural development
commun., 1957). was delayed. The farmers who remained acquired
The geologic significance of these archeologic re- larger holdings and gradually adjusted their farming
searches may now be considered. Radiocarbon dates methods to local conditions. Although corn remained
on charcoal from occupation levels in the lower part the most important crop, increasing amounts of
of the Stockville terrace indicate that deposition of wheat, oats, rye, barley, and sorghums were grown;
the Stockville terrace deposits began about 10,000 Census figures have not been published for the
years ago. If the Woodland sites on the Mousel ter- basin as a unit, but satisfactory approximations df
race do indeed belong to the Keith focus and if this the population may be made from the precinct
focus is correctly dated, then the terrace is older census. The total population was 7,750 in 1930,
than 1,300 years. The silt mantle ranging in thick- 6,400 in 1940, and 5,900 in 1950. The total popular
ness from 6 to 18 inches covers Woodland sites on tion of the basin decreased by 23 percent between
the Mousel terrace and Upper Republican sites on 1930 and 1950, and the proportion of persons living
the Stockville terrace. Most of the sites are not in the towns has increased from 42 percent in 1930
to 51 percent in 1950. Population density of the There is no evidence of extension of the drainage
basin in 1950 was 8.7 persons per square mile. system between the end of Sangamon time and the
The number of settlers and domestic animals in initial deposition of loess of the Peorian, although
the basin before 1875 was too small to have any the valleys may have been incised. After about half
significant effect on erosion and deposition. The of the total thickness of the Peorian had accumulated
sparseness of the grass cover on the uplands during on the uplands, the valleys were incised to bedrock.
1869-72 reported by surveyors cannot be attributed The incision was followed by an episode of deposition
to grazing by livestock. According to the Federal during which the upper part of the Peorian accumu-
census, cattle and horses in Frontier County in- lated in the valleys and on the uplands.
creased from about 30,000 in 1890 to about 43,000 The landscape as it existed at the end of deposition
in 1935. Farming could have had little geologic of the Peorian can be reconstructed (fig. 194) from
importance before 1879, when, according to the remnants such as those shown in figure 195 (upper) .
Federal census, only about 600 acres in Frontier In a few places, as for example at the eastern tip
County had been plowed. Only about 104,000 acres of Dry Creek, the valley heads of Peorian time have
of plowed land was reported in Frontier County in been preserved. Deposition of the Peorian Loess
1889, whereas about 425,000 acres was reported in and development of the Brady soil are tentatively
1929. assigned to the interval 60,000 to 12,000 years B.P.
The drainage system was developed to approxi-
DRAINAGE SYSTEM mately its present pattern and extent during the
EVOLUTION OF THE DRAINAGE SYSTEM episode of incision that followed deposition of the
According to the stratigraphic evidence presented Peorian (fig. 1945). The incision is perhaps asso-
in this report, major valleys of the Medicine Creek ciated with a rapid retreat of late Wisconsin ice
basin, such as Cut Canyon and Cedar Creek, were sheets and a rather abrupt climatic warming about
in existence during- the Kansan Glaciation, and their 11,000 years B.P., the evidence of which is presented
bedrock floors were at about the same depth as at by Broecker and others (1960). Valleys were prob-
present. In the Republican River valley also, the ably incised to about the same bedrock altitudes as
bedrock floor has not been deepened since Kansan during previous erosional episodes, but drainage
time according to a geologic section of the river was gradually increased. The extent of incision and
valley near McCook, Nebr. (Bradley and Johnson, increase in drainage density is strikingly shown in
1957 pi. 38). Drilling along two traverses between an area just outside the Medicine Creek basin,
the Republican and the Platte Valleys (figs. 183 and where the head of North Plum Creek was captured
184) revealed buried valleys, cut into the Ogallala by Deer Creek. Although the Deer Creek drainage
Formation and filled with pre-Wisconsin deposits. was incised about 100 feet below the surface of the
These valleys, which evidently had an eastward Peorian, the North Plum Creek drainage just east
trend, were probably filled with alluvium and aban- of the point of capture was not incised, and the
doned in Nebraskan or early Kansan time. Plum surface of the Peorian has been preserved almost
Creek, which flows eastward into the Platte River, intact. (See fig. 196.)
is probably a relic of the early Pleistocene drainage. The lengthy duration of the episode during which
Deposits of Pearlette Ash Member of the Sappa For- the Stockville terrace deposits accumulated is indi-
mation of Kansan age are distributed along the cated by extensive grading of valley-side slopes and
valley of Plum Creek and its former westward ex- valley heads to the level of the Stockville terrace.
tension, which was captured by Deer Creek after In profile, the valley-side slopes that join the Stock-
deposition of the Peorian. ville terrace to the upland are straight for most of
The valleys of most major tributaries (fifth and their length, gently convex upward at their inter-
higher order) were established by incision and prob- section with the upland and gently concave upward
ably by extension of the drainage system during at their intersection with the terrace. Most of the
the episode of incision that followed deposition of slopes are graded across the Peorian, but some are
the Sappa Formation. These incised valleys were graded across older rock units, including the Ogal-
subsequently filled with the Loveland Formation to lala. Accumulation of the Stockville terrace deposits
levels that are generally 10 or 20 feet above the and grading of slopes are assigned tentatively to
modern valley flats. Most valleys of lower order the interval between the climatic warming of about
than fifth were probably not in existence in Sanga- 11,000 years B.P. and the middle of the Altithermal,
mon time, for the Sangamon soil does not dip toward about 5,000 B.P. The Stockville terrace and slopes
them. graded to the terrace can be identified nearly every-
- :""K'u**
-' .' ' iV"-'**
FIGURE 194. Evolution of the relief since deposition of the Peorian Loess. A, Valleys were broad and valley sides gently sloping at end of
deposition of the Peorian, about 12,000 years B.P. B, Drainage system was incised, extended, and branched during episode that lasted from
about 12,000 to 11,000 years B.P. C, Valley sides and valley heads were graded during accumulation of Stockville terrace deposits, from about
11,000 to 5,000 years B.P. D, Relief was modified during the several episodes of incision and deposition that occurred during the past 5,000 years.
where in the basin and constitute a reference surface

to which the incision of later erosional episodes can
be related.
The Stockville terrace was formed at some time
before 2,200 years_B.P., which is the carbon-14 date
obtained from Mousel terrace deposits on Dry Creek.
Tentatively, the Stockville terracing is placed at
about 5,000 years B.P., and the cause is assigned
to drought during the Altithermal, which was a
well-established warm-dry climatic episode that
lasted from about 6,000 to 4,000 B.P. In minor
valleys (of fifth and lower order) the effects of this
terracing cannot generally be distinguished from
the effects of the Mousel terracing. During one or
the other of these two episodes of erosion, trenching
extended to the heads of many minor valleys. Other
minor valleys were trenched for only part of their
length, and still others escaped being trenched.
Deposition of the Mousel terrace deposits began
at some time before 2,200 years B.P. and ended at
some time before 420 years B.P. (carbon-14 date
from late Recent alluvium on Elkhorn Canyon).
Tentatively, this deposition is assigned to the inter-
val 4,000 to 1,000 years B.P., which corresponds
roughly with the dates of the "Deposition 2" de-
scribed by Miller (1958, p. 38) in his generalized
alluvial sequence in Western United States. (See
table 2.) The terracing of the Mousel terrace must
have been brief, inasmuch as the thickness and prop-
erties of the late Recent alluvium indicate that not
much less than a thousand years would be required
for its accumulation.
A tentative chronology of post-Sangamon deposi-
tional and erosional episodes in the Medicine Creek
basin is summarized below:
Tears B.P.
Local incision of valley bottoms ______. __.__ Present to 500.
Accumulation of late Recent alluvium _.______ Present to 900.
Terracing of Mousel terrace . ___ ___ 900 to 1,000.
Accumulation of deposits of Mousel terrace 1,000 to 4,000.
Terracing of Stockville terrace .__ ______._.____ 4,000 to 5,000.
Accumulation of deposits of Stockville ter-
race. 5,000 to 11,000.
Incision and extension of the drainage sys-
FIGURE 195. Remnants of the side, slopes and valley flats of former drainage
systems. Upper, Looking south from the Platte-Republican drainage tem to approximately its present pattern. 11,000 to 12,000.
divide (in the NE^ sec. 31, T. 11 N., R. 28 W.) along a fourth-order Accumulation of Peorian Loess and develop-
tributary to Fox Creek. Remnants of the valley sides of Peorian time are ment of Brady soil., 12,000 to 60,000.
preserved on the divides between side tributaries, and these remnants show
the general aspect of the former valley. Middle, Looking north along MORPHOMETRY
Well Canyon in the SW% sec. 27, T. 10 N., R.. 29 W. Gentle slope on hori-
zon (left) is slope of valley side of Peorian time. Steeper converging Detailed measurements were made of the drainage
slopes on horizon (left center) are slopes of valley side of Stockville time.
Remnant of the valley flat of Stockville time is in middle distance, and
basins of six major tributaries of the Medicine Creek
below this is a remnant of the valley flat of Mousel time. Narrow trench basin, the locations of which are shown in figure
on modern valley flat showed little or no change from 1937 to 1952. Lower,
Looking southeast from the upper end of a fifth-order tributary to Cut
176. The purpose of the measurements was to de-
Canyon, in the NEVi sec. 26, T. 10 N., R. 29 W. Crest of interstream termine the drainage-basin characteristics that cor-
divides between side tributaries in background indicate approximate level
of valley flat of Peorian time. Valley head in foreground was graded
relate most closely with water and sediment yield
in Stockville time, as were the steep side slopes leading to it. and with the development of gullies. Where gaging
100° 30'
40°35'20'
T.9N.
FIGURE 196. Capture of head of North Plum Creek by Deer Creek. In inset, which is reproduced from the Stockville NE quadrangle, relict topog-
raphy of the depositional surface of Peorian time is shown at right.
stations (also shown on fig. 176) are not at the Drainage density, D
mouth of a basin, only that part of the basin up-
stream from the station was measured. Relief ratio,
Ordering of the drainage system was carried out
according to the method of Strahler (1952, p. 1,120),
which is a modification of the method of Horton Elongation ratio, Re
r> = diameter of a circle
Re
(1945, p. 281). Strahler's method is not only more
objective and convenient, but it is also more valu- For practical purposes first-order channels in the
able for many hydrologic purposes. A channel seg- Medicine Creek basin are defined as the lowest order
ment of third or higher order obtained by Strahler's of channels represented by V-shaped bends in con-
method is of approximately uniform cross-sectional tour lines on the available topographic maps. For-
area throughout its length, but this is not true of tunately, the accuracy of the topographic maps in
channels obtained by Horton's method. The word representing small drainage channels is excellent,
"channel" is used here in the same sense that the and the scale (1:24,000) and contour interval (10
word "stream" is used by Horton and Strahler in ft) are also favorable for representation of small
connection with ordering. A channel is an established channels. Drainage patterns of several fifth-order
watercourse that may transmit water continuously, basins were plotted both on aerial photographs and
intermittently, or ephemerally. The flow of water on topographic maps, and comparison showed that
in most channels of the Medicine Creek basin is results obtained by the two methods were very
ephemeral, and the term "stream" is inappropriate similar.
for them. Nearly all the first-order channels represented by
Properties of fluvially eroded landforms and rec- V-shaped bends in contour lines are channels that
ommended symbols for these properties are sum- formed immediately before or during accumulation
marized by Strahler (1958, p. 282-283). The prop- of the Stockville terrace deposits, and the valley
erties discussed in this report and the symbols used sides leading to these channels show some degree of
are listed below. Measured properties are those grading. In only a few places has modern gullying
that can be directly measured or counted, and de- proceeded far enough to form channels where no
rived properties are those that cannot be directly channels had existed previously.
measured but must be computed from measurements. Each of the six major subbasins was separately
Measured properties: outlined on the topographic maps, and each subbasin
Total area of drainage basin, A area was measured with a polar planimeter. In
Area of upland in drainage basin, Act addition, the area of upland in each subbasin was
Area of valley system in drainage basin, A* (A* = A
Aa) measured. Upland is defined as undissected remnants
Channel order, u (for examples: u-i, first order; Ua, sec- of the land surface that existed at the end of deposi-
ond order) tion of the Peorian. For practical purposes upland
Number of channels of order u, Na is distinguished on topographic maps by smooth
Channel length, L contour lines that is, by contour lines that do not
Channel length, mean length of segments of order u; Lu
Basin relief, H have V-shaped indentations. Where areas of upland
Basin length, L& are continuous around the periphery of a drainage
Derived properties: system as in basins 1-1', A', 0-4', and C-2' of
Channel slope, Se figure 197 the area of upland was determined by
Valley slope, S, measuring with a polar planimeter the area actually
Bifurcation ratio, Rb occupied by the drainage system and subtracting
this area from total basin area.
For Dry Creek and Lime Creek, every channel of
Channel length ratio, RL
every order was drawn in color on topographic maps,
counted, and measured individually. Sampling pro-
Channel frequency, Fu cedures, similar to those described by Leopold and
Miller (1956, p. 16), were tested for the approxi-
mation of number, mean slope, and mean length of
Adjusted channel frequency, F* first- and second-order channels in the other sub-
basins. These procedures were used for Mitchell
Creek, Well Canyon, Fox Creek, and Brushy Creek.
CONTOUR INTERVALS 10 AND 20 FEET

FIGURE 197. Variations in drainage texture in the Medicine Creek basin. Basins shown are fourth or fifth order.
Two derived properties drainage density and shape, also devised by Schumm, is the elongation
channel frequency commonly are used but are am- ratio, which is derived by dividing the diameter of
biguous for so-called immature or youthful basins a circle having area equal to that of the basin by
in which areas of undissected upland remain. For the basin length, L6.
such basins a distinction must be made between the A measure of mean valley-side slope is difficult to
properties of the drainage system and the properties obtain because the side slopes leading to low-order
of the drainage basin. If a particular drainage sys- channels are steeper than side slopes leading to
tem were transferred without any change in prop- high-order channels. For example, measurements
erties to a larger drainage basin, the values of on topographic maps indicate that slopes leading
drainage density and channel frequency would to first-order channels on the lower part of Dry
change regardless of the constancy of the drainage Creek have a mean angle of about 25°, whereas
system. Drainage density is also ambiguous for slopes leading to the main channel (which is seventh
another reason: a drainage system consisting of order) have a mean value of about 12°. Slopes lead-
short channel segments may have the same total ing to the main channel of lower Dry Creek were
channel length as another drainage system that measured in the field during the preparation of 12
consists of fewer but longer segments. Melton transverse valley profiles. Of 23 slopes, 13 ranged
(1958, p. 36) has proposed that the ratio of F/D2 from 8 to 11°, only 2 were greater than 15°, and
is constant for mature drainage basins, but this is none were less than 6°. The mean angle was 11.5°;
not true of Medicine Creek, which for the most the mean deviation, 3.3°; and the mode, 10°. Mean
part is immature. To compare the channel frequency side slopes of third- and fourth-order channels prob-
of two immature drainage systems whose drainage ably represent the best approximation to mean
basins differ in percentage of upland, drainage basin values for the drainage system as a whole and are,
area must be adjusted by subtracting the area of therefore, quoted in table 3.
upland. Channel frequency, expressed as number
of channels per square mile of area actually occupied DRAINAGE TRANSFORMATION AND
by the valley system, is here called adjusted channel VARIATIONS IN DRAINAGE TEXTURE
frequency. Variations in drainage texture are accompanied by
Relief ratio was defined by Schumm (1956, p. 612) variations in most drainage-basin characteristics.
as the ratio between H, basin relief, and L6, the The major variations in drainage texture within the
longest dimension of the basin as measured parallel Medicine Creek basin are illustrated by the repre-
to the principal drainage channel. Basin relief is sentative fourth- and fifth-order drainage basins in
the difference in altitude between the highest and figure 20. Basins 0-2' and K-l' are typical of the
lowest points in the basin. A measure of basin upper part of Medicine Creek; 1-1' and 0-4', of the
TABLE 3. Measured and derived properties for the Medicine Creek basin and its major subbasins
First-order channels
Valley Mean
Area of Area of slope of valley Mean Relief Elonga- Adjusted
Drainage basin Total upland, valley main side upland ratio, tion Mean Mean freaueney,
area, A Aa flat segment slope slope Rh ratio, length, slope, F*u
(sq mi) (percent) (percent) (ft per ft) (ft per ft) (ft per ft) Re L Sc (channels
(ft) (ft per ft) per sq
mi)
11.6 37.0 14.0 0.179 0.026 0.0114 0.662 410 0.110 140
Mitchell Creek__ ... ..... ._ 52.1 51.8 11.5 0.00299 .200 .013 .0050 .416 370 .100 165
Brushy Creek
73.8 25.9 16.0 .00367 .412 .022 .0064 .734 192 .234 260
Dry Creek (above gage) :
Total______ _.. ___ .. 21.1 34.8 14.2 .360 .032 .0079 .475 267 .187 203
12.6 25.0 .00550 260 .195 214
Lower only 8.5 49.5 .00366 287 .160 185
Fox Creek:
Total.... ___ .. . __ . ._ 72.5 16.5 27.2 .00360 .363 .037 .0059 .445 250 .241 250
57.0 12.6 .00386
Lower only 15.5 30.4 .00310
Well Canyon:
Total ___ ....... ___ .. ___ 53.3 20.4 22.6 .00286 _______ .036 .0052 .335 245 .251 235
Upper only ____ 22.7 12.3 .00290 260
30.6 26.5 .00284 220
Medicine Creek above
549.0 30.0 .00435 200
500,000
central part; and A' and C-2', of the lower part.
Major differences in these small basins, as in the
larger subbasins of which they are representative,
100,000 -
lie in the percentage of upland, the length and fre-
quency of first-order channels, and relief ratio.
These differences can be conveniently related to
differences in geomorphic history.
10,000 -
In basins A' and C-2' valleys at the end of Peorian
deposition were broad and shallow, and the gently
sloping valley sides were graded all the way to the
interstream divides. Divides were rounded, and no
1000 -
areas of flat upland remained. These valleys were
deeply incised during the episode of erosion that
preceded deposition of the Stockville terrace de-
posits, and channel frequency was increased. During
accumulation of the Stockville terrace deposits, the 100
CHANNEL ORDER
transformed drainage system was smoothly graded,
FIGURE 198. Mean channel length in relation to channel order.
and the gently sloping tips of first-order tributaries
were extended nearly to the former drainage divides.
Basin C-2' represents a less advanced stage of In summary, variations in the modern drainage
grading than Basin A', which is nearer the mouth pattern are related to two conditions that existed at
of Medicine Creek; it was incised first, and conse- the end of deposition of the Peorian the extent of
quently underwent a longer period of grading dur- drainage transformation that preceded deposition
ing accumulation of the Stockville terrace deposits. of the Stockville terrace deposits and the amount of
Basins 1-1' and 0-4' are representative of the grading that took place during deposition of the
central part of the Medicine Creek basin, where the Stockville terrace deposits. To a minor extent, the
valleys at the end of deposition of the Peorian were drainage pattern has been modified by episodes of
more narrow than those in the lower part and were incision that followed deposition of the Stockville
separated by broad, nearly flat uplands. Later, terrace deposits. At the end of Peorian deposition,
these valleys were deeply incised, and the drainage valleys were broad and shallow in the lower part of
was extended into the upland; however, large areas the basin, narrow and separated by flat divides in
were left undissected. The sides and heads of first- the central part, and narrow but separated by nar-
and second-order valleys were smoothed and some- row divides in the upper part. Grading during
what reduced in angle during deposition of the deposition of the Stockville terrace deposits reached
Stockville terrace deposits, but they remained rather an advanced stage in the lower part of the basin,
steep. The surface into which basin 0-4' was incised and in general the stage of grading decreases in an
is unusually flat, because it represents the valley upstream direction.
flat and gently sloping valley side of Medicine Creek The curves representing the relation of mean chan-
at the end of deposition of the Peorian. nel length to channel order (fig. 198) can be ar-
In basin 0-2', valleys at the end of deposition of ranged in a sequence that illustrates some of the
the Peorian were narrow and steep sided (average major variations in drainage texture. The length
slope angle about 14°). The channel frequency, al- of channel segments (of orders one through five)
ready considerably higher than that in the lower shows a consistent decrease from Lime Creek, in
part of Medicine Creek, was greatly increased dur- the lower part of the basin, to Brushy Creek, in the
ing the episode of incision preceding deposition of central part. Also, RL between orders one and two,
the Stockville terrace deposits. Moreover, the newly as well as between orders two and three, is less than
incised valleys, although somewhat graded during RL between higher orders. The change in slope at
Stockville time, retained their steep sides and heads. the lower end of the curves applies only to orders
Later episodes of incision have further increased the one and two for Lime Creek, but it applies to orders
frequency of first- and second-order channels. Basin one through four for Dry Creek and to orders one
K-l' is similar in history, but it formed on a less through three for Well Canyon, Fox Creek, and
steeply sloping surface and underwent a greater Brushy Creek.
degree of grading during the deposition of the An upward concavity of the curves representing
Stockville terrace deposits. the relation of mean channel length to channel order,
fourth order. In a sense the change in slope of the

curve relating mean channel length to channel order
is a measure of disequilibrium of the drainage
system.
GULLY EROSION
DEFINITION AND CLASSIFICATION OF GULLIES
In a region such as the Medicine Creek basin,
where many of the drainage channels have steep,
unvegetated sides and are incised in massive silt,
the distinction of a gully from other drainage chan-
nels is difficult. According to general usage, the
essential features of a gully are its size (which is
larger than a rill), recency of extension in length,
steepness of sides and head, incision into unconsoli-
dated materials, and ephemeral transmission of flow.
Gullies in the Medicine Creek basin are incised into
the sides, head, or bottom of previously established
CHANNEL ORDER drainageways, which are deepened or extended by
FIGURE 199. Number of channels of each order in relation to channel order. development of the gully. The lower limit of depth
for a gully is placed at about 2 feet because channel
as plotted on semilogarithmic paper with channel head scarps that are less than 2 feet in height show
order as independent variable, was noted by Strahler a very slow rate of advancement. Most of the actively
(1952) and Broscoe (1959, p. 5). Broscoe concluded advancing head scarps in the Medicine Creek basin
that Horton's (1945) "law of stream lengths" may are greater than 6 feet in height. The lower limit
not apply when Strahler's method of ordering the of width is arbitrarily placed at about 1 foot. The
drainage system is used. However, the basins ana- criterion of recency of extension requires that some
lyzed by Strahler and Broscoe were of fourth or evidence be obtained for extension within a period
smaller order; basins of larger order must be ana- of a few years. Extension can be established by com-
lyzed for proper definition of the relation. In the parison of aerial photographs, by comparison of
Medicine Creek basin the relation is exponential, field measurements made at intervals of a few years,
but the exponent changes at second or third order or by indirect evidence observed in the field. If the
(fig. 21). sides and head of a drainage channel have a slope
If a drainage system having low density of chan- less than about 45°, the channel is probably not ac-
nels is transformed by the growth of new first- and tively advancing and hence would not be considered
second-order channels without extension of the a gully. In this report the following definition of a
higher order channels, a marked absolute shortening gully is followed: A gully is a recently extended
of lower order channels can take place with- drainage channel that transmits ephemeral flow, has
out much shortening of higher order channels. The steep sides, a steeply sloping or vertical head scarp,
order of higher order channels may change, but the a width greater than about 1 foot, and a depth
absolute length remains nearly constant. A trans- greater than about 2 feet.
formation of this kind took place in the Medicine
The sides and heads of gullies in the Medicine
Creek drainage basin at the end of deposition of the
Peorian. Furthermore, the transformation can take Creek basin are steep slopes, straight in profile,
place without much change in the bifurcation ratio. that are here called erosional scarps or simply scarps.
The bifurcation ratios of the same subbasin, as well In addition to the erosional scarps at the sides
from one subbasin to another, are remarkably con- and heads of gullies, other erosional scarps are
stant. (See fig. 199.) One result of a transforma- conspicuous elements of the landscape. These include
tion of the kind described is a change in slope, at step scarps on slopes (Brice, 1958), scarps that form
lower channel orders, of the curve relating mean the fronts of terraces, and scarps at the base of
channel length to channel order. The major sub- valley-side slopes. The inclination of a scarp ranges
basins that have undergone the greatest drainage from about 45° to nearly 90°, and the height ranges
transformation and, consequently, have the highest from a foot to about 40 feet. Generally, the top of
frequency of first-order channels are those that a scarp meets the surface into which it is cut at a
show the change in slope extending to third or even sharp angle, but the base of the scarp decreases in
slope as it merges with the surface below. The term Creek basin are considered to be younger than the
"channel scarp" is appropriate for a. scarp that Stockville terrace deposits. Some of the slopes
forms a break in the long profile of a well-defined graded to the level of the Stockville terrace are
channel. If a particular channel scarp is the head- steep, especially those in the upper parts of the
ward terminus of a gully and attention is to be drainage basin; but these slopes are not regarded
drawn to this fact, the channel scarp may be called as scarps because they have an inclination less than
a head scarp. Not all gully head scarps are channel 45°. Also, the steepest part of these slopes (in a
scarps. The term "head cut," although commonly particular area of the basin) is graded to about the
applied to the headward terminus of a gully, is un- same angle, and the upper parts of the slopes de-
satisfactory because it may refer either to the scarp crease in angle as they merge with the older Peorian
at the head of a gully to to the whole gully head. surfaces.
Moreover, a channel may have many scarps along its Regardless of the age of the surface it cuts, a
length, and the designation of several of these as scarp may be either active or inactive. A scarp is
head cuts is confusing. The scarps that form the considered to be active if it has migrated a distance
sides of a gully along its length are here called side that can be measured by comparing of aerial photo-
scarps. graphs made in 1937 and in 1952 or if it shows field
Study of gullies in the field and on aerial photo- evidence of recent movement. In the field, inferences
graphs of the Medicine Creek basin has shown that as to activity were made from the type of vegetation
the depth of a gully, its areal pattern, and its rate on the scarp, from the degree of inclination, or from
of growth are more closely related to the topographic evidence of recent or imminent slumping. A thick
position of the gully head than to any other single sod of native grass indicates an inactive scarp; a
factor. Of particular significance is the location cover of brush (such as wild rose, buckbrush, or
of the gully head in relation to the previously estab- currant) indicates a moderately active scarp; and
lished drainage system. On the basis of location, lack of vegetal cover, or a cover of weeds, indicates
gullies are classified as valley-bottom gullies, valley- an active scarp. In general, the more nearly vertical
head gullies, and valley-side gullies. Inasmuch as a scarp, the more likely it is to be active. However,
valley bottoms grade smoothly into valley heads and many scarps that were bare and nearly vertical
valley sides, the distinction among the different showed little or no migration during the interval
classes of gullies is arbitrary. Moreover, a valley- 1937-52. Even a bare and nearly vertical scarp was
bottom gully becomes a valley-head gully as its head not regarded as active unless recently slumped silt
scarp migrates into the valley head. The valley-head was observed at the scarp base or unless fissures were
gullies are by far the most numerous kind, and observed on the ground surface beyond the scarp.
nearly all of these are in steep valley heads that
border areas of upland. CHANNEL SCARPS AND VALLEY-BOTTOM
The size of a gully depends on its depth and areal GULLIES
dimensions; but because of the irregular shape of Varieties of valley-bottom gullies are distinguished
gullies, the areal dimensions cannot be expressed on the basis of position with respect to one another
simply and consistently. For complexly branching and to different topographic levels produced by
gullies, it is not clear whether or not the uncon- trenching of the drainage system. The main va-
sumed area between the branches should be con- rieties are indicated by numerals in figure 200. The
sidered part of the gully. The best approximation two gullies at upper left (1), which are in the same
to an expression of gully size seems to be the maxi- valley bottom but are separated by a reach of un-
mum width of the gully head. Inasmuch as gully dissected valley, are of the sort called discontinuous
depth increases roughly in proportion to width of by Leopold and Miller (1956, p. 29). When the
gully head, the width of the gully head is an indirect head scarp of a downvalley gully reaches the tail of
expression of gully depth. In an arbitrary ranking an upvalley gully, a stage of coalescence is reached;
of gullies according to width of gully head, valley- and the two gullies are integrated into a single
side and valley-bottom gullies must be considered trench as the downvalley scarp advances. Similarly,
separately from valley-head gullies, which are rela- the advance of a major scarp in a large valley is
tively wider in proportion to depth. commonly preceded by the advance of a minor scarp
(2). A scarp may be initiated on the floor of a
AGE AND ACTIVITY OF EROSIONAL SCARPS trench, where it may advance either into freshly
Obviously, a scarp is younger than the surface deposited sediment (3) or into previously undis-
into which it is cut. All the scarps in the Medicine turbed alluvium. Channel scarps on the floor of a
^/////T/T^^^^^^'.^-^-^^^r
'^rtifei^f^fc
_ " *7~ T- -« _ VXy'- '*!.>'' >^ ^ "^ ^*". m*?*?2r''Zl
:§< îrnw'X- ?^^^$$*
^^S^S^^^^-' ">s
^^^ffejj^^^J^. W L^T~ "" " >^^^
l ...
^^V^
:
FIGURE 200. Composite sketch, based on field sketches and photographs, showing the different varieties of valley-bottom gullies. Actively eroding
scarps are indicated by darker shading. Numbered features are described in text.
trench are here described as inset. Remnants of the Scarps are most numerous in the central and lower
original gully floor, left along the gully sides after part of the basin, and scarps in large valleys are
passage of an inset scarp, constitute an inset terrace commonly upstream from the confluence of a large
(4). Another variety of valley-bottom gully is illus- tributary.
trated by the two gullies (5) in the middle ground of The greater abundance of the scarps in the central
figure 200. Unlike the discontinuous gullies, which and lower part of the basin is attributed to the
have no relation to the topographic level formed by relative narrowness of the valley bottoms there.
trenching elsewhere in the drainage system, these (See table 4.) In general, the width of a valley
gullies were initiated at the side scarps of the gully bottom is much affected by the degree of trenching
in the foreground and are accordant with its floor. both before and after deposition of the Stockville
The location of the head scarps of major valley- terrace deposits. The branching valley at upper left
bottom gullies in the Medicine Creek basin is shown (6) in figure 200 was represented by a relatively
in figure 201. Most of the gullies are in the size narrow and shallow trench before deposition of the
category of. very large, but some are in the category Stockville terrace deposits. During deposition of
of large. The scarps range in height from 10 to 25 the Stockville terrace deposits the valley sides were
feet and are in valleys of fourth or higher order. graded, and the valley has been little affected by
Advance of the scarps during the period 1937-52 post-Stockville trenching. By contrast, the valley
ranges from a maximum of about 3,400 feet (for at upper right (7) was represented by a relatively
gully B4 on fig. 201) to a minimum of about 200 feet. deep and wide trench before deposition of the
Two significant facts relevant to the origin of the Stockville terrace deposits. Post-Stockville trench-
scarps are indicated by their areal distribution: ing has extended to the valley head, and only narrow
EXPLANATION
Heavy transverse
line indicates posi-
tion of head scarp
B8
Numbered gully
See text for description
FIGURE 201. Head scarps of major valley-bottom gullies.

TABLE 4. Drainage system properties, listed according to confluence of a large tributary. (See fig. 202.) On
channel order, for major subbasins of the Curtis Creek Canyon a conspicuous steepening of
Medicine Creek basin the valley profile just upstream from the confluence
of a large tributary is indicated on the topographic
Mean
Chan- Number Mean Mean width Area of map. The steepening is probably due to deposition
Subbasin nel of channel channel of val- valley
order channels length slope ley flat flat of a fan at the mouth of a tributary. This fan ponds
(ft) (ft per ft) (ft) (sq mi)
the drainage in the main valley and leads to deposi-
1 1,020 410 0.1100 25 0.37
tion upstream from the fan. Schumm and Hadley
2 233 670 .0517 50 .28 (1957) have shown that, in small drainage basins
3 42 1,600 .0185 105 .25
4 9 4,110 .0111 120 .16
in eastern Wyoming and northern New Mexico,
5 2 9,200 .00545 150 .10 discontinuous gullies can form on locally steep valley
6 1 26,000 .00384 500 .47
Mitchell Creek 1 3,940 370 .100 35 1.83
reaches.
2 950 537 55 1.00 The proposal that gullies are initiated at a local
3
4
172
31
1,170
4,970
90
140
.68
.77
steepening of the valley slope has as a corollary the
5 6 3,320 160 .11 proposal that local slope in an ungullied valley reach
6 1 114,000 400 1.63 may be adjusted to water discharge. In the absence
Dry Creek................ 1 2,824 267 .187 35 .95
2 725 458 .0673 60 .71 of discharge measurements for most valley reaches,
3 153 1,023 .0343 96 .54 the assumption may be made that the discharge in
4 30 1,816 .0155 125 .24
5 4 8,830 .0088 160 .20 a reach is proportional to the upstream drainage
6 2 8,600 .00576 200 .12 area. The relation of local valley slope to drainage
7 1 25,500 .00296 260 .24
Well Canyon .......... 1 10,310 245 .251 35 3.17 area for reaches along three valleys is shown in
2 2,460 480 70 2.85 figure 203. Cut Canyon and Dry Creek are free
3 545 772 110 1.75
4 104 2,195 140 1.23 of major channel scarps in their uppermost reaches,
5 24 5,375 190 .88 as shown in figure 202, but Coyote Creek has channel
6 1 126,000 545 2.18
Fox Creek . _ .. 1 15,900 250 .241 35 4.98 scarps throughout most of its length. In spite of
2 4,250 460 .100 70 4.90 the fact that the floors of all three valleys are well
3 904 725 .0465 120 2.82
4 177 2,040 .0215 150 1.94
above bedrock, local slope is not constant in a given
5 35 4,085 .0101 190 .97 valley reach several hundred feet in length. The
6 6 14,850 .0055 270 1.71
7 2 32,000 .00406 440 1.01
values of local slope plotted in figure 203 represent
8 1 45,000 .00274 880 1.40 the best approximation obtainable from topographic
Brushy Creek .... . 1 12,400 192 .234 35 2.99
2 3,500 370 60 3.39
maps for mean local slope in a reach bounded by
3 729 674 95 1.67 contours and having a fall of 20 feet.
4 147 2,040 135 1.44
5 29 5,450 165 .93
A general relation between local valley slope and
6 6 12,833 240 .66 drainage area is apparent from figure 203. Slope
orr
7 2 40,500 300 values applying to gullied reaches tend to plot above
the average curve representing this relation, but
remnants of the Stockville terrace remain along some slope values applying to ungullied reaches also
the valley sides. The valley at right in figure 200 plot above it. Aside from chance, factors other than
is typical of the upper part of the Medicine Creek local slope and drainage area seem to be involved
basin; and the valley at left, of the central and in the initiation of gullies, and the most important
lower part. of these is probably valley width. The upstream
The association of major channel scarps with reaches of Coyote Creek have a greater susceptibility
the confluence of a large tributary is notable on to gullying than reaches of Cut Canyon because of
Fox Creek, Cut Canyon, Curtis Creek Canyon, and the relative narrowness of the valley bottom of
Well Canyon. Although the present scarps are up- Coyote Creek.
valley from the tributary confluence, the possibility The main valley of Well Canyon is relatively free
is good that the scarps were initiated at the con- of gullies, and the few gullies present have grown
fluence, where the slope of the main valley may be slowly. The channel scarps in the long profile of
locally steepened. After a valley floor has been Well Canyon (fig. 202) did not advance during the
trenched, its long profile cannot be accurately re- interval 1937-52 by an amount measurable on aerial
constructed ; however, long profiles drawn from con- photographs nor did the intricate small-scale mean-
tours at 10-foot intervals on the topographic maps derings of the trenched channel reaches show any
indicate that the main valley profile is generally, observable change. The aspect of a trenched channel
but not everywhere, steepened upstream from the on Well Canyon is shown in figure 195 (middle).
700 -r-
600 --
500 --
- 400 - -
uT
< 300
O
200 --
100 --
0 -L-
20 15 10
HORIZONTAL SCALE, IN MILES
FIGURE 202. Long profiles of valley and channel for some major tributaries to Medicine Creek. A break in the channel profile represents a channel
scarp, and an arrow on the valley profile indicates the point of confluence of a large tributary. Profiles drawn from topographic maps.
The general slope of the valley profile, although DEVELOPMENT OF VALLEY-BOTTOM GULLIES
somewhat less than that of Dry and Curtis Creeks, Photographs of typical valley-bottom gullies are
is not significantly different from that of Mitchell reproduced in figure 204. After a gully has become
and Fox Creeks. The tributaries received by Well large, neither the exact way in which it was initiated
Canyon are relatively short, however, and the valley nor the point of initiation can be established. How-
of Well Canyon is relatively wide at the point of ever, study of small gullies in various stages of de-
confluence of the two largest tributaries. Upstream velopment indicates that valley-bottom gullies may
from the confluence of the northernmost of these
begin as small depressions scoured by flowing water
tributaries, the valley profile of Well Canyon shows
local steepening and the presence of a channel scarp. on the valley bottom. Breaks in the sod cover such
Neither local steepening nor a channel scarp occurs as might result from an animal burrow, a trail, or
at the confluence of the other tributary. The inac- an excavation by man are likely spots at which
tivity of channel scarps in the main valley of Well scour can take place. The upstream side of a de-
Canyon is attributed to the lack of large tributaries, pression can evolve into a scarp and advance up-
to relatively great valley width, and to the generally valley, meanwhile growing in height. Although most
low valley slope. scarps in major valleys have apparently been ini-
I I I I I I I l_
EXPLANATION
Upper Cut Canyon
0.01
0.001
1 10 100
DRAINAGE AREA, IN SQUARE MILES
FIGURE 203. Local valley slope in relation to drainage area for reaches along upper Cut Canyon, upper Dry Creek, and Coyote Creek.
tiated in a locally steepened valley reach, a scarp, or no cohesive effect. Gully side scarps regress by
once initiated, can continue to advance into reaches the same process. For a distance up to 12 feet from
where the value of local slope is conservative in the edges of a gully, the ground is broken by cracks
relation to drainage area (water discharge). More- marking the boundaries of large blocks that have
over, many gullies occur in valleys of fourth or slumped because of saturation of the silt at their
smaller order in which no evidence of local steepen- base. These blocks will eventually slump into the
ing can be found. The value of the ratio of slope to gully, as illustrated in figure 200. Head scarps can
drainage area that is associated with the initiation be maintained in massive silt whether or not they
and rapid growth of a gully is not sharply defined. are capped by sod or some other resistant layer.
Advance of a head scarp takes place as the mas- Drainage is conveyed to many head scarps by a
sive silt at the base of the scarp becomes saturated narrow channel, the bottom of which is bare of
and disintegrates. In this region the permeability vegetation and is cut below the soil profile. A deep
of undisturbed dry loess in place is only about 0.8 notch is formed by the intersection of this channel
foot per day (Holtz and Gibbs, 1952), but this value with the head scarp. The notch in the head scarp of
has no significance for a vertical face of loess that gully Bl on Dry Creek changed very little between
is immersed in water, as at the edges of a plunge 1953 and 1957.
pool. The loess in contact with the water continually Downstream from a head scarp, long profiles of
sloughs along vertical joints and is removed by the gullies change from year to year; the change de-
turbulent water of the pool. In figure 204 (upper pends on the amount of runoff. V. I. Dvorak (writ-
left) the recently collapsed base of a scarp may be ten commun., 1962) has compiled measurements of
seen at right of plunge pool. The cohesiveness of the long profiles of three major gullies on Dry Creek,
massive silt depends to a considerable extent on a which are designated gullies Bl, B2, and B3 in
cement formed by clay minerals that coat the silicate figure 201. Changes in the long profile of gully
grains, and when wet the clay coatings have little B2 are typical and are represented in figure 205.
Upper left. Gully in main channel of Dry Creek, designated gully Bl in

figure 201. Photographed on July 5, 1956. Lower left, Oblique aerial
photograph of gully Bl, September 4, 1965. Upper right, Center, valley-
bottom gully below large valley-head gully, which is designated by numeral
3 on figure 214. Note cowpath erosion on side slope of valley. Oblique
aerial photograph, September 4, 1965. Lower right, Discontinuous valley-
bottom gullies of a tributary to East Curtis Creek Canyon, SE% sec. 31.
T. 10 N., R. 27 W. Oblique aerial photograph, September 4, 1965.
FIGURE 204. TYPICAL VALLEY-BOTTOM GULLIES
According to Dvorak, the relatively high runoff in The cross profile of a valley-bottom gully changes
1951 was accompanied by gully slopes distinctly less in a direction downstream from the head scarp, but
than valley slopes, but during the succeeding drier the nature of the change is not consistent from gully
years the gully slopes increased. Gully slope was to gully. Typically, gully depth decreases very
steeper than valley slope for gully B2 in 1956 and gradually in a downstream direction, whereas gully
for gullies Bl and B3 in 1960. width reaches a maximum within a thousand feet
The maximum height attained by a head scarp is of the head scarp and then decreases in a down-
evidently controlled mainly by the steepness of the stream direction. Downstream changes in the cross
slope into which the head scarp is advancing. The profile of a typical rapidly advancing valley-bottom
highest head scarps in the basin, which reach a gully (gully B4 in fig. 201) are shown in figure 206.
maximum of about 35 feet, are all in steep valley The width of gully B4 remains nearly constant for
heads, whereas head scarps on the gently sloping about 5,000 feet downstream, but for other gullies
valley bottoms all range from 10 to 25 feet in height. the width decreases downstream at a much faster
The height that a scarp can attain is also limited rate than depth. For example, in 1952 the width
by the rate of deposition at, and downstream from, of a large valley-bottom gully on Curtis Canyon
its base. (gully B6) decreased from about 105 feet at a
500 1000 1500 2000

DISTANCE FROM ARBITRARY REFERENCE POINT, IN FEET
FIGURE 205. Changes in long profile of a large valley-bottom gully on Dry Creek, 1951-60. Gully is designated B2 in figure 201. From V. I. Ovorak.
point 900 feet downstream from the head scarp to ratio greater than 6, which was the value measured
about 45 feet at a point 5,000 feet from the head for gully B5 at a point 390 feet downstream from
scarp. Through this 5,000-foot distance, the depth its head scarp. However, none of these gullies had
of the gully changed little. Neither gully B4 nor a drainage area greater than about 16 square miles.
B6 has a well-defined downstream terminus at which Gully B9, the drainage area of which was about
the gully floor merges with the valley floor; each 1.5 square miles, had in 1956 a width-depth ratio
merges downstream with a narrow but apparently that did not exceed a value of 1 for a distance of
stable channel that continues for miles along the 1,000 feet downstream from its head scarp.
valley bottom. The advance of major valley-bottom gullies during
From his study of valley-bottom gullies on Dry the period 1937-52 is given in table 5, together with
Creek, Dvorak (written commun., 1962) concluded pertinent information as to gully size and topo-
that widening downstream from the head scarp graphic setting. Gully advance and width of valley
effectively ceased when the width-depth ratio was bottom were measured on aerial photographs; drain-
in the range of 3.5 to 5. At great width, the depth age area was measured on topographic maps; and
and velocity of flowing water would be insufficient height of head scarp was measured in the field.
to remove material that accumulated from bank Slope of the valley bottom, which applies to the
slumping, and the channel bottom would assume a slope upvalley from the 1952 position of head scarps,
parabolic shape. was measured by field surveys for some of the gullies
As is generally the case with channels in nature, and on topographic maps for others. In general,
the width-depth ratio that can be maintained for a the maximum depth of a gully is about equal to, or is
gully probably depends on the magnitude of water greater than, the height of its head scarp. No cor-
discharge. Small gullies in the Medicine Creek basin relation is apparent between the rate of advance
have low values of the width-depth ratio, and large and any of the other measurements. The estimate
gullies have larger values. In 1956 none of the large of the year at which active advance of the gully
valley-bottom gullies had attained a width-depth began is made by dividing gully length by average
TABLE 5. Measurements relating to the size and topographic low rainfall (fig. 179) and, as observed on aerial
setting of major valley-bottom gullies photographs, most of the gullies were inactive in
[Gully locations are indicated on fig. 201]
1937 and were partly filled with sediment. A sub-
stantial part of the advance during the period 1937-
Width Slope Height Advance Esti-
Drainage of of of head of head mated 52 took place in 1951 as a result of high runoff.
Gully area valley valley scarp scarp, year of
in 1952 bottom bottom in 1956 1937-52 acti- For example, of the 750-foot advance made by gully
(sq mi) (ft) (ft per ft) (ft) (ft) vation
B2 during the period 1937-52, about 350 feet was
Bl 6.1 215 0.0056 24 900 1850 made during 1951. Differences in rate of gully ad-
B2 __.___.__._____ 5.6 210 .0055 22 750 vance (table 5) are probably in part due to unequal
B 3 ___. ___ .8 120 .0105 20 610 ?
B4.. ____ 3.4 140 .0082 16 3,400 1900 distribution of rainfall during 1951.
B5
B6.____ _____
Q
«7. 7
9.8
1 200 .0058 18 2,400 1920 The development of a locally continuous channel
160 .0062 18 1,750 1920
B7._ _______ 15.9 180 .00415 18 2,800 1910 by the coalescence of two discontinuous valley-bot-
B9._... _____
9.5 150 .0050 20 2,100 tom gullies, as described by Leopold and Miller
1.4 120 .0125 23 2,100 1920
(1956, p. 31), is illustrated in figure 207. Only the
annual rate of advance during 1937-52. For Dry upper reach of the gullied valley bottom is shown
in the illustration. The shallow channel shown at
Creek, which has a continuous channel downstream
left continues downvalley for 1,300 feet and termi-
from the head scarp, the point of origin is assumed
nates at the head scarp of gully B3. No evidence of
to be at the confluence of East Fork. downvalley extension of gullies during the period
Valley-bottom gullies advance only during periods 1937-56 was observed at this locality, and in general
of runoff. As a result of rainfall on July 4 and 5, downvalley extension of a gully is uncommon. The
1956, which totaled about 1.9 inches, gully B5 ad- long profile of a single discontinuous gully is rarely
vanced about 15 feet and gully Bl advanced 20 feet. smooth, but it is ordinarily broken by one or more
The year 1937 was preceded by a period of relatively inset scarps.
20 40 60 80 100 120 FEET
VERTICAL AND HORIZONTAL SCALE OF CROSS PROFILES

I-
\ /
N
J-
1000 2000 3000 4000 5000 6000

DISTANCE, IN FEET
FIGURE 206. Long profile and cross profiles of a large valley-bottom gully in a tributary to Curtis Creek Canyon. Gully is designated B4 in figure
201. Based on a field survey, 1956.
160
EXPLANATION
Profile of valley in 1956

120
Profile of channel in 1956
Probable profile of channel in 1937
80
40
4000 3500 3000 2500 2000 1500 1000 500

DISTANCE FROM VALLEY HEAD, IN FEET
FIGURE 207. Changes in long profile of a gullied tributary to Dry Creek, 1937-56. For the long profile in 1937, the position of head scarps was taken
from aerial photographs, and other parts of the profile were estimated. Long and cross profiles in 1956 are based on a field survey.
The gradient of a discontinuous gully is less steep 209 refer to distances in feet from the mouth of
than the gradient of the valley floor into which it isDry Creek, are represented in figure 208.
advancing, and Leopold and Miller attribute this to In general, the upper reaches of Dry Creek are
the narrower width of the gully. As gullies coalesce ungullied (fig. 210, upper left), the middle reaches
and widen, a continuous channel is finally formed (as represented in fig. 210, lower left and upper
that has a gradient nearly parallel to the original right) have been twice gullied in late Recent time,
valley floor. The gullies of the Medicine Creek and the lower reaches (as represented in fig. 210,
basin, however, have not reached this final stage; lower right) were ungullied until about 1937. The
channel scarps are advancing in succession along inset terrace (figs. 209 and 210, lower left) provides
the valleys, and final regrading of a given reach willevidence for the two episodes of gullying. The inset
probably involve the passage of many channel scarps. terrace deposits accumulated in the wake of a major
channel scarp, and the inset terrace was formed
HISTORY OF LATE RECENT GULLYING ON as these deposits were trenched by the advance of a
DRY CREEK
second scarp. The first channel scarp probably began
Dry Creek was selected for detailed study of gully at about station 10,000 for the inset terrace becomes
erosion by agencies involved in the Medicine Creek indistinct at this point. A reasonable date for the
Watershed Investigations because the rate of erosion beginning of this first scarp is about 350 years ago,
there seemed to be more severe than elsewhere in the which is the date suggested (on the basis of a
basin. Periodic surveys were made of selected carbon-14 age determination) for gullying on Elk-
reaches of valley-bottom gullies to provide a basis horn Canyon.
for calculation of gross erosion by these gullies. A The approximate age of the inset terrace deposits
detailed geomorphic study of the Dry Creek channel can be inferred from a cottonwood stump rooted
and terraces was made by the writer, with the on the terrace at station 43,300 (fig. 210, lower left)
objective of deciphering the history of gullying in and from a tobacco tin buried to a depth of 5 feet
late Recent time and relating this history to white in the deposits at station 43,700. The growth rings
settlement and occupation. on the cottonwood stump indicate that it was about
A detailed long profile of Dry Creek, based on a 40 years old at the time of cutting (1950) ; therefore,
field survey made by personnel of the Bureau of the inset terrace deposits had already accumulated
Reclamation in 1951, is shown in figure 208. The in 1910. The tobacco tin, which is about 400 feet up-
terrace profiles in figure 208 and the cross profiles valley from the stump, was perhaps buried about
in figure 209 are also based on this survey, supple- this time or somewhat later. Because the inset
mented with field surveys made by the writer. terrace deposits accumulate in the wake of an ad-
Station references used in the text and in figure vancing head scarp, their age decreases in an upvalley
2900
EXPLANATION
Stockville terrace East Fork Dry Creek
Mousel terrace
Valley flat
Flow line, 1951 channel

1111ii ii
Bedrock outcrops in channel
Position of channel scarp in 1937

VERTICAL SCALE IS EXAGGERATED 100 TIMES
2450
30 40 50 60
CHANNEL LENGTH, IN THOUSANDS OF FEET
FIGURE 208. Long profiles of channel, valley flat, and terraces on Dry Creek.
direction. The channel scarp that is trenching the cultivated upland at the valley head, and to the
deposits and thus forming the inset terrace was at activities of man and livestock. The valley at right
station 37,400 in 1937 and had advanced 8,300 feet has steep sides and head because it was deeply
by 1951. trenched both before and after deposition of the
A new channel scarp began at or near the mouth Stockville terrace deposits. The steep valley head,
of Dry Creek at a time not long before 1937, accord- which receives heavy runoff from a field planted in
ing to its position on the 1937 aerial photographs. row crops, is a favorable place for the formation of
This scarp had reached station 6,000 (fig. 210, lower a gully. A large valley-head gully (gully 3, fig. 211),
right) by 1952, and it is moving rapidly upvalley which began on the upper part of the valley head,
at present. has branched along the fence and into the field. The
step scarps (4) on the steep valley sides began at
VALLEY-HEAD AND VALLEY-SIDE GULLIES animal trails along the slope contour and have grown
Varieties of valley-head and valley-side gullies in height as they slowly migrated upslope. The step
are distinguished mainly on the basis of shape of scarp at extreme right has evolved into a valley-side
gully head in plan view. The main varieties in their gully (5). At the left side of the valley, two narrow
characteristic topographic situations are illustrated valley-side gullies (1) are forming from cowpaths,
in figure 211. A gully head that is on a relatively and another valley-side gully (8) is advancing along
steep slope and receives a concentration of flow a road ditch.
from a single direction, as from a narrow channel, The valley at center is shallow and has a relatively
tends to be narrow and pointed (gully 1, fig. 211). gently sloping head because it was trenched to shal-
If the slope is less steep or the flow less concentrated, low depth before deposition of the Stockville terrace
the gully head will be broadly lobed (2). If, on deposits, and post-Stockville trenching has not ex-
the other hand, the slope is gentle and the flow comes tended far upvalley. In addition, the valley head
from diverse directions, the gully head will be com- borders a divide rather than a tract of upland and,
plexly branching (3). therefore, has a small drainage area. The three
Situations favorable for the initiation of a gully gullies in the valley head (6) are lobed and have
are related to valley shape (as determined by past grown slowly. In the background beyond the center
erosional history of the valley), to the presence of valley, two gullies (7) in another valley head are
STA 30,760
STA_24,210
1 I I I | I I I I I I I I I "NyT" ~ ~~ ~
20 20 40 60 80 100 FEET
VERTICAL AND HORIZONTAL SCALE
EXPLANATION
Stockville terrace Valley flat Inset terrace

Dashed lines represent valley profiles restored
by projection of terrace slopes
FIGURE 209. Cross profiles of Dry Creek. Station numbers are distances in feet from mouth of creek.
Upper left. Upper reach of East Fork Dry Creek, 1953. Narrow incised
channel is shown in middle ground. Lower left, Channel and inset ter-
race at station 43,300 in 1956. Cottonwood stump, near center, is rooted
at outer edge of inset terrace. Upper right, at gaging station (sta.
17,660), August 17, 1953. Discharge about 300 cfs. Lower right. Channel
at station 6,000 in 1953, looking downstream. On downstream side of
cottonwood tree is a large channel scarp, advance of which has since
felled the tree.
FIGUGE 210. DOWNSTREAM CHANGES IN CHANNEL OF DRY CREEK
advancing rapidly one in the direction of drainage the largest in the basin. As is typical, these gullies
from the road and the other in the direction of drain- border areas of cultivated upland. The gully in figure
age from the barnyard. 212 (upper left) extends for about 900 feet along
The two valleys at extreme left are typical of a fence line. A narrow but deep valley-head gully,
many valleys in the upper part of the basin. Severe whose head scarp attains a height of 31 feet, is
post-Stockville trenching has removed most of the shown in figure 212 (lower left). Valley-side gullies
deposits of Stockville age and left the valley sides typical of the sharply dissected upper part of the
raw and steep. The valley-side and valley-head basin are shown along the sides of the three valleys
gullies (2), which are lobed in plan view and have in figure 212 (upper right). Some are moderately
head scarps tens of feet in height, are advancing active and others are inactive. The growth rate of
very slowly because their drainage area is small. such gullies is slow because they border narrow
Photographs of valley-head and valley-side gullies divides and hence have small drainage areas. The
are shown in figure 212. The valley-head gullies in valley-side gullies in figure 212 (lower right) orig-
figure 212 (upper left and upper right) are among inated from cowpaths and an old field road.
- V < *.- jn /W» IJ** I'm V'.IV/ *
FIGURE 211. Composite sketch, based on field sketches and photographs, showing different varieties of valley-head and valley-side gullies. Actively
eroding scarps are indicated by dark shading. Numbered features are described in text.
MEASUREMENT OF VALLEY-HEAD AND tion of volume was complicated by the irregular

VALLEY-SIDE GULLIES ON DRY CREEK shape of most gullies. No volume increases less than
In the summer of 1953 measurement of the volume about 30 cubic yards are reported because increases
increase of valley-head and valley-side gullies during less than this amount are considered to be unrecog-
the period 1937-52 was made by field surveys and nizable by comparison of aerial photographs. Vol-
comparison of aerial photographs. On aerial photo- ume increases, however, were measured for nearly
graphs enlarged to a scale of 1 inch to 660 feet, all the gullies on Dry Creek that seem to be active,
reference lines across or near gully Heads were and erosion from gullies that seem to be inactive is
drawn between fixed points, such as trees, fence probably not quantitatively important.
corners, houses, or other landmarks. The position According to these measurements, the enlargement
of gully head in 1937 was marked on a 1952 aerial of all active valley-head and valley-side gullies on
photograph, which was then taken into the field to
facilitate measurement of the volume increase. Dry Creek (216 gullies) for the period 1937-52
Changes in gully depth could not be determined from totaled 106,500 cubic yards. A plot of the frequency
aerial photographs; but this is probably not a sig- distribution of gully enlargements on logarithmic
nificant source of error, inasmuch as neither height probability paper indicates a logarithmically normal
of head scarp nor gully depth changes very much distribution, and the skewness at the lower end of
during a moderate advance. The horizontal dimen- the curve is attributed to the omission of perhaps
sions of gullies were measured by pacing, and the 10 gully enlargements less than 30 cubic yards.
depths were measured with a steel tape. Determina- (See fig. 213.) The median enlargement is about
FIGURE 212. Valley-head and valley-side gullies. Upper left. Center, large valley-head gully in Curtis Creek subbasin, NE*4 sec. 10. T. 8 N., R.
28 W. Oblique aerial view, September 4, 1965. Lower left. Valley-head gully in Dry Creek subbasin, center sec. 33, T. 9 N., R. 27 W., September,
1953. Upper right, Large valley-head gullies and smaller valley-side gullies in Curtis Creek subbasin, SW */4 sec. 30, T. 10 N., R. 27 W. Oblique
aerial view, September 4, 1965. Lower right, Valley-side gullies in Well Canyon subbasin, SW *4 sec. 13, T. 9 N., R. 29 W.
Geometric mean of
200 cubic yards, and the range is from 28 to about GuUy size Enlargement range, enlargement range
4,200 cubic yards. 1937-52 (cu yd) (cu yd)
A satisfactory approximation to the measured Small _____ 50-100 63

gully enlargement was reached by another method, Medium _____ 100-600 245
which is based on the premise that active gullies Large ___ 600-3,600 1,470
can be recognized on aerial photographs and also Very large 3,600-7,200 5,100
on the premise that small gullies have small volume
increases and large gullies have large volume in- The number of gullies in each category on Dry
creases. Several years after the gully measure- Creek is shown in table 6. By multiplying the num-
ments were made on Dry Creek, active gullies in the ber of gullies in each category by the geometric
Medicine Creek basin were studied on aerial photo- mean of the enlargement, a total enlargement of
graphs and ranked into four categories according 97,300 cubic yards was obtained, which is a reason-
to size (not according to enlargement). On the able approximation to the measured enlargement.
basis of the measured gully enlargements on Dry Although this method is obviously subjective, no
Creek, each size category was assigned an enlarge- means of applying rigorously objective sampling
ment range, according to the following scheme: methods to such irregular objects as gullies is
CO
o
O5
PERCENTAGE ENLARGEMENT PERCENTAGE ENLARGEMENT

vi 01 on £» w ro
§ o o o o o o p
fo p
*-
O
M
8
m o H
2
2{
H
ss
2
ii
25
02
H
§
^
H
2!
H3
TABLE 6. Number of gullies according to type and size in the Medicine Creek basin and
in major subbasins
Number of gullies
gully head Medicine
Gully type Size (range in Mitchell Dry Fox Brushy (subbasins
feet) (subbasin C) (subbasin I) (subbasin K) (subbasin N) A-O)
Valley head ____..._____ Small ___ _____ 1-15 47 70 39 36 417

Medium _______ 16-60 99 110 90 302 1,357
Large _____ _ 61-180 28 35 16 90 360
Very large _ _.___ >180 1 2 _ ______ 3 20
Valley side __ __ _ Small ...... 1-10 22 3 7 10 112
Medium ____ __ ._ 11-25 21 14 17 66 279
Large __ _ __ 26-50 6 2 2 7 44
VPTv 1 s T* &{* >50
Valley bottom __ ___ Small . _______ 1-10 56 38 11 24 563
Medium .___ ______ 11-25 33 27 23 82 477
Large _____ ._ 26-50
Very large ______ >50
} » 10 9 19 152
apparent. In the ranking of gullies into categories, Discontinuous valley-bottom gullies of small size
a particularly difficult subjective decision lies in are indicated by the two arrows at left of the nu-
deciding whether the complex branches of a severely meral 4, and others of medium size are indicated by
eroded valley head should be regarded as one gully arrows in the lower left corner of the photograph.
or several. In the Dry Creek measurements, most Small discontinuous valley-bottom gullies are also
of the branching valley heads were divided into indicated by arrows on the valley flat of Dry Creek.
several gullies, whereas in the later ranking of The major valley-bottom gully on Dry Creek begins
gullies according to size, the branching valley heads about 1 mile downstream (to the right) from this
were regarded as one gully. For this reason the locality. A valley-side gully of medium size, which
geometric mean of the largest size category (5,100 is advancing along a fence, is indicated by the
cu yd) is larger than the largest measurement re- numeral 6.
ported (4,200 cu yd). Among the notable relief features shown on the
The topographic situations of gullies, as well as photograph are the differences among valleys in
most of the topographic features characteristic of flatness of bottom and steepness of side. The valleys
the Medicine Creek basin, are illustrated by the at right of numeral 4 have flat bottoms and steep
vertical aerial photograph reproduced as figure 214. sides because post-Stockville trenching extended to
Sec. 17, T. 9 N., R. 27 W., is approximately in the the valley heads; the valleys at left and below
center of the area, and the valley of Dry Creek crosses numeral 4 have narrow bottoms and gentle sides
it from left to right. Gullies numbered 3 and 5 on because their trenching was incomplete. Also nota-
the photograph are in the very large category and ble, in the field at right center on the photograph,
have complexly branching heads. Total enlargement is a dark round spot that indicates a depression.
of all branches amounted to about 6,950 cubic yards Depressions of this sort, sometimes called buffalo
for gully 3 during the period 1937-52 and about wallows, are common on the Great Plains. Many
5,610 cubic yards for gully 5. Figure 204 (upper hypotheses have been proposed as to itheir origin,
right) is a view of gully 3 as seen from the air. but they probably originate in more th&n one way,
Gullies 1 and 4 are in the large category, and the and no hypothesis has been established as generally
lobed shape of their heads indicates that they have valid. In the Medicine Creek basin, many of the
not advanced beyond the steeper parts of the valley depressions probably mark the positions of sink-
heads, although the head of gully 4 is beginning to holes in the underlying Ogallala Formation.
branch. Enlargement during the period 1937-52
was 1,320 cubic yards for gully 1 and 2,250 cubic AREAL DISTRIBUTION OF GULLIES
yards for gully 4. The group of four gullies indi- Experience in the recognition of Dry Creek gullies
cated by the numeral 2 are in the medium category, on aerial photographs and in the estimation of their
and their enlargements ranged from 107 to 225 size and activity was used in the collection of infor-
cubic yards. mation on the distribution of gullies in the Medicine
FIGURE 214. Vertical aerial photograph of a severely gullied area on upper Dry Creek. Numbers indicate gullies described in text. Arrows indicate
small- and medium-size discontinuous valley-bottom gullies described in text. Photograph by U.S. Commodity Stabilization Service.
Creek basin as a whole. Aerial photographs of the central part (fig. 215). The frequency ratio of
whole basin were studied stereoscopically, and all valley-head and valley-side gullies to valley-bottom
gullies in the basin that seemed to be active were gullies ranges from about 0.6 in the lower part of
plotted on topographic maps. Decisions as to the the basin (subbasin A) to about 6 in the central
activity of a gully were checked by reference to part of the basin (subbasin 0-4). The distribution
1937 aerial photographs. The gullies were grouped of gullies according to type and frequency is con-
according to type (valley head, valley side, or valley trolled mainly by the steepness of valley heads, the
bottom) and according to size (small, medium, large, narrowness of valley bottoms, and the amount of
very large). (See table 6.) upland drained by a valley head. In general, the
A map showing the frequency distribution of steepness of valley heads is directly proportional,
gullies according to subbasin was prepared (fig. and the narrowness of valley bottoms is inversely
215). The basin was divided into 24 subbasins, proportional, to first-order-channel frequency.
corresponding generally to subbasins used by the Gully frequency in relation to percentage of area
Agricultural Research Service in their land-use tab- in upland and to adjusted frequency of first-order
ulations, that are reasonably homogeneous in texture channels is shown in a semiquantitative way in
of relief. The frequencies of valley-head, valley-side, figure 216. Moderate to high frequencies of valley-
and valley-bottom gullies (excepting large and very head and valley-side gullies are associated with
large valley-bottom gullies) in each subbasin are values of first-order-channel frequency and percent-
given in table 7. Gully frequency refers to number age of area in upland that plot to the right of the
of gullies per square mile of valley system rather dashed line. On the other hand, values that plot to
than to number per square mile of drainage area. the left of the dashed line tend to have a higher
Gullies are directly related to the valley system and frequency of small and medium valley-bottom gullies.
are not present in the upland except at its edges. Points representing high gully frequency would
Area of upland and characteristics of the valley probably be more widely separated from points rep-
system should be regarded as independent variables resenting low gully frequency were it not for the
in their relation to gullying. fact that the subbasins are not entirely homogeneous
The frequency of active valley-head and valley- in topographic character. The severity of post-
side gullies is low in the lower and upper part of Stockville trenching not only varies from the upper
the basin and ranges from moderate to high in the to the lower part of the basin, but also from one
minor side tributary to the next (fi£. 214). The point
TABLE 7. Gully frequencies and related data for subbasins of indicated by a gully frequency of 10.8 and repre-
the Medicine Creek basin senting subbasin 0-4, at upper right in figure 216,
is most anomalous. In spite of the relatively low
Fre- Fre-
Fre- nuency quency gully frequency in this subbasin, it includes an area
Drainage Area in Area of quency of valley- of
Sub- area upland vallev of first- head and valley- of exceptionally severe erosion that is about 3 miles
basin (sq mi) (percent) system order vallev-side bottom
(sq mi) channels gullies gullies northeast of Maywood. This area is also exceptional
in that its drainage was greatly extended by post-
A ______ 27.2 45 15.0 113 3.2
2.9
5.3
4.2
Stockville erosion. Several large gullies there made
B-l _ . __ 43.4 38 26.4 150
B-2 _____ 11.6 37 7.3 140 3.7 5.5 advances ranging from 60 to 180 feet during the
C-l _____
C-2 ____ _
22.3
29.8
50
52
10.8
14.4
190
150
16.0
3.5
4.3
3.4
period 1937-52.
D 74.4 45 40.9 160 5.7 2.6
E 10.9 42 6.3 165 5.2 6.1 FORMATION OF VALLEY-HEAD AND
F 58.9 34 35.3 170 6.9 4.1 VALLEY-SIDE GULLIES
G . 23.2 42 13.4 195 12.4 5.0
H ___ __ 16.5 48 7.4 185 11.7 4.2 The head scarps of many valley-head gullies orig-
1-1 _____ 12.6 40 9.5 214 18.5 5.1
inate on the steep upper part of the valley head and
1-2 ______ 8.5 49 4.3 185 14.2 5.3
J-l ____ 17.6 12 15.5 275 3.3 3.0 advance toward the upland. This is indicated by the
16.4 4.6
J-2 _____
K-l._._ ___
22.6
57.0
45
12
12.6
49.8
200
260 1.5 4.2
position of short gullies and by the shape in plan
K-2 _____ 15.5 30 10.8 200 9.1 1.6 view of long gullies. The width of many large valley-
L-l __ _
L-2_ ___ .
,22.7
30.6
12
26
19.9
22.5
260
220
.5
5.2
,2
.9
head gullies decreases sharply toward the lower part
M ______ 27.5 30 19.3 180 6.8 1.6 of the valley head, and the gullies are connected to
N_____
0-1 __ - -
73.8
14.6
26 52.4 260 10.0 2.3 the valley flat by a narrow trench. (See fig. 217
O-2 _____ 15.0 10 13.5 450 .7 .4 and gully 5 in fig. 214.) Gully 5 narrowed down-
O-3 _____ 35.4 12
55
31.1 350 2.1 .4 valley (in 1956) to a particularly deep trench about
O-4_____ 22.0 9.9 350 10.8 1.8
25 feet deep, 15 feet wide at the top, and 5 feet wide
« fill. Ill
Frequency of gullies per square

mile of valley system
FIGURE 215. Areal frequency distribution of active valley-head and valley-side gullies in the Medicine Creek basin. Subbasins illustrated in figure
197 are shown by heavy outlines. See table 7.
60
1 1 I
50
'\ \H.2Q 0
o117 (
O
10.8
A
30.4
\
o52 \ 124o
16.4 A
34.6
40 I
o
3.7 O2 '
.. \ \
\
GULLY FREQUENCY
MODERATE TO HIGH
6.8 \ 9.1
i 30 °\ O
-
GULLY FREQUENCY \ 5.2 10 '.

LOW TO MODERATE \ 0 0
,
23.2
\ A
^
20
4.4
^^ _____ Al ,3
A
EXPLANATION - _______ <
o2-1 "o 83- 3 2.1
10 Subbasin 1.5 1.2

o°'7 -
A1 ' 2
Small basins less than 2
square miles in area i
Number beside symbol refers to gully frequency

A2
per square mile of valley system
0 1 1 1 i
100 150 200 250 300 350 400 450 500
ADJUSTED FREQUENCY'OF FIRST-ORDER CHANNELS
FIGURE 216. Frequency of active valley-head and valley-side gullies in relation to area of upland and adjusted frequency of first-order channels.
at the bottom. Set in the bottom of this trench was Creek, 9 percent; Fox Creek, 6 percent; Lime Creek,
a winding notch, sections of which passed under- 45 percent; and subbasin A, 65 percent. Thus, in
ground. the narrow valleys of the lower part of the basin, a
Other valley-head gullies begin at an upvalley greater percentage of valley-head gullies begins on
riser of the Stockville terrace, such as the one shown the valley bottoms.
in the valley at center in figure 211, or as discon- The head scarps of valley-head gullies reach a
tinuous valley-bottom gullies, such as the ones shown maximum height of about 35 feet, a value about
in the valley at upper left in figure 200. In the tabu- 10 feet higher than the maximum height of head
lation of all active gullies in the basin, valley-head scarp measured for large valley-bottom gullies. This
gullies that are connected by a continuous, rather greater height is evidently related to the steeper
wide trench to the valley flat were given a separate slopes into which the valley-head and valley-side
designation because they probably began on the gullies are advancing and to the correspondingly
valley bottom rather than in the valley head. The steeper slopes at the bsise of their head scarps. (See
percentages of valley-head gullies that fall in this figs. 217 and 218.) Material does not tend to accu-
category are as follows for several subbasins: Dry mulate at the -base of the head scarp as it does in
Creek, 7 percent; Brushy Creek, 22 percent; Mitchell large valley-bottom gullies. In long profile, the head
190
pproximate profile o
Stockville valley fill
100
100 200 300 400 500 600 700 800 900
DISTANCE ALONG GULLY, IN FEET
HORIZONTAL AND VERTICAL SCALE

Cross profiles
FIGURE 217. Long profile, plan view, and cross profiles of a large valley-head gully, designated gully 4 in figure 214. Shaded area on plan view
indicates enlargement during the period 1954-57. Based of field surveys. Location of cross profile H is beyond limit of area shown in plan .view.
scarps are nearly vertical or even overhanging at rim of the head scarp has relatively small volume
the rim, and the channel profile downstream from and velocity, and silt beneath the resistant rim is
the head scarp is typically broken by several channel disintegrated by back trickle of water. Underground
scarps. The channel scarps deepen the gully as they drainage, which probably enters the ground through
advance and probably increase the height of the rodent burrows upslope from the head scarp, is con-
head scarp by merging with it. The modal value for veyed into the gully head at some point below the
height of head scarps of active valley-head and resistant rim. The abundance of rodent burrows
valley-side gullies on Dry Creek is about 10 feet.
intersected by some head scarps, as well as the over-
The mechanism of head-scarp advance is not
hanging of the resistant rim, is shown in figure 219.
identical with that described for large valley-bottom
gullies. Most valley-head gullies are advancing into Head scarps less than 4 feet high do not advance
surfaces underlain by a well-developed soil profile rapidly because the resistance of the soil profile and
and, except for gullies advancing into cultivated the sod cover extend to about this depth. !
upland, protected by a sod > cover. The rills or Material that slumps from the head scarps and
trenches that lead to the gully head have not been side scarps accumulates on the gully floor arid is
incised through this profile; hence, the head scarp gradually removed by runoff. Deposition is uncom-
has a resistant rim. On the other hand, plunge mon in the trench that drains the gully head, but
pools are not conspicuous at the base of the head much deposition takes place where this trench enters
scarps, and saturation of the massive silt by plunge the valley flat. The number of valley-head and
pool action is less important tjian that for large valley-side gullies that discharge directly into the
valley-bottom gullies. The water flowing over the channels of major tributaries is insignificant, and
no
100
1937 profile of gully (approximate)
100 200 300 400 500 600 700 800 900

DISTANCE, IN FEET
FIGURE 218. Long profile of a large valley-head gully, designated gully 1 in figure 214.
most of the material eroded is probably deposited GULLYING AND LAND USE
on valley flats. That most of the active valley-head and valley-side
Although the general conditions of topography and gullies in the Medicine Creek basin are man induced
vegetation that are conducive to the formation of is highly probable. Many of the valley-side gullies
valley-head gullies can be stipulated, the localization are directly associated with cultural features such
of a gully in a particular valley head rather than as roads, buildings, and fences and nearly all the
in an adjoining valley head depends on events that valley-head gullies have begun in valley heads that
cannot be reconstructed. Such events include the drain areas of cultivated upland. During past nat-
past methods of cultivation of the upland and the urally induced episodes of erosion, gullying does not
past land use of the valley head, as well as such seem to have begun in valley heads. The two post-
fortuitous events as the digging of burrows by Stockville episodes of trenching stopped short of
animals. Moreover, no correlation whatever is ap- the valley heads or else progressed from downvalley
parent between the size of a gully and its topo- to the valley heads. The drainage system is presently
graphic situation. Study of the topography of seven extending itself beyond the valley heads and into
of the largest valley-head gullies in the basin showed the upland, a process that last took place before
accumulation of the Stockville terrace deposits. The
that the drainage areas ranged from 4 to 37 acres
lack of gullying in two tributary heads on Dry
and 1 that the slope of the upland draining into the Creek, densely vegetated because of protection by
valley head ranged from 0.02 to 0.06 foot per foot. fences, indicates that land use has caused gullying
The slope of the valley heads could not be accurately elsewhere. (See fig. 182.)
reconstructed, but all appeared to have been rela- The role of land use in the initiation and growth
tively steep. Other valley heads, similar in drainage of large valley-bottom gullies is uncertain. Geologic
area and steepness to the severely gullied valley evidence indicates that valley-bottom gullying has
heads, contain small gullies or none. taken place on Elkhorn Canyon and probably, on-;
drought in the past and by overgrazing since 1875.

Bryan (1941) also concludes that a slight change
in climate from the dry toward the less dry is ade-
quate to convert ephemeral streams from a condition
of erosion to alluviation.
These conclusions are reasonable, although the
specific climatic conditions that prevailed during
past episodes of cutting and filling cannot be recon-
structed by any presently known method. During
dry years in the Medicine Creek basin, gully advance
has been negligible and alluviation has taken place
in large valley-bottom gullies. Gullying does not
literally take place during dry years, but rather
during the wet years that occur from time to time
during a general period of drought. A temporary
effect of drought may be an increase in the rate of
accumulation of valley fill, as is suggested by the
character of the terrace deposits on Elkhorn Canyon;
but valley fill accumulated during drought is subject
to almost immediate incision. As to the climatic
conditions that prevailed during episodes of filling,
good faunal evidence indicates that the Stockville
terrace deposits accumulated during a climatic re-
gime very similar to the present climate. Past epi-
sodes of cutting are probably due to a series of
droughts more rigorous than any recorded since
white settlement.
CONTROL OF GULLIES
Valley-head gullies of moderate to large size have
been stabilized in some parts of the basin. The most
effective methods of stabilization have involved the
FIGURE 219. Head scarp of a small valley-head gully on Dry Creek. Note
construction of an arcuate ditch above the gully
animal burrows and overhanging rim of head scarp. head and the diversion of drainage along grassed
waterways on either side of the gully. As is true of
Dry Creek within the past 350 years and before the gully control structures generally, these must be
arrival of settlers. There is no evidence of late maintained; and lack of maintenance has resulted
Recent valley-bottom gullying elsewhere in the basin, either in renewed activity of the gully head scarp or
except that valley-bottom gullies, probably formed in trenching of the grassed waterways.
since settlement, are advancing rapidly at one or A heavy vegetal cover in a valley head inhibits
more localities in the valleys of major tributaries formation of new gullies because gullies begin on
to Medicine Creek. The formation of these large the valley head rather than on the upland. The cul-
valley-bottom gullies entirely from natural causes tivation of upland to the very edges of valley heads
is improbable. The condition of the vegetal cover has been unfortunate. Restoration of native vege-
in relatively undisturbed areas indicates that the tation in the valley heads and in the area of upland
climate since settlement has not been particularly adjoining the valley head would probably be effective
conducive to the formation of gullies. in preventing new gullies and in stabilizing gullies
Antevs (1952) concludes that arroyo cutting and of small to moderate size.
filling in the semiarid southwest are controlled main- No large valley-bottom gully in the Medicine
ly by vegetation, which is in turn controlled natu- Creek basin has been effectively controlled, and only
rally by climate. Channeling is a result of drought; one effort toward control was noted. Concrete blocks
filling takes place during climatic transitions; and were placed in the head of the main gully on Dry
soil development takes place during relatively moist Creek, but these had no apparent effect. The control
periods. Arroyo cutting was caused mainly by problem is complicated by the fact that a given head
scarp may have in its wake a series of channel station, but the maximum exceeded 20,000 ppm dur-
scarps, all of which must be controlled. A reduction ing only 3 years of combined station record. The
of runoff on the valley bottom is the only apparent maximum daily concentrations were larger than
solution to the problem, and this can probably be about 1,000 ppm for the combined record, and about
achieved by construction of terraces on the upland, one-third were between 10,000 and 20,000 ppm.
restoration of native vegetation of some parts of the The material that makes up suspended sediment
valley bottoms, and more moderate grazing. is relatively fine because of the general lack of
SUSPENDED SEDIMENT AND THE HYDRAULIC
sources of coarse material in the basin. The per-
GEOMETRY OF CHANNELS centages of clay (finer than 0.004 mm) and silt
(0.004 to 0.062 mm) are variable, but clay and silt
By CLOYD H. SCOTT
together make up 90 percent or more of the sus-
Medicine Creek has a well-sustained low flow. pended sediment. Ten percent or less of sand
Fox Creek and Brushy Creek also receive ground- (0.062 to 2.00 mm) is generally finer than 0.125
water inflow, although the volume is not so great mm except at the two stations on Medicine Creek,
as the volume received by Medicine Creek. Fox where a few samples contain 1 or 2 percent of sand
Creek has a well-sustained low flow, but Brushy ranging from 0.125 to 0.250 mm.
Creek usually stops flowing for some time during Stream-channel cross-sectional shapes are rather
the summer months. The other gaged tributaries, diverse within the basin, particularly the shapes of
Dry Creek and Mitchell Creek, and most of the channels of ephemeral streams. In cross section,
ungaged streams in the basin flow only after rainfall. the channels of ephemeral streams range from shal-
Typical hydrographs of large flows are shown in low swales on the valley floor to deeply incised
figure 220. The magnitude of these flows was such trenches, either wide or narrow. Figure 210 (upper
that the daily mean discharge was equaled or ex- left} illustrates a broad valley floor in which a nar-
ceeded about 1 percent of the time. Medicine Creek row channel is forming. In figure 204 (upper left)
at Maywood rises and falls more slowly than the a narrow trench is shown upstream from the head
others because of reservoir regulation upstream. of a gully, and this trench is being widened and
Medicine Creek and its tributaries tend to be deepened by the advance of the gully head scarp.
"flashy" streams; that is, the discharges increase The broad trench formed by advance of the head
rapidly in response to rainfall, and peaks are of scarp is only about twice as deep as the narrow
short duration. This flashy condition is probably trench, but a short distance downstream from the
the result of several factors, the most important of head scarp the width of the broad trench is several
which is the intensity of thunderstorms. One storm times that of the narrow trench. Head scarps are
produced 1.70 inches of rain in 45 minutes at advancing in some valley floors where narrow chan-
Tobiassen gage. Such intense storms do not occur nels are poorly defined or do not exist at all. As a
often, but storms that produce as much as three- result of the advance of these head scarps, the shape
fourths of an inch of precipitation in an hour or less of the channel is changed from broad and shallow
are common. to broad and deep.
The thunderstorms result in runoff having mod-
Dry Creek (fig. 210, upper riqht) and Brushy
erately high suspended-sediment concentrations. The
Creek both have broadly incised channels at the
maximum observed concentration in the basin during
gaging stations, and Brushy Creek has a poorly de-
the approximately 37 years of combined station
fined low-flow channel in the broadly incised channel.
record was 192,000 ppm (parts per million), and
Medicine Creek and Fox Creek, which are perennial
the maximum observed suspended-sediment concen-
streams, have well-defined low-flow channels at the
tration was more than 100,000 ppm in 5 of the 37
years of combined station record. All the annual gaging station.
maximum observed concentrations were in excess
HYDRAULIC RELATIONS AT A SECTION AND
of 10,000 ppm for the period of record. Maximum IN A DOWNSTREAM DIRECTION
concentrations in excess of those observed undoubt-
edly occurred during most years. Because concen- As the discharge at a particular cross section
trations do not remain high for very long, the maxi- (called "at a station" by Leopold and Maddock, 1953,
mum daily suspended-sediment concentrations are p. 4) in a stream increases, the width, depth, and
much less than the maximum observed concentra- velocity increase. Previous investigators (Leopold
tions. The maximum daily concentration during the and Maddock, 1953; Wolman, 1955; and Leopold
period of record was 37,600 ppm at the Fox Creek and Miller, 1956) have shown that, within limits of
1400
1300
1200
1100
1000
-Medicine Creek above Harry Strunk Lake
900
O 700
uj 600
I
O
CO
O 500
400
300
-Fox Creek
200
Medicine Creek at Maywood
100
Dry Cre ek-
12! ,=t
0.2 4 6 8 10 12 14 16 18 20 22 24
TIME, IN HOURS
FIGURE 220. Typical hydrographs of large flows.
bankfull discharge, these variables are simple power The product of mean velocity, width, and mean
functions of the water discharge of the form depth must equal discharge; therefore, the sum of
uaQm the exponents b, f, and m must equal 1.00.
daQf Points on graphs of width, depth, and velocity
waQ^ versus discharge for the gaging stations in the Medi-
where u is mean velocity (fps),: cine Creek basin scatter rather widely from the
d is mean depth (ft), . mean, particularly for Brushy, Dry (fig. 221), and
w is width,(ft), and ; % Mitchell Creeks. Scatter at the lower values of
Q is water discharge (cfs). discharge is caused, in part, by natural variations
10 100 600
DISCHARGE, IN CUBIC FEET PER SECOND
FIGURE 221. Change of width, depth, and velocity with increasing discharge at a section, Dry Creek near Curtis.
in channel shape at the different sections used for on reaches where flows are over loess, which covers
making discharge measurements by wading. Flows much of the basin. The assumption implies that the
too deep to be waded were measured at a single channels of two streams are similar at the points
section at each of the stations, and the scatter on along their lengths where the discharges are the
the plots is somewhat reduced at the higher dis- same at equal frequencies of occurrence.
charges. There would be some scatter on the plots The values of b, f, and m for changes of width,
even if all measurements were made at a single depth, and velocity in a downstream direction ob-
cross section. tained from figures 222 and 223 cannot be considered
The slopes of curves relating width, depth, and reliable because too few stations are available to
velocity to discharge at each gaging station were establish averages. The values can be used to show
fitted by eye and adjusted so that the values of general differences between the types of streams
b, f, and m summed to 1.00. The adjustments were represented by the two groups.
small for all sections. The relations of width, depth, The downstream relations of width, depth, and
and velocity to water discharge at the individual velocity to discharge could best be defined by meas-
sections have little meaning; therefore, the curves, urements along a stream at several stations for which
without the defining points, were placed on one at-a-station relations of width, depth, and velocity
graph, and the discharges that were equaled or ex- to water discharge and flow-duration curves are
ceeded 1, 2, and 25 percent of the time were indi- available, tiowever, gaging stations of the number
cated on each of the curves. Discharges equaled or required for such a study are found on few if any
exceeded 1, 2, and 25 percent of the time were streams. When better data are lacking, discharge
selected because the 1-percent discharge represents measurements might be made at several sections in
approximately bankf ull flow at the Fox Creek station a downstream direction at discharges of unknown
but less than bankfull flow for the other stations. frequency at each of the sections. Even though the
The 25-percent discharge is approximately the modal frequency of the discharges were the same at all
discharge for the Fox Creek and both Medicine the sections measured, the plot of width, depth, and
Creek stations, and the 2-percent discharge is about velocity against discharge would scatter from the
the lower limit for which reliable measurements of average line because of variations in shape that exist
width, depth, and velocity are available for the Dry, from place to place in a natural channel. Of course,
Brushy, and Mitchell Creek stations. it would not be possible to obtain measurements of
The slopes of width-discharge relations for Dry discharges of the same frequency at all sections,
and Brushy Creeks are greater than those for the especially for channels of ephemeral streams.
other sections, and lines joining points of equal- Assume that a series of measurements are to be
discharge frequency indicate that the sections should made in a downstream direction on Dry Creek at dis-
be divided into two groups (figs. 222 and 223). The charges that are equaled or exceeded about 2 percent
fact that Mitchell Creek does not fit with either of the time at each of the sections. Further assume
group is attributed to the presence of a concrete that the downstream relation of width to discharge
control at the section and to a low-water road cross- for Dry Creek is the same as that shown on figure
ing a short distance downstream, which causes back- 223 for discharges equaled or exceeded 2 percent of
water at the section. Therefore, the Mitchell Creek the time. If a measurement is made at the gaging"*
section will not be considered further. station at a discharge assumed to be equaled or
exceeded 2 percent of the time, but the discharge
Lines that join points of equal frequency of dis-
is actually the one equaled or exceeded 1 percent
charge on the curves of the at-a-station relations
represent the downstream relations of width, depth, of the time, then the width will be greater than that
indicated by the downstream relation at discharges
and velocity to discharge, if the assumption is made
equaled or exceeded 2 percent of the time by about
that the relation at sections not on the same stream
40 percent plus or minus the deviation from the
is representative of relations at different sections
at-a-station relation (fig. 221). About two-thirds
on the same stream. This assumption may be justi-
of the points that defined the at-a-station relation
fied for two reasons. First, the discharges of equal
for the Dry Creek station were within plus 35 and
frequency for the two Medicine Creek stations and minus 35 percent of the mean line. As the difference
the Fox Creek station plot about on a straight line between the slopes of the lines for the at-a-station
even though the uppermost section on Medicine relations and the downstream relations of width,
Creek, at Maywood, is somewhat affected by reser- depth, and velocity to discharge becomes less, the
voir regulation. Second, the gaging sections are scatter from the average line attributable to differ-
100
I I I I I I I I
100 1000
EXPLANATION
Percentage of time daily flow
Medicine Creek above equaled or exceeded:
Harry Strunk Lake
1
A
O
Fox Creek at Curtis 25
FIGURE 222. Change, of width, depth, and velocity in a downstream direction for channels of perennial streams.
10 100
EXPLANATION
Percentage^ time daily flow
equaled or exceeded:
Dry Creek near Curtis
Brushy Creek near Maywood 2
FIGURE 223. Change of width, depth, and velocity in a downstream direction for channels of ephemeral streams.
ences in frequency of discharge at the various sta- at a station. Because the discharge at a constant
tions becomes less. Establishing fairly reliable frequency increases downstream, the width must
downstream relations of width, depth, and velocity increase rapidly. However, the width for the broadly
to discharge is possible if a sufficient number of sec- incised ephemeral streams increases very little down-
tions are selected to provide a reliable average and if stream. 'The increase in discharge is accommodated
measurements of extremely high or low discharges by relatively large changes in both depth and ve-
are avoided. locity in a downstream direction. The downstream
The slopes of the relations of width, depth, and change in width of the broadly incised ephemeral
velocity to discharge (table 8) indicate differences streams may be greater than that indicated on
figure 223, but the change still would not be so large
TABLE 8. Values of b, f, and m for streams in Medicine as the change of the perennial streams. Figure 204
Creek basin (upper left) shows the main-stem channel of Dry
Creek just below the gully scarp that forms the end
Perennial streams Ephemeral streams of the broadly incised channel, and figure 210 (upper
right) shows the main-stem channel at the gaging
At a station 1 Downstream 2 At a station 1 Downstream 2
section. The top widths of the channel just below
b. 0.24 0.69 0.35 0.03
the gully scarp and in the vicinity of the gaging
f .56 .12 .43 .48 section are 40 to 50 feet. The distance is too short
m . _ __ ____ .20 .19 .22 .45
and the measurements are too rough to indicate
1 Unweighted average. the change in width along the Dry Creek channel;
2 At discharge equaled or exceeded 1 percent of time.
but if the width increased as rapidly as the indicated
change of about 5 percent per mile on Medicine
between the perennial and ephemeral streams in the Creek between Maywood and the gage above Harry
Medicine Creek basin. The at-a-station slope values Strunk Lake, the width at the gage on Dry Creek
may be considered reliable, but slopes for the down- should be more than 10 feet wider than the width in
stream relations probably only indicate the correct the vicinity of the head scarp.
order of magnitude. The differences between the
at-a-station relations for the perennial and ephemeral SUSPENDED-SEDIMENT DISCHARGE RELATIONS
streams reflect the differences in channel shapes. The curves relating suspended-sediment load to
The channels of the perennial streams have fairly water discharge for the perennial streams (fig. 224)
steep banks, and the width does not increase as have breaks in slope at water discharges that are
rapidly as the depth. The channels of the ephemeral higher than normal but below bankfull stage. The
streams also have steep banks; but the bottoms of curves applying to ephemeral streams show no cor-
the channels are somewhat parabolic, and the width responding breaks in slope (fig. 225), and in value
can increase rapidly as discharge increases until the of slope they are generally similar to the upper part
limits of the steep confining banks is reached. How- of the curves applying to perennial streams. The
ever, a relatively large discharge is required for the breaks in the suspended-sediment rating curves re-
width of water surface to reach the confining limits sult from a rather complex and uncertain set of
of the steep banks in the broadly incised channels of conditions. The concentrations, peak and mean, for
ephemeral streams. The changes of velocity with runoff periods are highly variable. Even two events,
increasing discharge are about the same for peren- which may be very similar with respect to peak
nial and ephemeral streams; therefore, the change discharge and total discharge, may have very differ-
of width is less and the change of depth is greater peak suspended-sediment concentrations and mean
with increasing discharge for perennial streams than concentrations. The peak concentration generally
for the broadly incised ephemeral streams. occurs before the peak water discharge for stations
Depths of water in a channel normally do not in the Medicine Creek basin, but the length of lead
increase very rapidly in a downstream direction. is variable. The length of lead may be from 1 to 3 or
That is, it would not be expected that the depth at or more for the perennial streams and is generally
a cross section on a stream would be much greater an hour or less for the ephemeral streams. The
than the depth at another section several tens of shorter lead time, in general, occurs with a small
miles upstream. The downstream change in depth peak discharge. The variability in concentration
for perennial streams in Medicine Creek basin is of suspended sediment and in' length of lead results
relatively small, and the downstream change in because the suspended-sediment load is made up
velocity is about the same as the change in velocity principally of fine material derived from sources
other than the streambed. The amount of fine ma- analyses were not weighted because a plot of water
terial that is transported by the streams during a discharge against percentage of suspended sediment
given storm depends on intensity and duration of finer than 0.004 mm and coarser than 0.062 mm
rainfall, soil conditions at the time of the storm, indicated no relation between instantaneous water
antl tjt&er factors. The leading1* coneentratitHtf indi- diseharge-gilfd ârticle-size distribution. -The average
cates that the sediment rating curve should describe size distributions for each year for each of the
a loop, but because of the variability of concentration stations did show an inverse relation to annual water
from storm to storm and the variability of the lead discharge; that is, the percentage finer than a given
time of concentration, the loop would not be defined size decreased as water discharge increased. The
by an average of many measurements for many relation was not very well defined, and the indicated
storms. Instead, the points defined by the measure- change was not large; but the relation of annual
ments would scatter with increasing discharge. (See water discharge to the average percentage finer than
fig. 224.) The measurements that define the points a given size does indicate the possibility of a relation
of figures 224 and 225 are somewhat biased, however, between the percentage finer than a given size and
because only a few measurements were obtained instantaneous water discharge. The large vari-
before peak water discharge and practically none ability of particle-size distributions for a given water
were obtained near peak sediment concentration. discharge, however, prevent detection of a relation.
The generally rapid rises and mud roads prevented Average particle-size distributions of suspended
arrival of workers at the stations much before the sediment and bed material (figs. 226 and 227) indi-
occurrence of the peak water discharge, particularly cate little difference between size distributions of
on the ephemeral streams. Nearly all the measure- suspended sediment and a large difference between
ments at those stations were obtained after the peak size distributions of bed material for Medicine Creek
water discharge; therefore, the scatter of points is at Maywood and Medicine Creek above Harry Strunk
fairly uniform throughout the range of water dis- Lake. Medicine Creek above Harry Strunk Lake
charge. has a small percentage of material in the 0.125- to
The lower part of the curves for the perennial 0.250-mm range in suspension; such material is
streams is represented mostly by measurements made normally not in suspension at Maywood. Although
during small rises. A small increase in water dis- the size distributions of suspended sediment are
charge is accompanied by a relatively large increase similar at the Medicine Creek stations, the distribu-
in sediment concentration, and the lead time for tions of bed material are very different. The bed
smaller rises is generally short. As a result, the material at Maywood has 36 percent of the material
slope of the relation of water discharge to sediment finer than 0.062 mm, 44 percent between 0.062 and
discharge increases more rapidly for low water dis- 2.0 mm, and 20 percent coarser than 2.0 mm; at the
charges than for the high water discharges. station above Harry Strunk Lake only 8 percent is
The fact that the slope of the water-sediment dis- finer than 0.062 mm, 80 percent is between 0.062
charge relation is slightly less than unity for the and 2.0 mm, and only 12 percent is coarser than
ephemeral streams indicates that the concentration 2 mm. The increase of material in the 0.062- to 2.0-
remains about the same or decreases slightly in a mm range is probably from the tributaries draining
downstream direction, but the fact that the slope the area to the southwest of Medicine Creek. There
of the water-sediment discharge relation is slightly are, however, no analyses of bed-material samples
greater than unity for perennial streams indicates from any of these tributaries. The average distri-
that the concentration increases in a downstream bution of only three samples, each of bed material
direction. Because so few records are available to from the low-flow channels of Brushy and Fox Creeks
establish good average downstream relations, it is (fig. 50), indicates that the bed material at those
not possible to state that an actual difference exists stations is similar to bed material at Medicine Creek
between the slopes of the water-sediment discharge at Maywood, especially for the sizes finer than about
relations in a downstream direction for perennial 1.0 mm. Suspended-sediment size distributions for
and ephemeral streams. Brushy, Fox, Dry, and Mitchell Creeks (fig. 226)
show that there is an apparent increase in fine
PARTICLE-SIZE DISTRIBUTIONS OF SUSPENDED material available for transport as suspended sedi-
SEDIMENT AND BED MATERIAL ment from the west to the east side of the basin.
Average particle-size distributions of suspended Brushy Creek, on the west side of the basin, has a
sediment were obtained from unweighted averages median suspended-sediment particle size of about
of all size analyses at each of the stations. The size 0.014 mm, whereas Mitchell Creek, on the east side
100,000
100 1000 10,000

WATER DISCHARGE, IN CUBIC FEET PER SECOND
FIGURE 224. Change of suspended-sediment load with increasing discharge at a section, perennial streams.
100,000
10,000
1000
100
Mitchell Creek above Harry Strunk Lake
I I I I I I I I I I I I I I I I
10 100 1000
WATER DISCHARGE, IN CUBIC FEET PER SECOND
FIGURE 225. Change of suspended-sediment load with increasing discharge at a section, ephemeral streams.
en 99.0 -
x: Medicine Creek above

Harry Strunk Lakei
0.002 0.01
DIAMETER, IN MILLIMETERS
FIGURE 226. Size distribution of suspended sediment at gaging stations in Medicine Creek basin.
of the basin, has a median particle size of about probably represents 95 percent or more of the total
0.004 mm. The decrease in median particle size load for the period of record.
from west to east reflects the increasing distance
from the sand-dune area of the extreme western and RELATIVE SEDIMENT CONTRIBUTIONS OF THE
MAJOR TRIBUTARIES
northwestern parts of the basin.
The measured loads for all the stations are nearly In any basin or subbasin, erosion and deposition
the total loads because the suspended sediment con- are taking place simultaneously, and sediment data
tains little sand and the fine suspended sediment collected at any point along a stream represent the
should have fairly uniform distribution in the ver- net of erosion and deposition upstream from that
tical. Computations of total load were made by the point.
Colby method (Colby, 1957) for the two stations A measure of the net erosion and deposition in
on Medicine Creek, both of which have sand beds. a basin is the discharge-weighted mean concentra-
Not many sets of data are available for computation tion of suspended sediment. This is the concentration
that .would result if all the water passing a point
of total loads, but the average percentage of meas-
during some period of time were mixed with the
ured load to total load is indicated. At low flow the suspended sediment passing during the same period.
measured load amounts to alkmt 90 percent of the If the cumulative water volume is plotted against
total load, but at high flow it 'amounts to more than cumulative weight of sediment, the slope of a line
95 percent. Because all the' suspended sediment at that joins points of the graph defines the discharge-
the Dry Creek gage results from storm runoff and weighted mean concentration for the period of time
a large part of the suspended sediment at the Brushy for which the data are cumulated. The annual sedi-
and Fox Creek gages results from storm runoff, the ment loads and storm runoff for the station were
measured suspended sediment for those stations. reduced to a per-square-mile basis and are shown
Medicine Creek above Harry Strunk Lake
10
5 -
2 -
1.0
0.05
DIAMETER, IN MILLIMETERS
FIGURE 227. Size distribution of bed material at gaging stations in Medicine Creek basin.
on figure 228. Reducing the values to a per-square- the curve probably occurs because the ratio between
mile basis does not change the constant of propor- the variables is not constant at all rates of cumula-
tionality and presents a comparison of storm runoff tion. If the ratio were constant at all rates of
and sediment yields. Such a comparison is valid if cumulation, the break would have to be explained
precipitation over the entire basin is assumed to be by a physical change that would cause a greater
uniform in amount. Because of the relatively small reduction in sediment yield than in water yield
size of the basin and the low relief, the assumption (Searcy and Hardison, 1960). A break in the curve
of uniform precipitation probably is valid if several is drawn at the 1956 water year for the Dry Creek
years record are considered. The" differences in total data; and breaks at the same year are indicative
precipitation for the years 1951-58 recorded at the for Fox and Mitchell Creeks, but they are not drawn.
Wellfleet, Curtis, and Stockville rain gages were less For hydrologic data plotted in this manner, breaks
than 10 percent. As the time period is shortened, that persist for less than about 5 years should be
however, the probability of uniform precipitation attributed to chance and ignored unless there is
becomes less. definite reason for believing that such a break should
Each of the stations shows a definite break in occur. The high rainfall and runoff during the 1951
slope following the 1951 water year. The break in water year explains the break at the end of that
Slopes of lines representing

discharge-weighted concentrations
Data for 1951-58 water years

except Mitchell Creek which
is for 1952-57 water years
reek above Harry Strunk Lake
40 60 80 100 120 140 160 180 20C

WATER DISCHARGE, IN CUBIC FEET PER SECOND DAYS PER SQUARE MILE
FIGURE 228. Cumulative water discharge and sediment discharge per square mile at gaging stations in Medicine Creek basin. Year at which a curve
changes in slope is indicated.
year. Data are available for only the part of the For drainage basins that are incised in a level or
year when rainfall was extremely heavy. The reason gently sloping upland surface to approximately the
for the break at the end of the 1956 water year is same local relief, a correlation of relief ratio with
obscure. Runoff for the 1957 water year was gen- size and shape properties is apparent. Basin relief
erally greater than normal, but the break is in the increases at a much slower rate than basin length;
therefore, relief ratio is reduced both by increase
wrong direction on the basis of the experience for
in basin size and by increase in basin length. Because
the 1951 water year. Land-conservation work was of the decrease in valley-side slope with increasing
being actively carried on during the period of the in- channel order, the lower relief ratio of high-order
vestigation, but there are too few data to evaluate basins is accompanied by a lower mean value of side
the significance of the break. slope. Similarly, because of the decrease in channel
The ephemeral streams Brushy, Dry, and Mitch- slope with channel order (fig. 229), the lower relief
ell Creeks have higher discharge-weighted mean ratio of higher order basins is accompanied by a
concentrations and generally yield larger amounts lower mean channel gradient. However, the rela-
of sediment per square mile for storm runoff than tion of channel slope to channel order may not be
the perennial streams. consistent; therefore, relief ratio may not express
adequately the slopes of low-order channels.
GEOMORPHIC PROPERTIES IN RELATION TO WATER In a discussion of the interrelations of drainage-
AND SEDIMENT DISCHARGE
basin characteristics, Gray (1961) presents evidence
One major objective of quantitative geomorphic to show that, for small basins, the properties of
measurements is to select drainage-basin properties area, length of main stream, and length to center
that have the highest possible degree of correlation of area are highly correlated. In a region of homo-
with runoff and sediment yield. If runoff and sedi- geneous relief, a correlation also exists between these
ment yield are known for a large number of drainage properties and the slope of the main stream. Al-
basins in the same region, statistical methods can be though not discussed by Gray, relief ratio probably
used to evaluate the significance of different drain- will express adequately all the above interrelated
age-basin properties. If, as in the Medicine Creek properties for some regions. In addition, relief ratio
basin, runoff and sediment yield are known for only will express differences in relief between basins of
five subbasins, evaluation of the properties cannot about the same size. The effectiveness of relief ratio
be made rigorously by statistical methods. Ideally, as a property probably depends on, among other
the properties selected should be independent of one things, the constancy in shape of long profile among
another, unambiguous, and not unduly time con- the basins being compared. A basin whose main
suming to obtain. For the Medicine Creek basin,
properties that meet these qualifications and offer
promise of good correlation with runoff and sediment
yield are relief ratio, adjusted frequency of first-
order streams, and percentage of area in upland.
Relief ratio was shown to correlate with mean
annual sediment accumulation for small drainage
basins in the upper Cheyenne River basin by Hadley
and Schumm (1961, p. 173). Maner (1958) reports
that relief ratio correlated more closely with sedi-
ment delivery rates than did size of sediment con-
tributing area, drainage density, basin shape, or
weighted-average land slopes. The drainage basins
studied by Maner, which are in the Red Hills area o.oi -
of Oklahoma and Texas, range from 332 to 0.036
square miles. For these basins, relief ratio showed a
correlation, as based on inspection of scatter dia-
grams, with basin size, basin shape, average land
slopes, and drainage density. Maner concludes that
0.001
sediment delivery rate in the Red Hills area is a CHANNEL ORDER
function of several drainage-basin properties that

FIGURE 229. Mean channel slope in relation to channel order for Lime,
evidently are expressed adequately by relief ratio. Dry, and Fox Creek subbasins.
stream is strongly concave upwards in long profile 450
may have the same relief ratio as another basin, of

equal size, whose main stream is nearly straight in
long profile. However, such differences in long pro-
file surely affect runoff and sediment yield. LENGTH
FRAGE
OF
IRST-ORD
w en o
Frequency of first-order streams was found by CHANNEL
FEETS,
IN
Morisawa (1959) to give a high correlation with

peak intensity of runoff for drainage basins in the
Appalachian Plateaus. The properties used by Mori-
sawa in multiple regression analysis were relief
ratio, basin circularity, and first-order channel fre-
quency. She considers that these three properties
express the shape, relief, and network composition cucji
9~ oo
of a drainage basin, that they vary independently of
each other and of area, and that they do not dupli-
cate any other geomorphic factor. For the Medicine
| 0. 25
Creek basin, relief ratio is considered to be an ade-
quate expression of basin shape, but first-order Q u.
o: LO
90-
channel frequency is an important property not only
ciS
in relation to runoff and sediment yield but also in £0
u. u. cji
relation to gully erosion. o~
L. Z
9 H-
Frequency of first-order channels expresses the S 3

small-scale properties of a drainage basin and there- Q. LU
21
wl
by complements relief ratio, which expresses the
larger scale properties. Channel frequency must be " 0.05
adjusted for immature basins according to their 0.45
differing percentages of upland area. In figure 230
scatter diagrams show the relation of adjusted fre-
quency of first-order channels to mean valley-side
slope, slope of first-order channels, and mean length W § 0.35
of first-order channels. The number of points is too LL) U.
small to warrant the calculation of correlation co- (75 LU

efficients, but the existence of a relation is apparent
not only from the scatter diagrams but also from
examination of the topographic maps. In addition, z£
0.25
the reason for the correlation is apparent from geo-

morphic history. In subbasins where post-Peorian *" X
dissection was most intense, first-order channels are
shorter, steeper, and more numerous. Much of the 0.15
120 160 200 240 280
scatter of points in figure 230 is due to the fact that ADJUSTED FREQUENCY OF FIRST-ORDER CHANNELS,
some of the subbasins are less homogeneous than IN CHANNELS PER SQUARE MILE
others in texture of drainage. EXPLANATION

The relation of length to channel slope for first- O -- AX
Brushy Creek Fox Creek Dry Creek Mitchell Creek
order channels in the lower part of Dry Creek is
+ A V
shown in figure 231. Each point represents the
Well Canyon Upper Dry Creek Lower Dry Creek Lime Creek
mean slope and mean length of 10 channels in a
particular length group. Noteworthy is the fact FIGURE 230. Adjusted frequency of first-order channels in relation to mean
that the shorter channels have relatively steeper valley-side slope, mean slope of first-order channels, and mean length of
first-order channels.
slopes than the longer channels. In addition, the
shorter channels have steeper valley heads and more
concave profiles. This is of particular importance order streams should give indirect expression to
in gully erosion because steep valley heads promote another important variable the percentage of area
the formation of new gullies. in valley flat. An unusual geomorphic feature of
In combination, relief ratio and frequency of first- the Great Plains generally is the flatness of valley
330 EROSION AND SEDIMENTATK
0.240
1 i i ation from one subbasin to the next, is related to
the slope of channels. In general, the slope of the
"\ - upland is less than the slope of first- and second-
order channels, about equal to the slope of third-order
5o channels, and greater than the slope of fourth-order
u.
K 0.200
UJ
\
and higher order channels. (See tables 3 and 4.)
CL
O
u.
\ The percentage of upland gives a reasonably good
indication of land use. (See table 9.) Similarly, the
Z
UJ
£ 0.160
\ \
percentage of area in side slope and valley flat shows
a correspondence with the percentage of area used
for hay and pasture.
CO
Ni The probable effect of the selected properties
UJ
Z
Z
_ \ (relief ratio, frequency of first-order streams, and
I
o
Z 0.120
X percentage of area in upland) on runoff and sedi-
ment yield can be inferred. On the premise that
UJ
lower slopes, wider valley flats, and larger (or
longer) drainage basins all tend toward reduction of
- runoff and sediment yield because of greater oppor-
tunity for infiltration and for deposition of sediment,
runoff and sediment yield should be positive func-
n nan I i i tions of relief ratio. According to the graphs of
100 300 500 700
MEAN LENGTH OF FIRST-ORDER CHANNELS, IN FEET Hadley and Schumm (1961), the function is expo-
nential where mean annual sediment accumulation
FIGURE 231. Slope of first-order channels in lower Dry Creek subbasin
in relation to length. in reservoirs is the dependent variable. According
to the graph of Maner (1958, fig. 3), the function is
exponential where sediment delivery rate is the
bottoms along channels of both low and high order.
dependent variable. Sediment delivery rate is the
The percentage of area in valley flat for each channel
ratio, expressed as a percentage, between annual
order in each major subbasin has been calculated rate of sediment yield and annual gross erosion rate.
from the product of average width of valley flat and
average length of channel. (See table 4.) The per- An increase in frequency of first-order channels
centage distribution of valley flat according to chan- is associated with an increase in both valley slope
nel order differs considerably from one subbasin to and side slope for lower order channels; therefore,
the next and depends mainly on the distribution of increase in water and sediment discharge should be
channel length according to order. In Well Canyon a positive function of frequency of first-order chan-
subbasin, for example, about 18 percent of the total nels. For drainage basins whose main channel is
area of valley flat is in the wide bottom along the very long and bordered by wide valley flats, a high
main channel, whereas in Dry Creek subbasin only yield of water and sediment from lower order chan-
8 percent is along the main channel. The probability nels may be largely dissipated along the main
of deposition of sediment on the main valley flat channel.
seems greater for Well Canyon than for Dry Creek,
and the probability of rapid runoff is less because TABLE 9. Areas in upland and valley flat in comparison with
of the lower slope angle of the Well Canyon valley. areas in two categories of land use
The relief ratio gives an indication of the percentage [Land-use percentages are mean values from inventories taken in 1954,
of valley flat along channels of fourth and higher 1955, and 1957]
order, whereas the frequency of first-order channels

Area in
gives an indication of the percentage along channels Area in Area in row crops, Ratio of
of first, second, and third orders. (See fig. 198, in side slope hay and Area in small row crops
Subbasin and pasture upland grain, and to small
which the slope of the curve relating channel length valley flat
(percent)
(percent) (percent) fallow
(percent)
grain
to channel order changes at about third order.)

Because of the immaturity of drainage in the Mitchell Creek
Dry Creek ____ ....
48.2
65.2
51.3
69.7
51.8
34.8
46.6
28.7
1.72
1.43
Medicine Creek basin, a variable is needed to express Well Canyon . 79.6 83.4 20.4 15.6 1.34
the difference among subbasins in percentage of Fox Creek _________ 83.5
74.1
86.2 16.5 12.7 1.57
Brushy Creek __._ 77.7 25.9 21.7 .98
upland. The slope of the upland, which shows vari-
The effect of percentage of area in upland depends 10
on the land use of the upland. If the upland is used
mainly for row crops, runoff and sediment yield
probably would be a positive function of percentage
of area in upland. If, on the other hand, the upland
were thickly sodded, the function probably would be
negative. A positive function would be expected for
the Medicine Creek basin.
The preceding inferences as to the geomorphic
properties most highly correlated with runoff and
sediment yield cannot, unfortunately, be rigorously
tested for Medicine Creek because of the small
number of subbasins for which data are available.
Nevertheless, a trial multiple linear regression of
sediment discharge in relation to the three selected
variables (fig. 232) is given in order to illustrate
the numerical effects of these variables. The trial
regression is based on data in table 3 and figure 228.
The uncertainties involved in correlation are ap-
parent from the graphs in figure 232, and clearly 0.6
no quantitative significance can be attached to the 0.003 0.005 0.007 0.009
RELIEF RATIO
results.
+2
In figure 232 the assumption is made that the
sediment discharge of each subbasin for the period
1952-58 is correctly represented by the data. Data +1
for 1951 were not used because no measurements
for Mitchell Creek were made during that year. The
1958 suspended sediment for Mitchell Creek is esti-
mated. Relief ratio is plotted against sediment dis-
charge on semilogarithmic paper because previous I- O
Z U.
work has indicated that this relation is exponential.
Percentage of upland and adjusted frequency of -2
first-order streams are plotted against sediment
discharge on rectangular coordinate paper. When 10 30 50 70
PERCENTAGE OF UPLAND
only relief ratio is considered (top graph, fig. 232),
the sediment discharge of Fox Creek is very low and
that of Mitchell Creek is high relative to the trial
curve. This is perhaps accounted for by the fact
that the percentage of upland on Fox Creek is lower
than the average, and the percentage on Mitchell
Creek is higher. When relief ratio and percentage
of upland are held constant (bottom graph), a reas-
onably good correlation with adjusted frequency of
first-order channels can be obtained for all the sub-
basins except Medicine Creek above Harry Strunk
140 180 220 260
Lake, which shows a high sediment discharge not ADJUSTED FREQUENCY, FIRST-ORDER CHANNELS
only for this trial regression curve but for the other EXPLANATION
two as well.
O X A
The relatively high sediment discharge of Medi- Brushy Creek Fox Creek Dry Creek
cine Creek above Harry Strunk Lake is attributed A
to yet another factor the presence of raw vertical Mitchell Creek Medicine Creek
banks along a continuous incised channel for a dis-
tance of several miles upstream from the gaging FIGURE 232. Trial graphic multiple linear regression of sediment
discharge in relation to relief ratio, percentage of upland, and
station. In general, the banks along Mitchell, Fox, adjusted frequency of first-order channels.
and Brushy Creeks upstream from the gaging sta- deposition (about 11,000 to 5,000 years B.P.), de-
tions are more sloping and more protected by vege- posits of the Stockville terrace accumulated in the
tation than the banks along Dry and Medicine valleys and the valley sides were graded. The de-
Creeks. But Dry Creek is ephemeral, whereas the posits were incised (about 5,000 to 4,000 years B.P.)
continuous flow along Medicine Creek obtains a con- to form the Stockville terrace, and the deposits that
tinuous supply of sediment from bank erosion. subsequently accumulated in the valleys (4,000 to
A trial graphical linear regression, in which water1,000 years B.P.) were later incised (about 1,000 to
discharge was plotted as a dependent variable against 900 years B.P.) to form the Mousel terrace. Mousel
the three geomorphic factors used in the sediment terracing was followed by accumulation of the upper
discharge regression, yielded results similar to thoseRecent alluvium that underlies the present valley
of the sediment yield regression. As with the sedi- flats. In some valleys the late Recent alluvium has
ment discharge, the water discharge of Medicine been incised intermittently during the past 350
Creek above Harry Strunk Lake is much higher than years. Thus, within the past 12,000 years, the valleys
expected. Probably, Medicine Creek receives a larger were trenched three times prior to the trenching
proportion of its water discharge from ground water that is taking place in some valleys at present. No
than do Fox and Brushy Creeks. In addition, the clear evidence was obtained as to the causes of past
assumption of uniform distribution of rainfall over episodes of erosion or of deposition. However, snails
the various subbasins during the period 1952-58 from the Stockville terrace deposits probably lived
may be incorrect. under climatic conditions similar to the present,
and the terracing of the Stockville terrace corre-
SUMMARY AND CONCLUSIONS sponds approximately with the Altithermal, a well-
Annual precipitation at Curtis for the period established dry climatic period that lasted from
1895-1960 averaged 21.5 inches but ranged from 11 about 6,000 to 4,000 years B.P.
to 38.2 inches. More than half of the precipitation Morphometric analysis of the drainage system
falls during 4 summer months, and much of this is shows that the slope of the exponential curves re-
thunderstorm rainfall. The 5-year moving average lating mean channel length to channel order changes
for Curtis indicates a long-term trend toward de- at the lower channel orders, whereas the curves re-
creasing annual precipitation. Rainfall intensity- lating number of channels to channel order have a
duration-frequency curves indicate that the rainfall nearly uniform slope. This change in slope of the
regimen was satisfactorily represented at Curtis curve relating channel length to channel order is
during the years 1951-58, although precipitation attributed to post-Peorian drainage transformation
was lower than the long-term average. Potential in which the lengths of first- and second-order chan-
evaporation is about three times average annual nels were reduced relative to third-order channels.
precipitation. The Medicine Creek basin lies in a In order to compare the channel frequencies of im-
transitional zone between prairie and plains grass- mature drainage systems whose drainage basins
land; the climax plant community consists of a differ in percentage of undissected upland, drainage
dominance of midgrasses and an understory of short basin area is adjusted by subtracting the area of
grasses. upland. Channel frequency, expressed as number
The immediate effects of drought, as during the of channels per square mile of area actually occupied
years 1933-40, are to reduce the grass cover in such by the valley system, is called adjusted channel
a way as to leave bare patches of ground, but grasses frequency.
are rapidly replaced by weeds. As a result of a Erosion by scarp retreat takes many forms in the
secular climatic shift toward dryness, a climax com- basin, and gullies are regarded as a type of scarp
munity of short grasses would become established. erosion. Gullies are defined and are classified as
Climatic change since the arrival of settlers has had valley-bottom, valley-side, and valley-head gullies.
little effect on the climax community, as indicated Many large valley-bottom gullies apparently origi-
by the character and abundance of vegetation in nate on valley reaches that are locally steepened at
areas that are neither grazed nor cultivated. the entrance of a major tributary. However, the
Geomorphic and stratigraphic evidence indicates ratio of local slope to drainage area that is critical
that the drainage system grew to approximately its for the initiation of a gully is not sharply defined
present pattern and extent during the episode of because other factors (such as width of valley bot-
erosion (tentatively assigned to the interval 12,000 tom) are important. In general, trenching of a
to 11,000 years B.P.) that followed deposition of the valley reach to bedrock does not take place by the
Peorian Loess. During the subsequent episode of coalescence of two discontinuous gullies, but by the
passage of several channel scarps in succession. dry, and the ephemeral streams have higher con-
Saturation of massive silt at the edges of a plunge centrations than the perennial streams.
pool is the most important factor in the advance of An evaluation of geomorphic properties indicates
channel scarps. that relief ratio, adjusted frequency of first-order
channels, and the percentage of area in upland would
Steep-sided valleys that drain areas of cultivated
be expected to have the highest correlation with
upland are most susceptible to valley-head and runoff and sediment discharge. No rigorous
valley-side gullying. The most significant geomor- statistical correlation can be made because of the
phic factors in the areal frequency distribution of small number of subbasins for which data are avail-
valley-head and valley-side gullies are percentage of able, but a trial graphic multiple regression indicates
area in upland and adjusted frequency of first-order that differences in runoff and sediment discharge
channels. can be accounted for reasonably well by differences
Generalizations regarding suspended sediment and in these three properties. They are not only impor-
the hydraulic geometry of channels are based on tant in themselves, but they also give indirect expres-
data from six gaging stations, two of which are on sion to other important properties. For example,
the main channel of Medicine Creek and four on the percentage of area in upland is related to area in
tributaries. At the gaging stations on Dry, Mitchell, row crops, small grain, and fallow.
and Brushy Creeks the flow is categorized as ephe- Active valley-head and valley-side gullies, and all
meral ; at the others, as perennial. Water discharge but a few of the active valley-bottom gullies, are
tends to increase rapidly in response to rainfall, attributed to land use since settlement. During the
much of which falls during summer thunderstorms; past two naturally induced episodes of erosion, gully-
and the discharge peaks are of short duration. ing was mainly confined to valley bottoms. Although
Hydraulic geometry of the channels is expressed some valley incision has taken place within the past
by the slope of curves relating width, depth, and 500 years and before settlement, major episodes of
velocity to discharges equaled or exceeded 1, 2, and the past are probably associated with droughts more
25 percent of the time. At a station, the low-flow severe than any since settlement. Restoration of
channels of perennial streams are well defined and native vegetation to the heads of valleys, together
have steep banks; with increasing discharge, width with conservation measures on the upland, would
does not increase as rapidly as depth. By contrast, be effective in the control and prevention of valley-
the low-flow channels of ephemeral streams are head gullies, which are the most numerous in the
shallow and transient, and width increases rapidly basin. These steps, if accompanied by more moderate
with discharge until the steep sides of the incised grazing in the vallevs, would probably prevent the
trench are reached. In a downstream direction, the initiation of new valley-bottom gullies; but no eco-
incised trenches of ephemeral streams do not widen nomically feasible means for arresting the present
appreciably, and increase in discharge is accommo- large valley-bottom gullies is apparent.
dated by relatively large changes in depth and ve-
REFERENCES
locity. The increase in discharge of perennial streams Antevs, E. V., 1952, Arroyo-cutting and filling: Jour. Geol-
is accommodated by relatively rapid increase in ogy, v. 60, p. 375-385.
width and by a slow increase in depth and velocity. Bacon, S. R., and others, 1939, Soil survey of Frontier County,
Curves relating suspended-sediment load to water Nebraska: U.S. Dept. Agriculture, Bur. Chem. and Soils,
discharge for the perennial streams are character- ser. 1935, no. 9, 34 p.
Bradley, Edward, and Johnson, C. R., 1957, Geology and
ized by a break in slope at water discharges above ground-water hydrology of the valleys of the Republican
normal but below the bankfull stage; curves for the and Frenchman Rivers, Nebraska: U.S. Geol. Survey
ephemeral streams show no corresponding break. Water-Supply Paper 1360-H, p. 589-713.
Peak sediment concentrations usually occur before Brice, J. C., 1958, Origin of steps on loess-mantled slopes:
peak water discharge, but the length of lead is vari- U.S. Geol. Survey Bull. 1071-C, 29 p.
Broecker, W. S.. and others, 1960, Evidence for an abrupt
able and is generally shorter for smaller discharge change in climate close to 11,000 years ago: Am. Jour.
events. In general, 90 percent or more of the sus- Sci., v. 258, p. 429-448.
pended sediment is silt and clay, and the measured Broscoe, A. J., 1959, Quantitative analysis of longitudinal
(suspended) load represents 90 to 95 percent of the stream profiles of small watersheds: Columbia Univ.,
total load. No relation was discerned between in- Dept. of Geology Tech. Rept. No. 18, Office of Naval
Research Contract N6 ONR 271-30, 73p.
stantaneous water discharge and particle-size distri- Bryan, Kirk, 1941, Pre-Columbian agriculture in the south-
bution. In general, the concentration of measured west, as conditioned by periods of alluviation: Assoc.
suspended sediment is higher in wet years than in Am. Geographers Annals, v. 31, p. 219-242.
Clark, 0. R., 1937, Interception of rainfall by herbaceous systems and their representation in an E 4 phase space:
vegetation: Science, v. 86, p. 591-592. Jour. Geology, v. 66, p. 35-54.
Colby, B. R., 1957, Relationship of unmeasured sediment dis- Miller, J. P., 1958, Problems of the Pleistocene in cordilleran
charge to mean velocity: Am. Geophys. Union Trans., North America, as related to reconstruction of environ-
v. 38, no. 5, p. 708-717. mental changes that affected early man, in Climate and
Condra, G. E., and others, 1950, Correlation of the Pleisto- man in the Southwest: Univ. Arizona Bull. 4, v. 28,
cene deposits of Nebraska: Nebraska Geol. Survey Bull. p. 19-50.
15-A, 74 p. Morisawa, M. E., 1959, Relation of quantitative geomorphol-
Frankel, Larry, 1957, Relative rates of loess deposition in ogy to streamflow in representative watersheds of the
Nebraska: Jour. Geology, v. 65, p. 649-652. Appalachian Plateau province: Columbia Univ., Dept.
Frye, J. C., and Leonard, A. B., 1952, Pleistocene geology of of Geology Tech. Rept. No. 20, Office of Naval Research
Kansas: Kansas Geol. Survey Bull. 99, 230 p. Contract N6 ONR 271, 94 p.
Frye, J. C., and others, 1956, Stratigraphy of the Ogallala Phillips Petroleum Co., 1956, Series on pasture and range
formation (Neogene) of northern Kansas: Kansas Geol. plants, booklet 2 on native grasses, 37 p., and booklet 3
Survey Bull. 118, 92 p. on undesirable grasses and forbs, 45 p.
Frye, J. C., and Willman, H. B., 1960, Classification of the Rubin, Meyer, and Suess, H. E., 1956, U.S. Geological Survey
Wisconsinan Stage in the Lake Michigan glacial lobe: radiocarbon dates III: Science, v. 123, no. 3194, p.
Illinois State Geol. Survey Circ. 285, 20 p.
442-448.
Goke, A. W., and others, 1926, Soil survey of Lincoln County, Schultz, C. B., and others, 1948, Preliminary geomorphological
Nebraska: U.S. Dept. Agriculture, Bur. Chemistry and studies of the Lime Creek area (Nebraska) : Univ.
Soils, no. 35, 31 p.
Nebraska State Mus. Bull. 4, v. 3, p. 31-42.
Gray, D. M., 1961, Interrelationships of watershed character-
istics: Jour. Geophys. Research, v. 66, p. 1215-1223. 1951, A graphic resume of the Pleistocene of Ne-
braska (with notes on the fossil mammalian remains) :
Hadley, R. F., and Schumm, S. A., 1961, Sediment sources Univ. Nebraska State Mus. Bull. 6, v. 3, 41 p.
and drainage basin characteristics in upper Cheyenne
River basin: U.S. Geol. Survey Water-Supply Paper Schultz, C. B., and Stout, T. M., 1945, Pleistocene loess de-
1531-B, p. 137-198. posits of Nebraska, in Symposium on loess, 1944: Am.
Holtz, W. G., and Gibbs, H. J., 1952, Consolidation and related Jour. Sci., v. 243, p. 231-244.
properties of loessial soils: Am. Soc. Testing Materials, 1948, Pleistocene mammals and terraces in the Great
Symposium on consolidation testing of soils, Spec. Tech. Plains: Geol. Soc. America Bull. 59, p. 553-588.
Pub. 126, p. 9-33. Schumm, S. A., 1956, Evolution of drainage systems and
Horton, R. E., 1945, Erosional development of streams and slopes in badlands at Perth Amboy, New Jersey: Geol.
their drainage basins, hydrophysical approach to quanti- Soc. America Bull. 67, p. 597-646.
tative morphology: Geol. Soc. America Bull. 56, p. 275- Schumm, S. A., and Hadley, R. F., 1957, Arroyos and the
370. semiarid cycle of erosion: Am. Jour. Sci., v. 255, p.
Kivett, M. F., 1949, Archaeological investigations in Medicine 161-174.
Creek Reservoir, Nebraska: Am. Antiquarian, v. 14, p. Searcy, J. K., and Hardison, C. H., 1960, Double-mass curves:
278-284. U.S. Geol. Survey Water-Supply Paper 1541-B, 65 p.
Kubiena, W. L., 1938, Micropedology: Ames, Iowa, Collegiate Smith, H. T. U., 1955, Use of aerial photography for inter-
Press, 243 p. pretation of dune history in Nebraska, U.S.A.: Internat.
Leonard, A. B., 1950, A Yarmouthian molluscan fauna in the Cong., 4th, Rome-Pisa, 1953, Assoc. for study of the
midcontinent region of the United States: Univ. Kansas Quaternary, 7 p.
Paleont. Contr., Mollusca, art. 3, 48 p. Strahler, A. N., 1952, Hypsometric (area-altitude) analysis
of erosional topography: Geol. Soc. Am. Bull. 63, p.
1951, Stratigraphic zonation of the Peoria Loess in 1117-1142.
Kansas: Jour. Geology, v. 59, p. 323-332. 1954, Statistical analysis in geomorphic research,
1952, Illinoian and Wisconsinan molluscan faunas in Jour. Geology, v. 62, p. 1-25.
Kansas: Univ. Kansas Paleont. Contr., Mollusca, art. 4, 1957, Quantitative analysis of watershed geomor-
38 p. phology: Am. Geophys. Union Trans., v. 38, no. 6, p.
Leopold, L. B., and Maddock, Thomas, 1953, The hydraulic 913-920.
geometry of stream channels and some physiographic 1958, Dimensional analysis applied to fluvially eroded
implications: U.S. Geol. Survey Prof. Paper 252, 57 p. landforms: Geol. Soc. America Bull. 69, p. 279-300.
Leopold, L. B., and Miller, J. P., 1954, A postglacial chron- Thorp, James, and others, 1952, Map of Pleistocene eolian
ology for some alluvial valleys in Wyoming: U.S. Geol. deposits of the United States, Alaska, and parts of
Survey Water-Supply Paper 1261, 90 p. Canada: Geol. Soc. America Bull.
1956, Ephemeral streams hydraulic factors and their U.S. Department of Agriculture, 1948, Grass, the yearbook
relation to the drainage net: U.S. Geol. Survey Prof. of agriculture, 1948: Washington, U.S. Govt. Printing
Paper 282-A, 37 p. Office, 892 p.
Lugn, A. L., 1935, The Pleistocene geology of Nebraska: U.S. Weather Bureau, 1955a, Substation history Nebraska:
Nebraska Geol. Survey Bull. 10, 2d ser., 223 p. U.S. Dept. Commerce, Key to meteorol. records docu-
Maner, S. B., 1958, Factors affecting sediment delivery rates mentation no. 1.1, 90 p.
in the Red Hills physiographic area: Am. Geophys. 1955b, Rainfall intensity-duration-frequency curves:
Union Trans., v. 39, p. 669-675. U.S. Dept. Commerce, Weather Bur. Tech. Paper 25, 53 p.
Melton, M. A., 1958, Geometric properties of mature drainage 1957, Rainfall intensity-frequency regime, Part 1,
the Ohio Valley: U.S. Dept. Commerce, Weather Bur. 1944, Nature and degree of recovery of grassland
Tech. Paper 29, 44 p. from the great drought of 1933-1940: Ecol. Mon., v. 14,
1959a, Evaporation maps for the United States: U.S. p. 393-479.
Dept. Commerce, Weather Bur. Tech. Paper 37, 13 p. Wedel, W. R., 1941, Environment and native subsistence
economies in the central Great Plains: Smithsonian
1959b, Climates of the States Nebraska: U.S. Dept.
Commerce, Climatography of the United States, no. Misc. Colln., v. 101, no. 3, 29 p.
60-25. 1949, Some provisional correlations in Missouri Basin
Archaeology: Am. Antiquarian, v. 14, p. 328-339.
Weakly, H. E., 1940, Tree-rings as a record of precipitation 1961, Prehistoric man on the Great Plains: Univ.
in western Nebraska: Tree-ring Bull. 6, p. 18-19. Oklahoma Press, Norman, Okla., 355 p.
1943, A tree-ring record of precipitation in western Wedel, W. R., and Kivett, M. F., 1956, Additional data on
Nebraska: Jour. Forestry, v. 41, p. 816-819. the Woodruff Ossuary, Kansas: Am. Antiquarian, v. 21,
Weaver, J. E., and Albertson, F. W., 1943, Resurvey of p. 414-415.
grasses, forbs, and underground plant parts at the end Wolman, M. G., 1955, The natural channel of Brandywine
of the great drought: Ecol. Mon., v. 13, p. 63-117. Creek, Penn.: U.S. Geol. Survey Prof. Paper 271, 56 p.
INDEX
[Italic page numbers indicate major references]
Page Page
A Page
Acknowledgments _._ _ __. 258 Carbon 14 age determinations .... 268, Drainage-basin measurements ___ 286
Adjusted channel frequency _. 286, 288 272, 273, 278, 281, 284, 300 Drainage, density _______ 286,288,328
Carbonates in loess, significance __ 277 evolution _____ _._ _ _ 282
Aerial photographs, use - . .. 274,
Carbonized grass _________ 274, 279, 280 ordering _____.__ _ __ 285
280, 286, 291, 298, 299, 304, 305, 309
Castings, worm ____-___-__.-__ 273,274 properties _______ _ - 294
Agricultural development of
Cedar Creek _________ 265,278,282 texture variation _ __. 288
basin _____._.__________ _ .___ ________________ 281
Ogallala exposures in ____ _ 265 transformation _ .___ 288
Alluvial fan .._.__ _.__.____ 294
terrace sequence __________ - 270 Droughts ________________ 281, 314
Alluvium _____________________________________ 268
Census ____ __ _______ ___ 281 ancient _ __ __.._ 263,264,274
Rates of accumulation
Channel, cross-sectional shapes __ 315 effects ___________________ 263
compared _____________________ 273,279
definition __________________ 286 Drought-resistant grass _______ _ _ 263
See also Banded alluvium;
frequency ___ 286,288,289,329, Dry Creek ____ 278,279,282,286,289,
Recent alluvium; Unhanded
332 294, 296, 298, 300, 315, 330,332
alluvium.
geometry ______ _.__.-_____ 315 Dunes ____________ 268,250,325
Altithermal, warm-dry climatic
hydraulics __ __ _ 315 Dust interception __ _____ - 262, 263
eposide _________ ___ ___ 284, 332
length ________ 286, 329, 330, 332 Dyksterhuis, E. J., cited _ .___ 258,262
Annual temperature __..____ _____..____.. 258
Anselmo loam ______.______ ____...______ 280 order _____ 286,289,328,330,332
scarp _____ 278,279,290,25^,332 E
Archeology ..._...._____....____...____.._.... 280
Arroyo cutting and filling _______ 314 shape ___ ____________ 321 Earth worms ________ 274,277,280
Artifacts _.__..._._______ 279, 281 slope __________ 286, 328, 329, 330 East Fort ________.___ - 263
See also Trenched channel. Elkhorn Canyon .____ 272,278,284
B Charcoal samples ___________ 273, 281 head scarp -_._ _ 274
Banded alluvium ____-._-____....__ ^_____. 279 Class A pan evaporation ___ .______ 261 Recent alluvium __ _ _ _ ._ 279
Basin, circularity .___ .____________..___... 329 Climate _______ 258, 260, 314, 332 terrace sequence and valley
length ________...____....______....__________ 286 archeological evidence . .._.._ 281 fills ______________ 273
measurement of areas ___.... ___ 286 effect on by Wisconsin Ice Elongation ratio ._____.__ _ 286,288
properties ____.-..________ -. ......___ 288 retreat _________________ 282 Eolian silt, Peorian Loess _ - 275
relief _____________ 286,288 Climax plant community __ __._ 262, Ephemeral streams _ _ 321,322,328,
shape -_....____....____ ..______. 288, 328 263,332 333
Bed material _______....______ ._______._._. 322 Colby loam ______________ .. 278,280 channels ________________ - _ 315, 318
Bibliography ._..______._____..____._______...___ 333 Colluvial soil __ ______ _. _________ 277 Erosion rate ___._______ ____- -___ ._ 330
Bifurcation ratio ____ ___ ___ _ 286 Conclusions __ _______ ___________ 332 Erosional scarps ._____-.____ 263,290,25i
Big bluestem ____......___-...____-____-..-_____. 262 Correlation of terraces ____. _ 268 Evaporation ____ __..____.__ 261,332
Bignell Loess ____ _._____.. 277, 278 Cowpaths _________________ 301, 303 Exposed rocks ^ __,__ _ ___ . 264
Blowouts .____-_____-________....__....______ 280 Coyote Creek ______________________ 294
Blue grama _______ . .._____......... 262, 263 Cretaceous bedrock __ ____-______ 265
Blue grass _ _._ ______-_____ _. 262 Crops, area in _______________ 333
Curtis, climatic data __ __ ___ 258 Fence posts, rate of burial - 279
Brady soil -______ __-___ 277,282
Curtis Canyon ________________ 279,294,297 Fire drives, use by
Branched gully head ___._.______ 301,307
Cut Canyon _____________ 272,282,294 prehistoric man ____-__ _- 280
Broscoe, A. J., cited __ _ ___ 289
First-order channels _ ___ 309,331
Brush, examples of thick ._________.. 263
D Flashy streams ___ _ _ 315
indicator of scrap activity .... 291
Flats __________________ 332
Brushy Creek . . .. 286,289,315, Deer Creek _ ____________ _______ 282 Formulas, landform _ _ _ 286
322,331 Deffenbaugh, J. L., cited _______________ 270 Fossils, list _____________ 271, 276
Buckbrush .__________.._.____.___..._....____.... 263 Dendrochronology -_ ______.264, 300
See also Index fossils.
Buffalograss .._______.______ _... 262, 263 Depth, relation to discharge ________ 316
Fox Creek _____ 265,271,278,279,286,
Bunchgrass ____. .___.__________________._______ 262 Discharge-weighted mean 289, 294, 315, 322, 331
Buried valleys -_ _____ _________ 282 concentration of suspended
Frye, J. C., and others, cited .__ 265
sediment...._____________________ 325
Discontinuous gullies _ 291,294,299 G
California method of Downstream hydraulic relations _ 315
plotting TT .. . .. _.... . ..._... 260 Drainage-basin characteristics, Gaging stations _________ 315,318,333
Cambridge ________-_ ____________ 279 interrelations _______ ____- 328 Geologic section, Cut Canyon .._ 272
337
Page Page Page

Geometric differences of ephemeral 0
and perennial streams ____________ 321
Land use _......_..._. 25J, £7.?, 330, 331 Ogallala Formation ______________ 265,
Geomorphic properties in relation
effects _______________ #03, 264, 301 270, 271, 274
to water and sediment
Landform properties ____________ 286 buried valleys in ___.____ - 282
discharge ______________________ 328
Leonard, >A.B., cited -. . - 268,271, Overgrazing _________________________ 263
Grading _______.__._______ ______ 282, 289
275, 276, 278 Oxidation _______________________ 280
Grand Island Formation __________ 270
Grass -_.-_-_-___.____.___ _ ____ _ __ 332 Leopold, L. B., and Maddock,
ability to prevent erosion ___ 263 Thomas,, quoted _________ . . __ _ 315
characteristics of native ~ 262 Lime Creek __ ______ 278, 281, 286, 289 Paleoclimate _-___-_--__ __ - 262
indicator of scarp activity ___ 291 Little bluestem ____ _____'.____.____ __ 262 Paleosol __...__-_-_____- _ 270
Gray, D. M., cited ________ 328 Load computations, B. R. Colby's Parabolic channel shape _ _ _ 321
Gross erosion ____-_ _-___ __ _ 279 - method ________________________________ . 325 Particle-size distributions of
Growth rings of trees __-______ ___ 300 Lobed gully head ...... ... ___ 301, 307 suspended sediment and bed
Gullies _-______ 264, 291, 300, 313 Loess ___..__..___._........ 256, 268, 274 material ____-___- 322, 333
advance ______________________ ______ 315 deposition rate ... _ _ ._.__. 277 Pearlette Ash Member of the
areal distribution -___ __ ___ 307 See also Bignell Loess. Sappa Formation _____________ 270, 282
coalescence _____________ 291, 299, 332 Loveland Formation ... 271,274,275, Pelecypod fauna _________ _ _ 278
control of ____________________ 314. 282 Peorian Loess __.___._ 268,271,272
Gully, definition and 275, 276, 277, 282, 332
classification ____._____.______ _ 290 M Peorian postdepositional landscape
erosion _ __._________ ____ _ 290 reconstructed ________________ 282
formation _____________ __________ 301 Maywood _ _______ _________ __ __ 265 Peorian valley heads _____ _.___ - 282
frequency distribution ___________ 309 McCoqk, climatic data __ _ _ 258 Percentage of area, in side slope . 330
man induced _______________________ 313 Mean channel length _ __ 289 in upland ______________ 328, 331, 333
measurement of advance _._.__ 298 Mean valley-side slope __ 288 in valley flat __________ ______ 330
profile changes _______________________ 296 Perennial streams _____ 315,321,333
Medicine Creek _.__.____ 256, 270,278,
size categories ________ ________ 305 Petrographic features, Peorian
slumping _________________________ 296 , 279, 315, 331, 332 Loess _______________________.... 277
See also Discontinuous gullies; Medicine Creek basin, climate 258 Platte Valley ____.._..... _..... -__--__ 282
Valley-bottom gullies; Niobrara Formation ____ __ 264 Pleistocene and Recent
Valley-head gullies; . Ogallala Formation _ _ __ 265 deposits _______ __.__.__ __ 265,268
Valley-side gullies. Melton, M. A., cited :__ __ 288 Pleistocene deposits ______ _______ 271
Gully head, branched ___ ____... 301, 307 Midgrass' _L _________ ___ 262,332 Pleistocene drainage, relic ______ 282
lobed _______________________ 301, 307 Mitchell Creek 286, 294, 315, 322, 331 Plum Creek ___ __________ .___.____ 282
Modern soil ______ ... L___ __...__ _______ 277 Plunge pools .__ _____ _.. 296, 312, 333
H Polar planimeter, use _______ _____ 286
Molluscan fauna, Peorian 275, 276
Harry Strunk Lake _______ 331, 332 Morisawa, M. E., cited _ ... .____. 329 Population _________________________________ 281
Niobrara Formation . ________ 264 Post-Sangamon, chronology _____ 284
Morphometric analysis ___________ _ 332
Head cut, use of term ______ 291 Prairie _______________________________ 332
Morphometry ________ _.___ _ __ 284.
Head scarps ____ 274, 297, 298, 303, 309 Precipitation _ _ 258, 263, 315, 326, 332
Mousel terrace __. _. _ _ 265, 278, 280, Pre-Pleistocene rocks ________ 264,268
advance -_-_____ __ 274, 290, 312
284, 332 Profile, Cut Canyon _..__ __ _______ _ 272
maximum height __________________ 311
carbon-14 dating _ ____. __ __ _ 284 main course of Medicine
Heavy minerals, Sangamon soil ___ 274
Holdredge soil ___________________________ 278 Woodland sites ___________ 280 Creek _____ ___________ 270
Horton, R. E., cited __________ 285,289 Multiple linear regression of Purpose of report _____ _____________ _ 255
Human occupation ____-_ ____ ___ 280 sediment discharge ___________ 331
Multiple regression analysis, R .
Hydraulic geometry _______________ 333
Hydraulic relations at a section properties used by Morisawa . 329, Radiocarbon dating. See Carbon-14
and in a downstream 333 age determination.
direction ________________ _ _ _ _ _ 315 Raindrop impact _______________________ 263
N Rainfall, intensity-duration-
frequency curves __-___-_ 260, 332
Native grasses ___________________________ 262
Index fossils ________ 268,271,272,275, interception __ _ _______ 262,263
Nebraskan deposits ___-_______ - 268 Ravines ____________________________ 262
276 correlation with ________________ 272
Inset terrace -_____ _____________________ 300 Recent alluvium ______ 265,272,273,
Needlegrass ____________________________ 262 278, 279, 280, 284, 332
K Niobrara Formation ____.___ _ ___ 264 Recent deposits. See Pleistocene and
North Platte. ancient Recent deposits.
Kansan deposits ___-___i_-_________ 268 droughts at _____________________ 264 Relief ratio _______________ 286,288,328,
Kansan Glaciation __,-______. 271,282 climatic data ___ _-____________ 260 329, 331, 333
Kulp, J. L., quoted _______________ 258,278 North Plum Creek ___________ 282 Republican River valley _________ 282
INDEX 339
Page Page Page
Reservoir regulation _ 315,318 Stream, capture by Deer Creek _.__ 282 Valley fill, Elkhorn Canyon ________ 273
Rodent burrows __________________. 312 hydraulics ____________________ 315 Peorian __ ____________ ___ 276
Rubin, Meyer, cited _______ 258,273 ordering method _._ '285, 286, 289 Valley flat ___________ 279, 329, 330
Runoff _____ 326,328,329,330,331 Stream lengths, R. E. Horton's Valley-head gullies _______ 291, 301,
Runoff, reduction ..._________________ 314 law of ___________________ 289 303, 307, 313,332,333
Upland, definition _________________ 286 formation _________ ____ 309
S Summary _______________________ 332 measurement ___________ 304, 309
Sand dunes ________________ 268,250,325 Sumac ____________________ 263 Valley incision, Ogallala
Suspended sediment - - 279, 315,321, Formation ______________ ._ 282
Sangamon soil _______ 271,274,282
325, 328, 329, 330, 331, 333 Valley-side gullies _____ 291,501,303,
Sappa Formation _________ 270,271,282
affect on hydraulic geometry 307,313,332,333
Saturation ____________________. 296, 312
of channels . ___ ___ ___ 315 formation ___,_______ __ 309
Scarps ____________________________ 290
contributions of major measurement ____ _______ __ 304
See also Channel, scarp;
tributaries ___________________ 325 Valley, slope ___________ ___.._____-_ 286, 329
Erosional scarps; Head
particle-size distributions ___ 322 trenching ___ _________________ 278
scarps; Step scarps; Side
Switchgrass ____________________ 262 Vegetation ________________ 262, 314, 332, 333
scarps.
role of, in gully control _____ 314
Schultz, C. B., and others, cited __ 265
Velocity, relation to discharge __ 316
Schumm, S. A., cited ________________ 288
V-shaped contour lines _______________ 286
Scour __________________ 295 Tall grass ______________________ 262
Settlement __________________ 281 Tazewell Stade, Wisconsin W
Short grass ___ _-___ _____ 262,332 Glaciation __________________ 275
Side-oats grama _ ___ __ _-___-_ 262 Temperatures __________________ 258 Water-discharge functions .__ _ 316
Side scarps -._ __ ______ ______ 291 Terraces, Cedar Creek __________ 270 Water-sediment discharge
Silt ___________________ 333 Elkhorn Canyon ____________________ 273 relation _____________ _______ 322
Slocum, J. L., quoted ___________________ 262 rate of formation ________ _ 274 Wayne, W. J., cited _ 258,271,275,
Slumping ____________ 298, 312 sequence ___ _______________ 265 276
Snails ______ 268,271,275,276,278, See also Inset terrace; Weakly, H. E., cited ___.___.__ 264
332 Mousel terrace; Stockville Weaver, J. E., and Albertson,
Sod _____________________ 263 terrace; Wellfleet terrace. F. W. cited ____________________ _ 263
effects of breaks in ___.___. 295 Topographic maps, use _____ 258,286, Well Canyon ... ... 265,278,279,286,
scarps ___________-_-___ _ 263 288,294,298 289,294, 330
Soil marker _____ ______ 268, 274 Trees ____________________ 262, 264 Wellfleet terrace ______ 265,268,280
Station hydraulic relations __________ 315 Trenched channel _______________ 279 Wheatgrass _________________ 262
Step scarps ______________ 264,290 Width of stream, relation to
U discharge _____._ _ ___ _ ____ 316
Stephenson, D. V., quoted ______ 262
Stockville terrace ______ 265, 268, 272, Unbanded alluvium .__.__.__ ____ 279 Wild plum -___ ______ _______ 263
978, 282, 286, 291, 292,311, 313, 332 Unexposed rocks __ ____________ 264 Wild rose ___________ ______. 263
age ________________ 281 Upland ___________ 286, 288, 330 Wimpf, Charles, quoted _._ _______ 262
carbon-14 dating ________ _ 284 slope _________________ 330 Wind speed and direction __. ___ 261
Early Lithic sites ___________ 280 Upper Republican Culture __ 274, 281 Windblown dust ________.. .__ _ 263
Strahler's, A. N., method of U-shaped dunes ______________ 280 Wisconsin deposits _.__ __-_____. __ 268
stream ordering ____ 285,286,289 Wisconsin ice, retreat _______________ 282
Stratigraphy, Cedar Creek near Worm burrows and casting _.__ 273, 277
Stockville ________________ 270 Valley-bottom gullies __ 264,291,295,
Cut Canyon near Curtis - 271 297, 298,307, 309, 313, 314, 332, 333
Pleistocene and Recent _ 268 Valley bottoms __________________ 264 Yarmouth soil _________ 270,271
U.S. GOVERNMENT PRINTING OFFICE: 1966 O 2I7-S78

BRICE 1966 Rill Gully

Uploaded by

Copyright:

Available Formats

BRICE 1966 Rill Gully

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

BRICE 1966 Rill Gully

Uploaded by

Copyright:

Available Formats

Erosion and Deposition in

the Loess-Mantled Great

Prepared as part of a program of the

EROSION AND SEDIMENTATION IN A SEMIARID ENVIRONMENT

GEOLOGICAL SURVEY PROFESSIONAL PAPER 352-H

Prepared as part of a program of the

UNITED STATES GOVERNMENT PRINTING OFFICE, WASHINGTON : 1966

For sale by the Superintendent of Documents, U.S. Government Printing Office

Page -i .'. Page

ABSTRACT change in width. Curves relating suspended-sediment load

Boundary of report area

FIGURE 176. Location map of Medicine Creek basin.

5-year moving average (plotted at

FIGURE 179. Annual precipitation at Curtis, Nebr., 1895-1960.

30 5-year moving average

1895 1900 05 10 15 20 25 30 35 40 45 50 1960

FIGURE 180. Annual precipitation at McCook, Nebr., 1895-1960.

the notes of Deputy Surveyor D. V. Stephenson,

valley bottoms. The termination of the small valley-

Mousel /*' st°c kvil| e

FIGURE 187. Profile and geologic section on Cedar Creek.

DATUM IS MEAN SEA LEVEL

Layer from which

DATUM IS MEAN SEA LEVEL

FIGURE 189. Profile and geologic section on Elkhorn Canyon.

alluvium is exposed in the walls of rapidly advancing

where in the basin and constitute a reference surface

CONTOUR INTERVALS 10 AND 20 FEET

fourth order. In a sense the change in slope of the

:§< ^irnw'X- ?^^^$$*

FIGURE 201. Head scarps of major valley-bottom gullies.

Upper Cut Canyon

Upper left. Gully in main channel of Dry Creek, designated gully Bl in

FIGURE 204. TYPICAL VALLEY-BOTTOM GULLIES

500 1000 1500 2000

20 40 60 80 100 120 FEET

VERTICAL AND HORIZONTAL SCALE OF CROSS PROFILES

1000 2000 3000 4000 5000 6000

Profile of valley in 1956

Probable profile of channel in 1937

4000 3500 3000 2500 2000 1500 1000 500

Stockville terrace East Fork Dry Creek

Flow line, 1951 channel

Position of channel scarp in 1937

VERTICAL AND HORIZONTAL SCALE

Stockville terrace Valley flat Inset terrace

FIGUGE 210. DOWNSTREAM CHANGES IN CHANNEL OF DRY CREEK

- V < *.- jn /W» IJ** I'm V'.IV/ *

MEASUREMENT OF VALLEY-HEAD AND tion of volume was complicated by the irregular

A satisfactory approximation to the measured Small _____ 50-100 63

PERCENTAGE ENLARGEMENT PERCENTAGE ENLARGEMENT

Valley head ____..._____ Small ___ _____ 1-15 47 70 39 36 417

Frequency of gullies per square

GULLY FREQUENCY \ 5.2 10 '.

EXPLANATION - _______ <

o2-1 "o 83- 3 2.1

10 Subbasin 1.5 1.2

Number beside symbol refers to gully frequency

Valley head __..._ Small _ _____ 1-15 47 70 39 36 417