Skip to main content

Full text of "DTIC ADA363623: Laboratory Investigation of the Kansas River Bend and Kansas City Reach, Conducted at Mead Hydraulic Laboratory, Mead, Nebraska"

See other formats


LABORATORY INVESTIGATION OF THE 
KANSAS RIVER BEND AND 
KANSAS CITY REACH 


MEAD HYDRAULIC LABORATORY 
• MEAD, NEBRASKA 




U. S. ARMY ENGINEER DISTRICT, OMAHA 
U. S. ARMY ENGINEER DISTRICT, KANSAS CITY 
MISSOURI RIVER DIVISION, OMAHA 
DECEMBER 1971 


OrSTRIBUTTON STATEMfiMtM 

for Public ReleaS^ 
Distribution Unlimitad 



19990525 064 


Form Approved 
0MB No. 0704-0188 




TAT 


Public reportinq burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, 
aatherinq and maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this 
collection of information, including suggestions for reducing this burden, to Washington Headquarters Services, Directorate for Information Operations and Reports, 1215 Jefferson 
Davis Highway Suite 1204 Arlington, VA 22202-4302, and to the Office of Management and Budget, Paperwork Reduction Project (0704-0188), Washington, DC 20503. 


1. AGENCY USE ONLY (Leave blank) \ 2. REPORT DATE 




3. REPORT TYPE AND DATES COVERED 


4. TITLE AND SUBTITLE , ^ 



7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) 


VfMLKpiA ^drOUilfCy 
^ A/E- 



8. PERFORMING ORGANIZATION 
REPORT NUMBER 


9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) 

^\£b/v. 

Om>^K ^ I ^ 


11. SUPPLEMENTARY NOTES 

a.5. 


Kan^ . /no 

(Ytr sjVin 


10. SPOKSORIWO/MONITORIWG 
AGENCY REPORT NUMBER 

vvil^D \\udraAx.nk. 

lc,bo>-a3r^S£n^^ 
fVo. & 


12a. DISTRIBUTION/AVAILABILITY STATEMENT (9/0 ^ A 

fod^ 


12b, DISTRIBUTION CODE 


13. ABSTRACT (Maximum 200 words) 


/N.-i KiVer yn)csoi.^n kHk 

Dfd bi^oh. urjyr iU^ 

^ra l su^sto<^ o/ 4U- aiJ.-fU 

Kiytr yfpT'o/?. 


15. NUMBER OF PAGES 

2J) 


16. PRICE CODE 


I’ rea.cW(?s 

Qj^o.f\nc{s 
rvLfer bgAdig> 

17. SECURITY CLASSIFICATION 18. SECURITY CLASSIFICATION'TToT'SECUmV CLASSIFICATION | 20. LIMITATION OF ABSTRACT 
OF REPORT A OF THIS PAGE . OF ABSTRACT 

MSN 7540-01-280-5500 Standard Form 298 (Rev. 2-89) 

Prescribed by ANSI Std. Z39-18 
298-102 












































GENERAL IMSTRUCTIO^JS FOR CQfVfPLETifMG SF 298 


The Report Documentation Page (RDP) is used in announcing and cataloging reports. It is important 
that this information be consistent with the rest of the report, particularly the cover and title page. 
Instructions for filling in each block of the form follow. It is important to stay within the lines to meet 

optica! scanning requirements. 












DEPARTMENT OF THE ARMY 

CORPS OF ENGINEERS 

LABORATORY INVESTIGATION OF THE KANSAS CITY REACH 

AND KANSAS RIVER BEND 

Conducted at 

Mead Hydraulic Laboratory 
Mead, Nebraska 


U. S. Army Engineer District, Omaha, Nebraska 
U. S. Army Engineer District, Kansas City, Missouri 
Missouri River Division, Omaha, Nebraska 


December 1971 


TABLE OF CONTENTS 


Page 


Introduction . 1 

Description of the Kansas City Reach and the 

Kansas River Bend . 2 

Purpose of Study . 2 

Model Design and Verification . 3 

Test Procedure . 6 

Description of Kansas River Bend Tests . 8 

Analysis of Model Results for Kansas River Bend Tests . 13 

Description of Kansas City Reach Tests . 15 

Analysis of Model Results for Kansas City Reach Tests . IT 

Conclusions . 20 


REFERENCES 


1. The Committee of the Hydraulics Division on Hydraulic Research, 
American Society of Civil Engineers, "Hydraulic Models," 

July 23, 1942. 


2. Missouri River Division, Corps of Engineers, "Operation and 
Function of the Mead Hydraulic Laboratory." MRP Hydraulic 
Laboratory Series Report No. 1 , March I969. 


3. Missoxiri River Division, Corps of Engineers, "Laboratory Investi 
gation of Underwater Sills on the Convex Bank of Pomeroy Bend." 
MRP Hydraulic Laboratory Series Report No. 2 , November I966. 












LIST OF FIGURES 


Figure 

Number Description 

!• Photograph of Kansas City Model in Operation. 

2. Map of Reach. 

3. Size Distribution of Prototype and Model Bed Materials. 

4. Photomicrographs of Bed Materials. 

5. Photograph of the Kansas - Missouri River Confluence 
After the 1951 Flood. 

6 . Test Locations for Dike and Sill 391.5- 


LIST OF PLATES 


Plate 

Number 

1 . 

2 . 

3. 

4. 

5. 

6 . 


7 . 


8 . 


Description 

Sketch of Modeled Reach. 

Bed Maps - Model Verification. 

Kansas River Bend Bed Maps - Left Bank Dike and Sill Tests. 

Kansas River Bend Bed Maps - Low Elevation Sill Tests. 

Kansas River Bend Bed Maps - Minor Realignment Tests. 

Kansas City Reach Bed Maps - Varying Degrees of Channel 
Confinement. 

Kansas City Reach Bed Maps - Vane Dike Model Tests. 

Map of Reach Showing Model Designed Structures. 


LABORATORY INVESTIGATION OF THE KANSAS CITY 
REACH AND KANSAS RIVER BEND 


INTRODUCTION 


This report describes the tests and results of a model study con¬ 
ducted at the Mead Hydraulic Laboratory of the Kansas City Reach and 
the Kansas River Bend of the Missouri River, The study was performed 
by personnel of the Hydro-Sediment Section of the Omaha District, 
Corps of Engineers, under the general supervision of the Kansas City 
District and the Missouri River Division, 



Figure 1. Photograph of Model Taken While Operating. 

Attempts to improve and control the Missouri River have been in 
progress for many years. Dikes, sills, and revetments of various 
arrangements have been constructed to control the overall river align¬ 
ment and to insure a river channel of adequate depth and width to 
permit navigation. The channel alignment of the Missouri River between 
Sioux City, Iowa and its mouth at St. Louis, Missouri, has in general 
been established with the shape of each major bend controlled by a com¬ 
bination of spur dikes and bank revetments. However, many times prob¬ 
lems still appear within this general alignment. Maintaining an 
adequate navigation channel through the Kansas City Reach and the 
Kansas River Bend is a typical problem. 



DESCRIPTION OF THE KANSAS CITY REACH 
AND THE KANSAS RIVER BEND 


This river reach is located within the Kansas City city limits 
and extends from Missouri River mile 363.5 to 368. The reach 
selected for the model extended from river mile 363 to 369 as shown 
on Figure 2. The concave (outer) banks are completely revetted with 
rock, wood piling, or concrete retaining walls and are therefore 
stable boundaries. The convex bank, or inside of the bend, is con¬ 
trolled by a series of spur dikes spaced intermittently throughout 
the reach. "L-Head" dikes have been constructed from the ends of 
some of the spur dikes. At normal river stages, the dikes are above 
water surface elevation. The "L-heads," however, have generally been 
constructed to the established Construction Reference Plane (CRP) 
which is 2 to 3 feet below normal water surface elevation. As of 
October I 966 , sills (also constructed to CRP) existed on the first 
and third dikes above the Kansas River Bend. These sills confined 

that portion of the river to about 600 feet. The channel width in 

the Kansas City Reach remained at about 8 OO feet. The only signifi¬ 
cant inflow within the limits of this reach is the Kansas River which 

enters the Missouri River on the right bank of the Kansas River Bend 

at Missouri River mile 367 . 5 . A location map of the study reach is 
shown on Figure 2, while Plate 1 shows the location of m1 existing 
control structures as of October I 968 . 

A review of hydrographic surveys reveals that a deep and narrow 
channel generally exists on the right bank of the Kansas River Bend, 
and a point bar often develops directly across from the mouth of the 
Kansas River. This restriction frequently results in a split channel 
at the confluence of the two rivers, with neither channel having dimen¬ 
sions suitable for navigation. The channel remains narrow throughout 
the remainder of the Kansas River Bend. Immediately below this bend 
in the upper segment of the Kansas City Reach near the Broadway Bridge 
(river mile 366.1), the flow leaves the right bank and a more unifonn 
cross-section developes. Continuing downstream, the channel meanders 
between the banks leaving scattered sand bars throughout the remainder 
of the Kansas City Beach. 


PURPOSE OF STUDY 


Design criteria for the effective control of river reaches similar 
to the Kansas River Bend and the Kansas City Reach are very limited. 

L-heads," low elevation sills both level and sloping, dike extensions, 
and vane dikes are types of control that may be effective. Each type 
of structure may have several successful combinations of structure 


2 


length, elevation, orientation and slope. Because of the many pos¬ 
sible solutions and the unknown results of each, the above reaches 
were selected for a detailed laboratory investigation. Emphasis 
was placed upon the selection of an effective system of structures 
that would develop the desired navigation channel both through the 
Kansas River Bend and the Kansas City Reach. 



Figure 2. Location Map of Study Reach. 

MODEL DESIGN AND VERIFICATION 


The bed material for this model, like previously completed studies 
at the Mead Hydraulic Laboratory, was finely ground walnut shells. The 
gradation and particle shape of this material is very similar to the 
sand found in the bed of the Missouri River, as illustrated in Figures 
3 and U. This material resembles and responds very much like sand, but 
has a specific gravity of only 1.33 compared to 2.65 for sand. 

This light-weight bed material permitted the model to be operated at 
low flow velocities, and still maintain a relatively high suspended sediment 
load which resulted in a desirable time-scale ratio. The bed material was 
transported both as bed and suspended load. 


3 




Figixre 3. Size Distribution of Prototype and Model Bed Materials 



Walnut Shells 


Missouri River Sand 


Figure it. Photomicrograph of Test Materials, Grid Size: 0.39 nim 






Verification of the model involved the determination of a set 
of model dimensions by vhidi the model would respond to structure 
changes in much the same manner as the prototype's response to 
similar changes. First, the length of the reach to be modeled was 
selected and a sufficient flume length added to both the upper and 
lower segments of the reach to minimize the Influences of undesirable 
entxance and exit conditions. Ibe horizontal scale, or length radio, 
was then determined by fitting the largest possible model within the 
physical limits of the laboratory building, ihe vertical scale, or 
height dimension, and the velocity scale were determined by a series 
of verification tests. 

Use of the Froude (1) relationship pirovldes a basis for logical 
selection of a reasonable opeimting depth and computation of the 
velocity ratio. However, in a movable bed model, an additloxial vari¬ 
able is present that does not appear in the Froude Number; the sedi¬ 
ment transport both in suspension and along the bed. Transport rates 
are a function of variables such as stream depth, velocity, individual 
grain characteristics, bed forms, int«isity of turbulence, and the 
width-depth ratio of the stream. The horizontal scale ratio neces¬ 
sitated by the physical, dimensions of the building dictated that a 
distorted model be employed(S). The distortion in channel dimensicois 
also results in distortions of such things as the bed forms and scour 
hole dimensions. 

Preliminary verification tests revealed that the Froude Criteria 
could not be strictly adhered to. Results of prototype surveys 
indicated that a very confined channel existed throughout the Kansas 
River Bend and that a meandering channel was the principal problem 
in the Kansas City Reach. It was ii^perative that the model reproduce 
these conditions. A series of tests was conducted in which operating 
depths and velocities were varied. Changes in these paiametexa were 
made until the model satisfacto]i.ly reproduced the river as shown by 
the bed maps on Plate 2. Using the results of these tests as a guide, 
a complete set of model scenes was then adopted. Because the average 
discharge dviring the navigation season is essentially the dominant bed 
forming discharge, nearly all of the subsequent tests utilized only 
this one flow. Exceptions to this are defined later under "Description 
of Tests." No attempt was made to reproduce a seasonal runoff, since 
a completely new set of scale relationships may have been necessary 
for each dlsdiarge involved. The scale relationships adopted for 
this model study are listed in Table 1. 


5 


TABLE 1 

MODEL - PROTOTYPE RELATIONSHIPS 



Missouri 

River 

Model 

Run No. 18 

Scale Ratio 

Prot otype/Model 

Discharge, c.f.s. 

40,000 

0.486 

82,300 

Average Depth, ft. 

10 .T 

0.198 

54 

Channel Width, ft. 

875 

5.8 

150 

Average Velocity, ft/sec 

4.76 

0.428 

11.1 

Slope, (ft/ft) X 10 ^ 

1.9 

7;2 

0.26 

Manning's "n" 

0.021 

0.032 

0.66 

Specific Gravity of Bed Material 

2.65 

1.33 

1.99 

D 35 , mm 

0.26 

0.23 

1.13 

D 50 , mm 

0.30 

0.26 

1.15 

D65, mm 

0.38 

0.30 

1.27 

Froude No. 

0.257 

0.169 

1.52 

Sediment Time Ratio* 



28 * 


Ratio based on measured prototype and model sediment transport rates. 


TEST PROCEDURE 


The basin used for this model study was a closed system in which 
both the water and transported sediments were recirculated. The water 
depth was controlled by regulating the stage at the midpoint of the 
model. No tailgate or depth control structure of any kind was usedj 
therefore,the water surface and bed slopes were free to adjust. Labor¬ 
atory testing procedures are further described in "MRD Hydraulic 
Laboratory Series Report No. 1 ”'^). 


6 


Table 2 is a siammary of the measured data and the hydraulic com¬ 
putations made for each test. The basic parameters; discharge, aver¬ 
age depth, energy slope, and transported sediment concentrations are 
average values of measurements taken during or after a test. The 
remaining items are functions of these basic quantities and can be 
used to compare one test with another. The ability of a given arrange¬ 
ment of structures to develop the desired navigation channel indicates 
the degree of control. Methods of relating these channel conditions 
to basic hydraulic calculations are very difficult. Because most of 
the tests were operated at basically the same depth and discharge, 
changes in the bed formations or hydraulic functions were assumed to 
be the result of changes in the structure layout. A description of 
each of the items shown in Table 2 is as follows: 


Columns (l) & (2) 


Model Test Nxnnber 


Columns (3) & (^) Prototype Discharge Represented 

Columns (5) & (6) Model Flows 


Coliimn (T) 


Colximn (8) 


Column (9) 


The average flow area of the measured 
cross sections. Twenty-three 
Missouri River sections were used to 
establish this area. 

Mean depth was determined for each 
section by dividing the flow area by 
top width. The mean depth for the 
model was determined by averaging 
the mean depths of all sections. 

Average velocity - Discharge divided 
by flow area. 


Column (lO) 


Energy slope observed in the model at 
the completion of each test. 


Column (ll) 


Column (12) 


Suspended sediment concentrations in 
parts per million by weight. Refer¬ 
ence 1 explains methods used to col¬ 
lect samples. The values presented 
here are the results of measurements 
made after the model had reached equi¬ 
librium. 

Manning’s "n” = 1.H86 . A . D^^^ . 

Q 


sl/2 


7 


Column (12) - Cont'd Where Q = Discharge in c.f.s. 

A = Average flov Eirea 
D = Average depth 
S = Energy slope 

Column (13) Froude Number F = V/ ^gD" represents 

the ratio of inertia forces to the 
gravitational force as they existed in 
the model. 


DESCRIPTION OF KANSAS RIVER BEND TESTS 

Many types of structures vere tested in an attempt to increase the 
width of the navigation channel throughout the Kansas River Bend. These 
included spur dikes, low elevation sills, vane dikes, low elevation sills 
extended from the concave bank, and minor channel realignment. Studies 
were also made to determine the best location of each of the test struc¬ 
tures . A description of the prototype problems and attempted model solu¬ 
tions follows. 

The narrow navigation channel in the Missouri River near the mouth of 
the Kansas River has been a continuing problem. Nearly all hydrographic 
maps indicate a deep, narrow channel at the confluence. This condition 
is the result of the point bar that develops near the left bank and ex¬ 
tends riverward. Occasionally, the degree of confinement develops to 
such an extent that a secondary channel develops along the left bank, 
with neither channel having suitable dimensions for navigation. Test 

391.95 (see Plate 3 ) was placed in the model in an attempt to 
prevent the development of this secondary channel. The tendency for the 
development of the secondary channel cam be seen on bed maps of the model 
verification runs shown on Plate 2 , 

A second problem in the Kansas River Bend occurs after higher than 
normal Kansas River flows. The Kansas River normally contributes only 
very low flows, but short—duration rises in the spring and summer are 
common. Because of the angle at which these two rivers merge, the higher 
Kansas River flows erode the left bank accretion below the confluence, 
and deposit the material throughout the downstream reaches. Figure 5 Is 
an aerial photograph of the Kansas River Bend after the record flood of 
1951. Although the left bank erosion depicted in this photograph is 
much more severe than that caused by the normal short duration increases 
in flow, it demonstrates the river's ability to erode the accretion 
material from this bank. Even though the erosion caused by these Kansas 


8 



TABLE 2 


SUMMARY OF HYDRAULIC COMPUTATIONS 


Run Number 
Series Test 

Discharge 

Prototype 

Missouri R. Kansas R. 

in C.F.S. 

Model 

Missouri R. Kansas R, 

Average 
Flow Area 
{ft2) 

Mean- 

Depth 

(ft) 

Average 

Velocity 

(ft/sec) 

Slope 

xlO^ 

Transported 
Sediment Cone 
(ppm) 

Mannings 

"n" 

Froude 

Number 

(1) 

(2) 

(3) 

(»♦) 

(5) 

(6) 

(7) 

(8) 

(9) 

(10) 

(11) 

(12) 

(13) 

1 

38,000 

2,000 

0.71*8 

0.038 

1.639 

0.286 

0.479 

5.6 


0.032 

0.158 

2 


38,000 

2,000 

0.1*12 

0.021 

1.338 

0.243 

0.323 

1.6 

12 

0.023 

0.116 

3 


38,000 

2,000 

0.6il* 

0.031 

1.252 

0.228 

0.514 

6.4 

284 

0.027 

0.190 

k 


38,000 

2,000 

0.500 

0.025 

1.275 

0.237 

0.411 

4.1 

68 

0.028 

0.i49 

5 • 


38,000 

2,000 

0.1*81* 

0.024 

1.069 

0.200 

0.475 

5.8 

180 

0.026 

0.187 

6 


38,000 

2,000 

0.1*60 

0.023 

1.080 

0.204 

0.447 

5.2 

26 

0.026 

0.174 

7 


38,000 

2,000 

0.1*60 

0.023 

1.136 

0.201 

0.425 

6.6 

233 

0.031 

0.167 

8 


38,000 

2,000 

0.1*60 

0.023 

1.063 

0.191 

0.454 

7.3 

362 

0.029 

0.182 

9 


38,000 

2,000 

0.1*01 

0.020 

1.238 

0.213 

0.34o 

7.0 

254 

0.04l 

0.130 

10 


38,000 

2,000 

0.1*10 

0.020 

0.935 

0.171 

0.460 

7.9 

583 

O.O2S 

0.196 

11 


38,000 

2,000 

0.1*62 

0.023 

1.245 

0.228 

0.390 

4.3 

32 

0.029 

0.144 

12 


38,000 

2,000 

0.1*62 

0.023 

1.100 

0.205 

0.441 

7.1 

169 

0.031 

0.172 

13 


38,000 

2,000 

0.1*62 

0.023 




5.8 

100 


0.211 

Ih 


38,000 

2,000 

0.530 

0.030 

1.085 

0.186 

0.516 

10.1 

1280 

0.030 

15 


38,000 

2,000 

0.1*62 

0.024 

1.096 

0.201 

0.444 

8.6 

419 

0.034 

0.175 

16 


38,000 

2,000 

0.1*62 

0.024 

1.030 

0.180 

0.471 

8.4 

564 

0.030 

0.192 

17 


38,000 

2,000 

0.1*62 

0.024 

1.145 

0.196 

0.424 

6.4 

324 

0.030 

0.168 

18 


38,000 

2,000 

0.1+62 

0.024 

1.133 

0.198 

0.428 

7.2 

315 

0.033 

0.169 

19 

A 

38,000 

2,000 

0.1*62 

0.024 

1.247 

0.211 

0.390 

6.2 

92 

0.034 

0.150 

B 

38,000 

2,000 

0.1+62 

0.024 

1.115 

0.192 

0.436 

6.5 

147 

0.029 

0.174 


c 

38,000 

2,000 

0.1+62 

0.024 

1.139 

0.204 

0.426 

5.6 

77 

0.029 

0.166 


D 

38,000 

2,000 

0.1*62 

0.024 

1.165 

0.203 

0.417 

6.5 

3l4 

0.031 

0.163 


E 

19,000 

38,000 

0.231 

0.462 




5.8 

880 




F 

38,000 

2,000 

0.462 

0.024 

1.128 

0.188 

0.431 

8.6 

420 

0.033 

0.175 


G 

36,000 

2,000 

0.462 

0.024 

1.162 

0.197 

0.417 

7.5 

230 

0.033 

0.166 


H 

38,000 

2,000 

0.462 

0.024 

1.134 

0.195 

0.428 

7.2 

250 

0.031 

0.171 


I 

38,000 

2,000 

0.462 

0.024 

1.067 

0.184 

0.456 

7.4 

420 

0.029 

0.187 

20 

A 

38,000 

2,000 

0.462 

0.024 

1.115 

0.192 

0.436 

7.2 

310 

0.030 

0.175 

B 

38,000 

2,000 

0.462 

0.024 

1.035 

0.178 

0.468 

7.7 

540 

0.028 

0.195 


C 

38,000 

2,000 

0.462 

0.024 

1.044 

0.178 

0.465 

7.8 

330 

0.028 

0.194 


D 

38,000 

2,000 

0.462 

0.024 




7.2 

300 



21 

A 

38,000 

2,000 

0.462 

0.024 

1.152 

0.202 

0.421 

5.9 

150 

0.030 

0.165 

B 

38,000 

2,000 

0.462 

0.024 

1.176 

0.205 

0.413 

6.2 

310 

0.031 

0.161 


C 

38,000 

2,000 

0.462 

0.024 

1.128 

0.199 

0.431 

6.0 

l40 

0.029 

0.170 

22 

A 

71,000 

2,000 

0.668 

0.024 

1.572 

0.251 

0.567 

5.5 

440 

0.025 

0.194 

B 

38,000 

28,000 

0.462 

0.347 

1.606 

0.259 

0.504 

4.7 

500 

0.028 

0.174 


c 

38,000 

iu,ooo 

0.462 

0.173 

1.348 

0.224 

0.472 

6.2 

460 

0.029 

0.175 


D 

38,000 

2,000 

0.462 

0.024 

1.213 

0.208 

0.400 

7.0 

220 

0.036 

0.155 


E 

38,000 

2,000 

0.462 

0.024 

1.242 

0.205 

0.391 

5.8 

110 

0.032 

0.152 

23 

A 

38,000 

2,000 

0.462 

0.024 

1.158 

0.199 

0.419 

6.7 

200 

0.031 

0.166 

B 

38,000 

2,000 

0.462 

0.024 

1.198 

0.202 

0.405 

6.8 

330 

0.033 

0.159 


c 

38,000 

2,000 

0.462 

0.024 

1.221 

0.208 

0.397 

6.5 

400 

0.033 

0.153 


D 

38,000 

2,000 

0.462 

0.024 

1.222 

0.206 

0.397 

6.4 

280 

0.033 

0.154 


E 

38,000 

2,000 

0.462 

0.024 

1.208 

0.207 

0.402 

7.9 

350 

0.036 

0.156 


F 

38,000 

2,000 

0.462 

0.024 

1.205 

0.206 

0.403 

7.4 

420 

0.035 

0.157 


G 

38,000 

2,000 

0.462 

0.024 

1.126 

0.199 

0.431 

8.3 

450 

0.034 

0.170 

2 U 

A 

38,000 

2,000 

0.462 

0.024 

1.280 

0.215 

0.379 

6.3 

210 

0.035 

0.142 

B 

38,000 

2,000 

0.462 

0.024 

1.203 

0.204 

0.403 

8.2 

480 

0.036 

0.157 


C 

38,000 

2,000 

0.462 

0.024 

1.219 

0.206 

0.398 

7.2 

46o 

0.035 

0.155 

25 

A 

38,000 

2,000 

0.462 

0.024 

1.143 

0.195 

0.424 

8.4 

530 

0.034 

0.171 

B 

38,000 

2,000 

0.462 

0.024 

1.168 

0.199 

0.4o8 

7.6 

310 

0.035 

0.161 


C 

38,000 

2,000 

0.462 

0.024 

1.225 

0.202 

0.396 

7.4 

190 

0.035 

0.155 

26 

A 

38,000 

2,000 

Q. 462 

0.024 

1.201 

0.198 

0.4o4 

7.6 

390 

0.035 

0.160 

B 

38,000 

2,000 

0.462 

0.024 

1.204 

0.198 

0.403 

6.5 

210 

0.032 

0.160 


c 

38,000 

2,000 

0.462 

0.024 

1.244 

0.204 

0.390 

6.7 

150 

0.034 

0.152 


D 

38,000 

2,000 

0.462 

0.024 

1.170 

0.196 

0.415 

7.0 

250 

0.032 

0.165 

27 

A1 

38,000 

2,000 

0.462 

0.024 




7.4 

110 



A2 

38,000 

2,000 

0.462 

0.024 




6.8 

260 

0.034 

0.146 


A6 

38,000 

2,000 

0.462 

0.024 

1.265 

0.215 

0.384 

6.1 

210 


A7 

38,000 

2,000 

0.462 

0.024 




7.2 

130 




B2 

60,000 

2,000 

0.729 

0.024 




7.0 

340 




Bh 

60,000 

2,000 

0.729 

0.024 




7.9 

490 




C 2 

38,000 

15,000 

0.462 

0.182 




5.2 

30 




Ch 

38,000 

15,000 

0.462 

0.182 




6.4 

370 




D1 

38,000 

30,000 

0.462 

0.365 




5.3 

340 




D2 

38,000 

30,000 

0.462 

0.365 




5.5 

340 




D3 

38,000 

30,000 

0.462 

0.365 




6.4 

380 




DU 

38,000 

30,000 

0.462 

0.365 




5.8 

4l0 




El 

38,000 

1*5,000 

0.462 

0.547 




5.5 

480 




E2 

38,000 

1*5,000 

0.462 

0.547 




5.3 

570 




E3 

38,000 

1*5,000 

0.462 

0.547 




5.2 

520 




Eli 

38,000 

1*5,000 

0.462 

0.547 




5.4 

450 




E7 

38,000 

1*5,000 

0.462 

0.547 




5.5 

470 




FI 

38,000 

60,000 

0.462 

0.729 




5.0 

800 




F3 

38,000 

60,000 

0.462 

0.729 




5.8 

540 




9 


TABLE 2 (Cont’d) 


SUMMARY OF HYDRAULIC COMPUTATIONS 


Run Num'ber 
Series Test 

DischarRe in C.F.S. 

Prototype Model 

Missouri R- Kansas R^ Missouri R. Kansas R. 

Average 
Flow Area 
(ft2) 

Mean- 

Depth 

(ft) 

Average 

Velocity 

(ft/sec) 

Slope 

xlO^ 

(ft/ft) 

Transported 
Sediment Cone, 
(ppm) 

Mannings 

"n" 

Froude 

Humber 



U) 

—nn 

C5l 


a) 

“TFT 

19) 

110) 

(li) 

(1^ 

“(isr 

2T 

FT 

38,000 

60,000 

0.462 

0.729 




5.3 

510 




G1 

38,000 

75,000 

0.462 

0.912 




4.6 

470 




G2 

38,000 

75,000 

0.462 

0.912 




4.6 

430 




G3 

38,000 

75,000 

0,462 

0.912 




4.5 

640 




G5 

38,000 

75,000 

0.462 

0.912 




4.3 

540 




g6 

38,000 

75,000 

0.462 

0.912 




4.5 

410 



28 

A 

38,000 

2,000 

0.462 

0.024 

1.089 

0.192 

0.445 

8.1 

620 

0.032 

0.179 


B 

38,000 

2,000 

0.462 

0.024 




8.1 

600 




C 

38,000 

2,000 

0.462 

0.024 

1.157 

0.204 

0.419 

8.3 

640 

0.035 

0.164 


D 

38,000 

2,000 

0.462 

0.024 




7.9 

380 




E 

38,000 

2,000 

0.462 

0.024 




7.4 

360 




F 

38,000 

2,000 

0.462 

0.024 




8.9 

650 




G 

38,000 

2,000 

0.462 

0.024 




7.8 

60 



29 

A 

38,000 

2,000 

0.462 

0.024 




8.6 

500 




B 

38,000 

2,000 

0.462 

0.024 

1.090 

0.184 

0.445 

8.8 

540 

0.032 

0.182 


C 

38,000 

2,000 

0.462 

0.024 




7.2 

420 




D 

38,000 

2,000 

0.462 

0.024 

1.063 

0.185 

0,456 

7.0 

410 

0.028 

0.187 


E 

38,000 

2,000 

0.462 

0.024 

1.120 

0.201 

<0.433 

8.5 

320 

0.034 

0.170 


F 

38,000 

2,000 

0.462 

0.024 




8.2 

430 




G 

38,000 

2,000 

0.462 

0.024 




5.9 

450 




H 

36,000 

2,000 

0.462 

0.024 




8.0 

490 




I 

38,000 

2,000 

0.462 

0.024 

1.192 

0.207 

0.407 

7.8 

560 

0.036 

0.158 


J 

38,000 

2,000 

0.462 

0.024 

1.167 

0.211 

0.416 

7.9 

480 

0.035 

0.160 


K 

38,000 

2,000 

0.462 

0.024 

1.178 

0.203 

0.412 

7.8 

310 

0.035 

0.161 


L 

38,000 

2,000 

0.462 

0.024 

1.177 

0.206 

0.412 

7.6 

240 

0.035 

0.160 


M 

38,000 

2,000 

0.462 

0.024 




8.1 

250 




R 

38,000 

2,000 

0.462 

0.024 




8.1 

410 




0 

38,000 

2,000 

0.462 

0.024 




8.1 

450 




P 

38,000 

2,000 

0.462 

0.024 




7.7 

370 




Q 

38,000 

2,000 

0.462 

0.024 




7.8 

460 




R 

38,000 

2,000 

0.462 

0.024 




8.5 

350 




S 

38,000 

2,000 

0.462 

0.024 




7.8 

330 




T 

38,000 

2,000 

0.462 

0.024 




8.8 

460 




U 

38,000 

2,000 

0.462 

0,024 

1,162 

0.204 

0.4l8 

7.7 

450 

0,034 

0.163 


V 

38,000 

2,000 

0.462 

0.024 




8.6 

450 



30 

A 

38,000 

2,000 

0.462 

0.024 




7.6 

460 




B 

38,000 

2,000 

0.462 

0.024 

1.136 

0.194 

0.427 

8.0 

420 

0.033 

0.171 


C 

38,000 

2,000 

0,462 

0.024 




8.2 

500 




D 

38,000 

2,000 

0.462 

0.024 




8.1 

520 




E 

38,000 

2,000 

0.462 

0.024 




8.5 

430 




F 

38,000 

2,000 

0.462 

0.024 




8.2 

400 




G 

38,000 

2,000 

0.462 

0.024 

1.097 

0.191 

0.442 

7.5 

230 

0.030 

0.178 


H 

38,000 

2,000 

0.462 

0.024 




6.8 

240 




I 

38,000 

2,000 

0.462 

0,024 




8.0 

360 




J 

38,000 

2,000 

0.462 

0.024 

1.070 

0.186 

0.453 

7.9 

320 

0.030 

0.185 


K 

38,000 

2,000 

0.462 

0.024 




7.7 

460 




L 

38,000 

2,000 

0.462 

0.024 




8.2 

420 




M 

38,000 

2,000 

0.462 

0.024 




8.4 

390 



31 

A 

38,000 

2,000 

0.462 

0.024 

1.127 

0.190 

0.431 

8.7 

560 

0.034 

0.174 


B 

38,000 

2,000 

0.462 

0.024 




8.4 

350 




C 

38,000 

2,000 

0.462 

0.024 

1.124 

0.190 

0.432 

8.7 

380 

0.033 

0.175 


D 

38,000 

2,000 

0.462 

0.024 




8.2 

540 




E 

38,000 

2,000 

0.462 

0.024 




8.0 

540 




F 

38,000 

2,000 

0.462 

0.024 




8.4 

370 




G 

38,000 

2,000 

0.462 

0,024 




8.6 

450 




H 

38,000 

2,000 

0.462 

0.024 




8.5 

330 



32 

A 

38,000 

2,000 

0.462 

0.024 

1.089 

0.193 

0.446 

1.5 

330 

0.030 

0.179 


B 

36,000 

2,000 

0.462 

0.024 




8.3 

490 




C 

38,000 

2,000 

0.462 

0.024 




8.6 

380 




D 

U0,000 

0 

0.486 

0.000 




8.1 

380 




E 

40,000 

0 

0.486 

0.000 




8.6 

420 




F 

40,000 

0 

0.486 

0.000 




8.8 




33 

A1 

73,000 

2,000 

0.983 

0.024 




7.2 

64o 




A2 

38,000 

2,000 

0.462 

0.024 




7.3 

380 



31* 

A 

38,000 

2,000 

0.462 

0.024 

1.100 

0,202 

0.442 

7.7 

270 

0.032 

0.173 


B 

38,000 

2,000 

0.462 

0.024 




6.9 

350 




C 

38,000 

2,000 

0.462 

0.024 




7.4 

250 




D 

38,000 

2,000 

0.462 

0.024 




8.0 

330 




E 

38.000 

2,000 

0.462 

0,024 

1.069 

0.192 

0.455 

8.2 

270 

0.031 

0.183 


F 

38,000 

2,000 

0.462 

0.024 




7.9 

310 




G 

38,000 

2,000 

0.462 

0.024 




8.4 

390 




10 


Figure 5, Photograph of the Kansas-Missouri confluence after the 

1951 flood. Note large amount of deposition immediately 
downstream of junction in Missouri River channel. 

River flows is not generally detrimental to the navigation channel in 
the Keinsas River Bend, it is imperative that the scour be controlled 
because of the adverse effects of the corresponding sediment deposition 

in the river downstream. Initial attempts in the model to control this ero¬ 
sion involved the addition of dike and sill 391.3 as shown on Plate 3. 

The purpose of this structure was to cause the left bank flows devel¬ 
oped during the higher Kansas River discharges to return to the right 
bank. Initial tests revealed that one dike and sill was not sufficient 
to control the left bank erosion. For this reason, another dike and 
sill were added. This structure (dike number 391*5) was placed in the 
model immediately below the confluence. The location and orientation 
of the dike and sill was altered as shown in Figure 6 in order to 
determine the most effective location. In addition to selecting the 
optimum location for the dikes, tests were conducted using various 
types of sills extending from the ends of the dikes. These included 
sills projecting directly into the flow, sills angled upstream and 
downstream, sills with "L" extensions, and curved sills. The greatest 
efforts centered on determining the best location and type of sill 
immediately below the confluence. In order to measure the effective¬ 
ness of the structure, tests were made using various combinations of 
flow from the Missouri and Kansas Rivers. The flows used in these 
tests are presented in Table 2, Series 22 and 27* 





Figure 6. Test Locations for Dike and Sill 391.5. 

The left bank dikes and sills discussed above were effective in 
controlling the erosion of the left bank and the split channel but 
were ineffective in improving the deep, narrow channel. The tend¬ 
ency for the point bar to develop directly across from the mouth of 
the Kansas River still existed. Attempts to widen the channe l 
reduce the thalweg depths were accomplished by constructing low ele¬ 
vation sills (representing depths greater than 12 feet) on the con¬ 
cave bank of the model. Structure layouts and resulting bed maps for 
two of the tests are shown on Plate 1*. These structures were tested 
to determine if they would be an effective method of reducing the 
highly concentrated velocities in the region and thus widen the 
navigation channel. Additional right bank roughness would assist the 
velocities in becoming more uniform over a greater width. Only three 
tests were conducted using these low elevation sills. In addition to 
these low elevation sills, a left bank longitudinal dike that extended 
from dike 392.1 downstream to beyond dike 391-5 was included in order 
to limit left bank scour. This structure began at the nose of dike 
392.1 and was constructed parallel to the right bank revetment. 


12 


The flow entering the Kansas River Bend is abruptly forced through 
this extremely short radius bend and a very deep and xiarrow light bonk 
channel results. A series of tests concentzated on a reedignment of 
the right bank revetment by sli^tly increasing the radius of curvature. 
Test 32c pres^ted on Plate 5 shows the minor realignment in the lower 
portion of the curve. Further attempts to widen the deep narrow channel 
were accomplished by constructing a very slight realignment of the bend 
in the vicinity of the confluence. These results (test 3^a) are also 
presented on Plate 

The realignment of the Kansas River Bend was limited by the presence 
of existing facilities and the exbranely hi^ costs associated with a 
ma;}or realignment. Studies wez^ conducted using a minor extension of 
the Missouri River ri^t bank revetment at the confluence. This structure 
is also shoTO on Plate 5 and was used to gain a more imiform flow dis¬ 
tribution throzigh the upper portion of the Kansas River Bend. The 
results of these tests were compared with previous realignment tests. 

The model test results Indicated that the dimensions of the channel 
through the Kansas River Bend was significantly influenced by the dis¬ 
tribution of flow entering from the upstream 3feadk». Consequently, tests 
were conducted using many structure arrangements in this upstream reach 
in order to develop a system that would stabilize this reach. After 
the upstream reaches had been stabilized in the model, changes in the 
Kansas River B^d channel geometry could be attributed to co3n:«sponding 
changes in the test structures in the Kansas River Bend. The extent of 
model construction works necessary to eliminate the meandering in this 
upstream reach was not used in the design prototype structures because 
the model results were influenced by model entrance conditions. They 
did, however, assist in the evaluations of the effectiveness of model 
test structures in the downstream reaches. 

ANALYSIS OF MODEL RESUITS FOR 
KANSAS RIVER BEND TESTS 


The model demonstrated that the split channel at the confluence 
can be controlled by constructing dike and sill 391*95» This is shown 
on Plate 3 by observing the bed maps of the Kansas River Bend for tests 
both with and without this dike and sill. 

Effective control for maintaining the left bank accretion in the 
Kansas River Bend was much more diffic\ilt. The model results indicated 
that the construction of dike and sill 391*3 vas not in itself effec¬ 
tive in maintaining control during hi^ Kansas River flows. Dike and 
sill 391*5 demonstrated its ability to maintain a desirable chann el. 
Tests conducted in which the location of this dike and sill was altered 
revealed that the structure (see Figure 6) was quite ineffective when 


13 


in position 2, because it intercepted too small a portion of the 
flow. Position 5 was found to be more effective. 

The differences between positions 1 and It were insignificant. 

Model results revealed that greater amounts of scour occurred at the 
end of the sill when placed in these positions than when oriented 
either 30° upstream or downstream. 

Some interesting observations were made when comparing the posi¬ 
tion 3 and 5 test results. During this study, as was noted during 
previous model studies in which submerged sills were tested, when 
flow passes over a sill it tends to pass over the sill perpendicular 
to it. This resulted in greater amounts of local scour below the 
sill when in position 3 than when in position 5. The difference, 
however, was quite local and only minor changes in the bed formation 
below the structure were noted. A slightly larger left bank scour 
hole existed below the number 3 position, although in nearly all cases 
it maintained a more suitable right bank channel for navigation. The 
differences in local scour between tests were too small to be depic¬ 
ted from the bed maps presented in Plates 2 and 3. 

The results of the model tests in which attempts were made to 
widen the navigation channel through the use of low elevation right 
bank sills are presented on Plate U. The tests demonstrated that the 
additional channel roughness caused by these sills resulted in greater 
turbulence in the vicinity of the structures, and severe scour adja¬ 
cent to the structures. In addition, these structures encouraged the 
development of a center bar which resulted in a left bank channel 
similar to the split channel that dike 391.9 was designed to elimin¬ 
ate. The turbulence which developed near these structures would be 
detrimental to navigation traffic, A left bank longitudinal dike 
extending from dike 392.1 downstream to beyond dike 391.5 was used in 
conjunction with these low elevation sills. This structure was, in 
effect, a left bank revetment. Although adequate depths were main¬ 
tained throughout the channel width using this dike, velocities were 
excessive and turbulence too great to permit safe passage of tows. 

The results of the very minor realignment of the Kansas River Bend 
are shown on the bed maps on Plate 5- Test 32c demonstrated the effects 
of the very minor relocation of the lower half of the curve. A very 
minor realignment of the upper portion of the curve was also tried in 
subsequent tests. These changes resulted in virtually no improvements 
because the increase in the bend's radius of ciirvature was insignifi¬ 
cant. A revetment extension of the confluence structure was found to 
be equally effective in streamlining the flow around the upper portion 
of the curve. This is demonstrated by the results of test 3itE, which 
are presented on Plate 5. 


The changes in bed forms and scour hole dimensions for the many 
types of sills and dike extensions were also documented. Similar 
tests of these sills were made in the straight Kansas City Reach. 

Test results for sills in both reaches were similar and are discussed 
under "Analysis of Model Results for Kansas City Reach Tests." 

DESCRIPTION OF KANSAS CITY REACH MODEL TESTS 


The model investigation of the Kansas City Reach involved the 
meandering characteristics of the channel throughout the long, straight 
reach. Prior to the initiation of the model study, the river channel 
was confined to an unobstructed width of about 800 feet, and the chan¬ 
nel meandered within these limits. The objective of the investigation 
was to determine the degree of confinement and extent of construction 
necessary to maintain an adequate navigation channel throughout the 
reach. The requirement to maintain the navigation channel through the 
right bank swing span of the AS&B Bridge necessitated that the naviga¬ 
tion channel be confined to the right bank throughout the entire reach. 
After the initial verification tests, changes were made in the model 
structures in an attempt to improve the navigation portion of the model 
channel. Plate 1 shows the structiores present in the prototype at the 
time of the model study. The bed maps of Plate 2 indicate the location 
of the channel and major shoaling areas existing in the prototype at the 
time of the model study. This plate also shows the model reach at the 
completion of verification test number l8 and illustrates the model's 
ability to reproduce itself (model test 21a). 

Initial attempts to produce the desired navigation channel centered 
on development of a series of dikes and sills that would contract the 
channel until the desired channel dimensions were achieved. This was 
accomplished by investigating many types of structures including: low 
elevation dikes called sills (both level and sloping crests) placed 
perpendicular to the flow; curved sills; and longitudinal structures 
called vane dikes. A description of each of the types of structures 
tested and the test results follow. 

Because the flow is heavily concentrated on the right bank through¬ 
out the bend immediately upstream of the reach (the Kansas River Bend), 
strong secondary currents are developed. These currents cause the 
highly concentrated flows to leave the right bank and cross to the left 
bank in the upstream portion of the Kansas City Reach. Preventing the 
crossing was of utmost importance in developing an 
adequate navigation channel. Initial attempts to limit right bank 
shoaling in the model involved the combination of a left bank dike and 
a low elevation sill extending to within 500 prototype feet of the right 
bank at river mile 365.9* The dike crest, like other dikes in the reach, 
was constructed above the normal water surface elevation, while the sill 


15 


was built to CRP which is about two or three feet below the normal 
water surface elevation. As testing progressed, additional sills were 
added to alternate existing dikes for each subsequent test, until 
sills extended from the end of every second dike throughout the entire 
Isngth of the Kansas City Reach. For this series of tests, the chan¬ 
nel was confined to a prototype width of 500 feet. Observations made 
during these tests revealed that after each sill was placed in the 
model, the right bank shoal would advance farther downstream into the 
reach. An exception to this was noted near the AS 8 eB Bridge. A right 
b^k shoal usually developed in the model when no sill was placed at 
dike 389 . 97 , and an adequate navigation channel existed throughout the 
remainder of the straight reach. This condition was corrected by add¬ 
ing a sill to this dike. The degree of channel control and the result¬ 
ing channel dimensions for this series of tests are shown on Plate 6 . 

After the determination was made of the number of sills necessary " 
to control the reach, future efforts centered on the degree of confine¬ 
ment necessary to maintain an adequate navigation channel. In order 
to do this, sills were again extended from the same dikes, but all were 
shortened so that the remaining channel represented a prototype width 
of 650 feet. The sills were then altered so that the channel was con¬ 
fined to 575 feet. The results of these tests were then compared to 
the original tests in which the confinement had been 500 feet. Com¬ 
parisons of these results are shown on Plate 6 and represent conditions 
for the indicated number of sills. A series of tests were run for var¬ 
ious numbers of added sills. The number of structures was varied for 
each degree of channel confinement using the procedure discussed above 
for the 500 -foot channel. The resulting bed maps for these tests are 
not included in this report. 

Tests were also conducted in which the crest elevations of the 
structures were varied. The variation in crest height ranged from CRP 
to above the normal water surface elevation. 

In still another series of tests, sloping sills were used in which 
the crest slopes of the test structures were varied. A total of nine 
model runs were completed in this series, using three different end 
elevations with each of three crest elevations at the root of the sills. 
The channel confinement remained constant at 500 prototype feet for 
this entire series. All structures were placed perpendiculeir to the 
flow. 

The sills projecting directly into the flow caused substantial 
scour near the channel end of the test structures. For each basic 
structure type tested, greater degrees of channel confinement generally 
resulted in greater scour. Curved sills were tested in an attempt to 
decrease the turbulence near the ends of the structures. These curved 



sills extended from the end of the dikes perpendicular to the direc¬ 
tion of the flow and then curved downstream through a 90^ bend to 
parallel the flow at the channel end. All structures used for these 
tests were constructed to CRP and confined the channel to 500 proto- 
type feet. 

The final series of tests in the investigation utilized longitu¬ 
dinal structures called ^*Vane Dikes.” These test structures were 
constructed to an elevation above the normal water surface, and were 
placed either parallel or slightly angled at 5^ or 10^ to the direc¬ 
tion of flow. The dikes varied in length as did the spacing between 
structures. Plate 7 demonstrates the structure layout for three of 
the model tests. Structures were initially placed at the upstream 
limits of the reach and subsequent structures added in a downstream 
direction. 

ANALYSIS OF MODEL RESULTS FOR KANSAS CITY REACH TESTS 


Developing and maintaining an adequate navigation channel in the 
Kansas City Reach of the Missouri River can be accomplished by con¬ 
fining the channel to a width less than 800 feet. This can be achieved 
through the use of one of several types of control structures. Low 
elevation sills placed perpendicular to the flow were found to be the 
most effective method tested, and required the least amount of addi¬ 
tional construction. These sills have a minimal effect on the flood 
control capacity of the channel, and also control meandering through¬ 
out the reach at normal navigation flows. 

Model tests revealed that an additional dike and sill should be 
constructed at river mile 365*9 as shown on Plate 8. This structure 
would be instrumental in preventing the flow from crossing from the 
right to the left bank and, therefore, would limit the highly concen¬ 
trated flow against dike 390.15. Tests were conducted in which two 
additional dikes and sills were placed between the AS&B Bridge (mile 
365 . 5 ), and the Hannibal Bridge (mile 366.1). No measurable channel 
improvements were noted when the results of these tests were compared 
with the results of tests in which only one structure, shown on Plate 
6, was used. Therefore, all subsequent studies utilized only the one 
test structure in this river segment. 

No significant improvements were noted in the model channel when 
the crest elevations were raised from CRP to above the normal water 
sxarface elevation. Slightly larger volumes of accretion developed 
behind the higher sills; however, the lower elevation sills allowed 
only a negligible amount of flow in this portion of the channel, with 
little net influence on the resulting navigation channel. Reduced 
amounts of scour near the ends of the sills was noted when the sills 
were constructed to the lower crest elevations. 


The amount of channel constriction was found to have significant 
influence on the ultimate shape of the navigation portion of the 
channel. This was clearly emphasized during the phase of the inves¬ 
tigation in which sloping sills were tested. The channel width 
between the ends of the structures remained constant for this series 
of tests, but the amoxxnt of channel constriction varied according to 
the elevation at the root and the slope of the crest of the test sills. 
The results of these tests were similar to those in which level sills 
were tested. For comparable channel constrictions, the percent of the 
total flow behind both types of sills was comparable. Only slight dif¬ 
ferences were noted in their ultimate effect on the channel. During 
the study of level sills, a small secondary channel or chute often 
developed between and in line with the ends of the dikes, while higher 
deposits were noted toward the channel. This apparently was due to the 
large amount of accretion that developed between the sills from the 
channel side due to movement of material along the bed. The accretion 
in this region would continue to increase, severely limiting the influx 
of transported sediment necessary to deplete the small remaining chan¬ 
nel. It was observed during the testing of sloping sills that accre¬ 
tion would often develop from the bank side and progress to the chan¬ 
nel end of the test structures; thus, eliminating this small left bank 
chute. Slight differences in the velocity distribution were apparently 
responsible for this change. This small channel was not particularly 
objectionable; therefore, there appeared to be no significant advan¬ 
tage in using the sloping sill. 

The degree of channel confinement (distance between the end of the 
left bank dike or sill, and the right bank revetments) was a prime con¬ 
sideration of this model study, as was the number of structures 
required to control channel meandering. The model study revealed that 
sills constructed from the end of a proposed new model dike at mile 
365.9 and from the ends of alternate dikes presently existing through¬ 
out the reach would adequately control channel meander and shoaling. 
Generally, additional sills did not result in further improvements in 
the navigation channel. One major exception to this is at Dike No. 
389.97 just below the AS&B Bridge. A sill was required at this struc- 
tiire to prevent a major right bank shoal from developing immediately 
downstream of this location. The recommended structure layout for the 
entire reach is shown on Plate 8. The selected structures do not neces¬ 
sarily represent the optimiom development of the reach, but represent a 
system of structures which will insure a suitable navigation channel 
completely incorporating the existing channel control structures. The 
economic necessity of utilizing existing structures limited the posi¬ 
tion and number of possible sills. 

The degree of channel confinement significantly influenced the 
dimensions of the channel along the right bank throughout the length of 
the Kansas City Reach. Verification tests (see Plate 2) indicate the 


presence of a large right bank shoal at mile 365.6 and a meandering 
channel downstream of this location. Tests in which the channel was 
confined to 65 O prototype feet (Plate 6) indicated only minor improve¬ 
ments over existing conditions. Fxxrther confinement to 575 prototype 
feet indicates a greatly improved channel, although a slight tendency 
to meander was still evident. Further confinement to 500 prototype 
feet resulted in additional improvements, although they were not con¬ 
sidered significant. Based on the results of these model tests, it is 
recommended that the channel be confined through the use of level sills 
to a width of 60 O feet. 

The use of curved sills (an example of these is shown on Plate 5) 
resulted in less turbulence and scour at the channel end than what was 
observed when the sills projected perpendicular to the flow. However, 
the ultimate improvement in the channel due to the reduction of this 
scour was not considered sufficient to Justify the additional construc¬ 
tion and maintenance costs associated with a structure of this nature. 

Plate 7 shows the channel geometry that existed at the completion 
of three of the model tests in which longitudinal struct\ires or vane 
dikes were tested. Many tests were conducted in which dike length, 
distance between dikes, and the angle between the dike and flow direc¬ 
tion were varied. Model tests revealed that the dikes should be 
slightly angled to the flow to prevent a large portion of the flow from 
passing along the backside of the structures. No reduction of flow was 
noted, however, when the angle between the dike and flow direction was 
increased from 5 to 10 degrees. 

The investigation revealed that there is an optimum length for the 
longitudinal structures. It was noted that model structures two feet 
long were much less effective than those three feet long. To achieve 
the same degree of control using two-foot dikes rather than three-foot 
structures, it was necessary to reduce the spacing between structures 
by more than one-third. This resulted in a greater total length of. 
structures required to achieve the same degree of control. The model 
results also indicated that when dike lengths were increased beyond 
three feet, the spacing between dikes could not be comparably increased. 

Although the tests showed that an adequate channel could be devel¬ 
oped in the reach through the use of vane dikes, this means of control 
was not recommended for this reach because of the high costs involved. 

This type of construction appears to be more adequately suited to 
river reaches where previously constructed control structures do not exist. 
Further model tests of an imcontrolled river segment are necessary to 
adequately define general design criteria for this type of stiucture. 


CONCLUSIONS 


Thfi model study of the Kenses River Bend and Kansas City Reach 
indicated the following: 

a. The Kansas River Bend split channel at the confluence can be 
controlled through the use of dike and sill 391. 9 , 

b. Removal of portions of the left bank accretion in the Kansas 
River Bend can be controlled by construction of dikes and sills 391.3 
and 391 . 5 . The most effective structure location for dike and sill 
391.5 is shown on Plate 8. 

c. Additional right bank roughness in the bend through use of low 
elevation sills is not an effective method for developing a wider, 
more uniform navigation channel aro\ind the bend. Tests using these 
structures resulted in severe turbulence and scour near the test struc¬ 
tures and in some cases, caused the development of a left bank second¬ 
ary channel. 


d. Minor realignments of both the upper and lower segments of the 
Kansas River Bend resulted in no significant improvement in the dimen¬ 
sions of the navigation channel. An equally effective improvement was 
demonstrated by a revetment extension of existing revetment 378.91 at 
the confluence. This extension is shown by run 34 e on Plate 5 and 
again on Plate 8. 

e. Tests of the Kansas City Reach indicated that dike and sill 
390.35 would limit the flows crossing in the upper segment of the 
reach. Further tests revealed that sills constructed on alternate 
existing dikes throughout the reach would limit channel meander within 
the reach except that an additional sill was necessary below the AS&B 
Bridge. The channel confinement must be reduced from the existing 
800 feet to 600 feet in order for this system to be effective. Level 
sills constructed to the Construction Reference Plane are as effec¬ 
tive as those built to above-normal water surface. 

f. Sloping sills demonstrated comparable improvements in navigation 
channel dimensions when compared with level sills where similar channel 
constrictions were used. The very sU^t difference in the accretion 
development was considered to be insignificant. 

g. Vane dikes have a definite place in river development, partic- 
in the original stabilization attempts. Because of the exist¬ 
ing construction in the modeled reach, this type of control was not 
feasible for the Kansas City Reach. Use of these structures warrants 
additional study. 

h. All of the structures that were demonstrated by the model to 
be the most effective and feasible are shown on Plate 8. 













Missoum tIVtR DCSICN STUDY 

KANSAS CITY NAVIGATION MODEL STUDY 

SKETCH OF MODELED REACH 

U. S. AIMY CNGINEER DISTRICT, OMAHA 
CORPS OF ENGINEERS 
OMAHA, NEIRASKA 


OECEMKEK 1971 



LEGEND 


aiSTING DIKE 
lEVEE 

ElOODWAll 
BIKE NUMIEK 
nVEI MUE 



DIKE (MODEL) 
SlU (MODEL) 



S92.S 
365 O 


NAVIGATION CHANNEL DEfTHS 
ADEQUATE m^^m. 
INADEQUATE KHtlllllll l li 







PLATE 2 











3 ^ 


I 


WITH DIKES & CURVED SILLS 


LEGEND 

niSTING DIKE 


»»[ (MOMl) •MiiM 

LEVEE 

A _ A _ A, 

V-f— ¥ 

SlU (RODIl) • • ■ 

FIOODWALL 


NAVIGATION CHANNEL DEPTHS 

DIKE NUMia 

3f2.3 

ADEQUATE NRRSBBSSSM 

RIVER MILE 

3A5 O 

MADIQUATI ttHlItlttHa 


MODU TEST 30K 







WITH DIKES & r TYPE SIllS 


MODEL TEST 30J 


MISSOUK Rivet OISION STUDY 
KANSAS CITY NAVIGATION MODEL STUDY 

KANSAS RIVER BEND BED MAPS 
LEFT BANK DIKE AND SILL TESTS 

u i AIMT (NOiNtei DISTIICT, Omaha 
CO tP) or fNOINIfIS 
OMAHA. NIIIASKA 


OeC^MtCR 1971 







LEGEND 


EXISTING DIKE 
LEVEE 

FLOODWALL 
DIKE NUMBER 
RIVER MILE 



DIKE (MODEL) 
SILL (MODEL) 


392.3 
365 O 


NAVIGATION CHANNEL DEPTHS 
ADEQUATE f ^ 
INADEQUATE iliU.tI|.niU.l.| 


NOTE: The concave bank sills shown on this plate were 
constructed below navigation depths. 











LEGEND 


EXISTING DIKE 
LEVEE 

FLOODWALL 
DIKE NUMBER 
RIVER MILE 



DIKE (MODEL) 
SILL (MODEL) 


392.3 
365 O 


NAVIGATION CHANNEL DEPTHS 
ADEQUATE ^ ; 

INADEQUATE imUmHHil 


MISSOURI RIVER DESIGN STUDY 

KANSAS CITY NAVIGATION MODEL STUDY 

KANSAS RIVER BEND BED MAPS 
MINOR REALIGNMENT TESTS 

U. S. ARMY ENGtNEER DISTRICT, OMAHA 
CORPS OF ENGINEERS 
OMAHA, NEBRASKA 

DECEMBER 1971 


PLATE 5 












































LEGEND 


EXISTING DIKE 
LEVEE 

ELOODWALl 
DIKE NUMBER 
RIVER MILE 



DIKE (MODEL) 
SILL (MODEL) 



392.3 
365 G 


NAVIGATION CHANNEL DEPTHS 
ADEQUATE 

INADEQUATE fllSanE:; : : 


MISSOUII tIVER DESIGN STUDY 

KANSAS CITY NAVIGATION MODEL STUDY 

KANSAS CITY REACH BED MAPS 
VARYING DEGREES OF CHANNEL CONFINEMENT 

U S AIMT iNGiNfEI DiSTtiCI. Omaha 
CO tPA or tNClNCEIS 
OMAHA. NEIIASRA 


DECEMIEt 1971 















KANSAS CITY REACH CHANNEL CONTROL STRUCTURES • 1968 





















PLATE 7