HomeMy WebLinkAboutAPA78TK
1425
.S8
A23
no.78
' I I .
I
"
----. _ AlASKA RESOURCES LIBRARY ~e!JlCd •'>'nh ... ARLlS .S . Department of the Intenor
--------~~~~~7-------~
ERim I
Alaska Resources
Library & Information Services
Ancl. ·e, .• \....iSka
f;USIT f~A HYDRCELECTP.aC PROJECT
TASK 6 -DESIGN DEVELOPMENT
SUBTASK 6.14
SCO~!P HOLE DEVELO~MENT
DOV.'NSTR~AM OF H ~GJ-f ·-H EAD DAMS
MARt:H "i 9 ·'32
I
l
I
I
I
I I I j
I I
I I
! I
I
I
i
I
I
I
.___ ___ A.LAS~Z~~ POWER AUTHORIT·y=-~-~=~
I
ALASKA POWER AUTHORITY
SUSITNA HYDROELECTRIC PROJECT
TASK 6 -DESIGN DEVELOPMENT
SUBTASK 6.14
SCOUR HOLE DEVELOPMENT
DOWNSTREAM OF HIGH-HEAD DAMS
TABLE OF CONTENTS
LIST OF TABLES
LIST OF FIGURES
1 -INTRODUCTION 1-1
2 -sUMMARy . . . . . . . . . . . . . . . . • . . . I • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • 2 -1
2.1 -Problem Definition ...............•........•................. 2-1
2.2-Method of Analysis .......................................... 2-1
2.3 -Results and Discussion .. .. . . .. .. . .. . . . .. . .. . .. .. . . .. .. .. .. . . 2-1
2.4-Conclusions and Recommendations ............................. 2-1
3 -PROBLEM DEFINITION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-1
3.1-Description ...............................................•. 3-1
3.2-Controlling Factors ......................................... 3-1
3.3-Existing Design Methods ......•...........•.................. 3-3
4 -METHOD OF ANALYSIS ................................................ 4-1
5 -RESULTS AND DISCUSSION . . . . . . . .. .. . . . .. . . .. .. . .. . . .. .. .. . . . . . . .. . . . 5-1
5 .1 -Res u 1 t s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-1
5.2 -Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-2
6-CONCLUSIONS AND RECOMMENDATIONS ................................... 6-1
6.1-Conclusions and Recommendations ............................. 6-1
6.2-Application to the Susitna Project .......................... 6-1
BIBLIOGRAPHY
TK
/42.5
. s~
A23
1110,78
LIST OF TABLES
Number
3.1
5.1
Title Page
Simple Scour Prediction Formulas ..................... ... 3-5
Prototype Scour Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-4
LIST OF FIGURES
Number
3.1
3.2
3.3
5.1
5.2
Title Page
Flip Bucket Definition Sketch . .......................... 3-6
Simple Scour Equations -50 Foot Head . . . . . . . . . . . . . . . . . . . 3-7
Simple Scour Equations-400 Foot Head.................. 3-8
Estimated Scour Depth -Heads Greater Than or Equal
to 100 Feet . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-6
Estimated Scour Depth -Heads Less Than 100 Feet 5-7
5.3 Prediction and Confidence Limits -Heads Greater Than
or Equal to 100 Feet . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-8
5.4 Prediction and Confidence Limits -Heads Less Than
100 Feet . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-9
6.1 Scour At Watana -725 Foot Head . . . . . . . . . . . . . . . . . . . . . . . . . 6-2
6. 2 Scour At Dev i 1 Canyon -565 Foot Head . . . . . . . . . . . . . . . . . . . 6-3
1 -INTRODUCTION
The purpose of this study is to investigate the development of scour holes in
unlined plunge pools downstream of high-head dams and to formulate a relation-
ship for determination of scour hole depth to assist in the feasibility level
spillway layouts of the Susitna Hydroelectric Project.
Numerous formulas are presently available to predict such depths. However,
these existing scour prediction relationships have been developed from analysis
of relatively low head dams and most of these are based on analysis of model
studies. The applicability of the available prediction methods is consequently
in doubt when dealing with high-head dams.
This report presents a parametric study of prototype scour data and incorporates
some observations at high-head dams. The derived statistical relationships
estimate the scour depth, given the unit discharge and head. It must be noted
that a number of variables which affect the extent of scour hole development are
not incorporated in this relationship and these must be considered for each
particular site. However, the formulas for estimating expected scour depth
proposed herein provide an acceptable basis for preliminary layout designs when
combined with appropriate engineering judgment.
1-1
2 -SUMMARY
2.1-Problem Definition
The extent of scour hole development in unlined plunge pools downstream of free
overfalls and spillway flip buckets is an extremely important design parameter
for high head dams. Underestimation of a scour hole may affect the structural
integrity of component structures of the dam and ultimately could lead to dam
failure. The extent of scour hole development is a function of numerous para-
meters which may be grouped into four categories: geotechnical, geometric,
hydraulic and duration of flows. No comprehensive scour prediction relationship
is currently available. Past efforts used to understand scour have focused on
physical model studies. This study is an attempt to develop an empirical scour
depth prediction formula for high-head dams based on prototype data.
2.2 -Method of Analysis
The general approach by Martins (18) and Chian Min Wu (8) was followed to
develop a statistically significant relationship between scour depth, unit
discharge and head. Other parameters affecting scour must be taken into account
through the application of engineering judgment. A total of 36 prototype data
sets have been assembled and grouped into 18 high-head (greater than or equal to
100 feet) and 18 low head (smaller than 100 feet) sets. Regression analyses
were run on both groups to develop scour depth estimation equations, confidence
intervals and prediction intervals.
2.3 -Results and Discussion
The resulting scour depth equations are presented in Section 5.1. The equations
were developed on statistical grounds and were assumed to be dependent on unit
discharge and head only. The actual scour depth for a particular prototype con-
dition may vary greatly from the regression estimate. Such variations, as shown
by the prediction limits, may be caused by:
-the extent of aeration of the discharge jet;
-the geometry of spillway chute and flip bucket;
-the downstream water levels;
-the distribution, variation and duration of the flow; and
-the geological conditions at the area of impact.
2.4 -Conclusions and Recommendations
This study resulted in relationships which, when applied with sound engineering
judgment, may be used to predict scour depth in plunge pools below overfall or
flip bucket spillways with sufficient accuracy for preliminary design. Due to
the nature of the study undertaken and the limited prototype data, it is
suggested that the preliminary design be based on the estimated scour depth
curves and that more detailed physical modeling and theoretical studies be done
for the final design. The preliminary design recommendation is that all loose
material in the plunge pool area should be removed to good rock. Deeper
excavations in good rock should be made as further studies in the final design
stage indicate.
2-1
3 -PROBLEM DEFINITION
3.1 -Description
The importance of estimations of the extent of scour hole development, particu-
larly with regard to maximum depth of the hole, has become apparent with the
construction of very high dams during the last 20 years. Excessive scour hole
development has affected the structural integrity of component structures of
many dams and ultimately could have led to dam failure. Extensive scour can
affect the net head available for power production when material removed from
the hole is deposited downstream, usually in the form of a bar.
For purposes of this report, scour hole development is considered in unlined
plunge pools downstream of free overfalls and spillway flip buckets. Figure 3.1
shows a definition of controlling parameters.
The extent of scour hole development is, amongst other factors, a function of
the duration of the scouring process. During the initial phase the majority of
energy in the incoming jet produces a dynamic loading on the riverbed. This
relatively strong dynanic loading 1 ifts rocks from the bed and may break off
chunks of rocks which are removed from the impact area. As the hole enlarges
and deepens, more of the energy of the jet is dissipated in turbulence before
striking the bottom. The dynamic loading is reduced and smaller rocks are
broken off and transported downstream. The intermediate and final phases are
characterized by rocks which are too large to be removed and which churn in the
bottom of the hole. The abrasive action loosens small particles which are
easily swept downstream. When the hole is large enough to dissipate all of the
jet•s energy through turbulence, rocks will not churn and the hole is in an
equilibrium condition.
The physical process that produces scour is an extremely complex phenomenon and
is dependent on various physical parameters. Head, discharge, unit discharge,
depth of water cushion, width ratio (width of incoming jet to width of river),
angle of incidence (of the incoming jet), air entrainment, duration of dis-
charge, and numerous geotechnical parameters all contribute to the depth, shape,
and location of the scour hole. No comprehensive scour prediction relationship
is currently available. Past efforts in understanding the scour phenomenon have
centered around physical model studies, focusing on only a few variables which
were thought to be significant by the author (or the specific conditions under
study). Very few studies are avail able for high-head dams.
3.2-Controlling Factors
Parameters influencing scour development can be grouped into four categories;
geotechnical, geometric, hydraulic and flow duration. Following is a discussion
of each category and a qualitative assessment of the parameters controlling the
scour process.
3-1
(a) Geotechnical
For purposes of scour development studies, soil conditions have been class-
ified into two major types, cohesive and noncohesive. When considering
scour development, as discussed in Section 3.1, it has been assumed that
the occurrence of truly cohesive riverbed material is extremely rare.
Therefore, this study deals primarily with noncohesive riverbeds. In fact,
all scour hole related model studies have been conducted using noncohesive
bed materia 1 •
Some investigators, especially those involved in model studies, have postu-
lated that the extent of scour in riverbeds composed of uniform rigid
blocks, such as defined by the jointing and shearing characteristics of
rock, is a function of block size. This approach has been found to be
questionable due to the scouring effect of the continuous churning and
grinding action of rocks in the scour hole. It appears reasonable, how-
ever, to assume that the block size significantly affects the rate of scour
development, more so than the extent of scour development. That is, the
rate of scour decreases with increasing block size.
(b) Geometric
The geometry of the spillway chute has a major bearing on the extent of
scour hole development. Of prime importance are the width of the chute,
the angle and shape of the flip bucket and the presence of chute acces-
sories such as training walls or flow dividers.
The width of the chute and the shape of the flip bucket govern the unit
discharge on impact which is the measure of flow concentration and there-
fore affects the extent of scour hole development.
The shape of the flip bucket can be chosen in suct1 a way that the concen-
tration of flow is reduced in the area of impact. This effectively reduces
the depth of the scour hole, but also increases the areal extent of scour
hole development.
Introduction of flow dividers or training walls to the spillway chute
affects the flow leaving the flipbucket and, therefore, influences the
scour hole development. The design of such spillway chute accessories is
normally established by scale model investigations.
The angle of the flip bucket determines the throw distance to the area of
impact. To safeguard against undermining of the structure, a maximl.ITl throw
distance is typically desired. Using a simple trajectory formula, this
distance is maximized with a flip bucket angle of 45°. In practice,
however, a smaller angle, in the range of 16.5° to 37.5°, is generally
found to be more desirable (24). The flip bucket angle also governs the
angle of incidence of the incoming jet. An incoming jet angle of incidence
in the range of 40° to 70° has little influence on the scour hole depth
(17). A greater angle would most likely increase the scour hole depth
while a smaller angle would decrease the depth. However, the smaller angle
would result in reduced plunge pool energy dissipation, leading to possible
erosion downstream.
3-2
The geometric properties of the chute and flip can also be significantly
influenced by the layout of the dam and tailrace locations. The topography
of the site generally determines site layout and will dictate, in
conjunction with other considerations, the preferred location of the jet
impact zone.
(c) Hydraulic
The hydraulic properties of the spillway flow have a major influence on the
develO!lilent of scour holes.
The unit discharge and the head determine the unit energy to be dissipated,
and are considered the major independent variables affecting scour develop-
ment.
Air entrained in the jet before it leaves the flip bucket affects the
extent of energy dissipation prior to impact and therefore influences the
scour hole development. Martins (17) reported that intermediate (25 per-
cent) and high (75 percent) air entrainment reduces scour depths by 10
percent and 25 percent respectively.
Tailwater elevations effect the total head of the flow and therefore affect
the amount of energy to be dissipated. Particularly for discharge condi-
tions in narrow gorges, the tailwater depth may be high, and therefore
depth variations need to be considered. Furthermore, the depth of water in
the area of impact tends to act as a cushion for the impinging jet and
mitigates the scouring action.
(d) Duration
General information on the progression of scour with time is seldom avail-
able. Although the duration of flow strongly influences scour develo!lilent,
the equilibrium scour depth is not affected. However, the more frequent a
spillway is used, the faster the scour develo!lilent takes place. Concep-
tually, the scour process is related to duration and geotechnical para-
meters. Poor rock format ions erode relatively quickly when compared to
good rock formations. Also, flow variation with time affects the rate of
scour hole development. Short-duration flows of high intensity may cause a
different scour hole geometry than long-duration flows of much lesser
intensity. The controlling parameters are site-specific and need to be
evaluated in terms of their effect on the equilibrium scour depth.
3.3 -Existing Design Methods
The numerous factors influencing the scour development process do not easily
allow the use of an analytical design approach. The historical method of
analysis has therefore been based on scale model studies. In these studies the
riverbed material was simulated by sand, gravel and concrete b 1 ocks. However,
in rocky riverbeds, the abrasion and resulting erosion with time of discrete
blocks of rock play an important role in the development of the scour hole.
Furthermore, the size of such blocks is a complex function of on-site geological
3-3
subsurface conditions. The effects are not easily simulated and the results of
the model studies, therefore, do not yield accurate prediction of scour hole
dimensions. However, investigators have attempted to formulate general equa-
tions to estimate scour depth and shape for given specific design conditions.
Table 3.1 lists some of these equations.
The need for better design criteria to predict the extent of scour development
has resulted in studies which considered actual prototype performances. The
availability of prototype data, however, is limited and specific experience
information, such as flow variation and duration of discharge, is difficult to
obtain. Martins (18) in 1975 assembled 18 data sets mostly of low-head spill-
ways and comprising unit discharge, head and scour depth only. A statistical
analysis carried out on these data sets yielded a general design formula (Table
3.1) which was presented as a basis for determining the estimated depth of scour
holes. Martins' derivation represents a broad spectrum of case studies without
considering the effects of specific geotechnical, flow duration and geometric
parameters. It must be noted that Acres attempts have failed to rederive this
relationship using the same data base* as Martins. Figures 3.2 and 3.3 show
plots of all equations presented in Table 3.1 for constant heads of 50 and 400
feet respectively.
*No contact with Martins has been established, and the data base used to
rederive the relationship is taken directly from the literature(18).
3-4
TABLE 3.1: SIMPLE SCOUR PREDICTION FORMULAS 1
Author Equation ---Units Reference
Martins y = 1• 5 q o. 6 Ho. 1 Metric ( 18)
Veronese y =
0.54 0.225 1. 32 q H English (7)
Chee y
1.235 90.67 H0.18
= d 0.063
English (7)
Patrashev y = 3.9 q 0.5 (~)0.25 Metric (7)
d
Schoklitsch y
3.15 90.57 Ho.2
= d 0.32
Metric (7)
Damle y = 0.36 qo.5 Ho.5 English (9)
Vyzgo y = AK q0.5 H0.25 Metric (26)
Chian Min Wu y = 1•18 q0.51H0.235 Metric (8)
where: H is the head measured from reservoir level to spillway tailwater
level.
q is the unit discharge on impact.
Y is the depth of scour.
A, K are constants dependent on air entrainment and flip bucket angle
respectively.
d is block diameter
Different formulas use slightly different definitions of parameters, see
references for application techniques.
3-5
w
I
0"1
FLIP BUCKET DEFINITION SKETCH
FIGURE 3.1.
350
y
~0 /1
300 / ( /I / V; /
250 /
VI v
-.:: w / 1/ w u.. -
J:200
b: I I v w
/ 0
a:: VI v /
,..... / :::>
0 I ~ u / ~ CJ) 150
II I ~ ~ ,.....,-~ v-s-/ ./ ~
I I y , ~ .......
~ ~ ~ ~ 100 / -/;; ~ / ~ t;:::: --~9
~ ..--:::::::: :::::---~
~ ~ :.---
~ ~ ~ ~ ~ ---
50
·~
~ ~ ~
If'
00 200 400 600 900 1000 1200 1400 1600 1800 2000
UNIT DISCHARGE
(CUBIC FOOT PER SECOND PER FOOT)
EQUATION NUMBER AUTHOR REMARKS
I VERON ESE EQUILIBRIUM SCOUR
2 CHEE EQUILIBRIUM SCOUR; d= 5 FT.
3 PATRASHEV EQUILIBRIUM SCOUR i d = 1.52 FT.
4 SCHOKLITSCH EQUILIBRIUM SCOUR; d = 1.52 m
5 DAMLE EQUILIBRIUM SCOUR
6 MARTINS REGRESSION ESTIMATE
7 VYZGO EQUILIBRIUM SCOUR
8 CHIAN MIN WU REGRESSION ESTIMATE
9 REGRESSION REGRESSION ESTIMATE
10 REGRESSION UPPER PREDICTION LIMIT
SIMPLE SCOUR EQUATIONS -50 FOOT HEAD 3.2. 3-7 FIGURE
1-
UJ
UJ
lL
J:
1-
0..
UJ
0
0:
:::;)
0
(.)
(/)
550
500
450
400
350
300
250
200
150
100
50
200 400 600 800 1000 1200 1400 1600 1800 2000
EQUATION NUMBER
I
2
3
4
5
6
7
8
9
10
SIMPLE
UNIT DISCHARGE
(CUBIC FOOT PER SECOND PER FOOT)
AUTHOR
VERON ESE
CHEE
PATRASHEV
SCHOKLITSCH
DAMLE
MARTINS
VYZGO
CHIAN MIN WU
REGRESSION
REGRESSION
REMARKS
EQUILIBRIUM SCOUR
EQUILIBRIUM SCOUR; d: 5 FT.
EQUILIBRIUM SCOUR; d: 1.52 FT.
EQUILIBRIUM SCOUR; d = 1.52 m
EQUILIBRIUM SCOUR;
REGRESSION ESTIMATE
EQUILIBRIUM SCOUR
REGRESSION ESTIMATE
REGRESSION ESTIMATE
UPPER PREDICTION LIMIT
SCOUR EQUATIONS -400 FOOT HEAD
3-8 FIGURE 3Ji]
4 -METHOD OF ANALYSIS
In order to arrive at a simple relationship to estimate scour depths downstream
of high-head spillways, the general approach taken by Martins (18) and Chian Min
Wu (8) was followed in which the unit discharge and the head of the approach
flow are the independent variables. Martins' original data base of 18 sets was
combined with the 6 prototype data sets of Chian Min Wu and extended to incor-
porate 12 additional case studies. The resulting 36 sets were then grouped
arbitrarily into ranges of high head (greater than or equal to 100 ft) and low
head (smaller than 100ft) in order to more accurately develop scour hole
formulas for high-head spillways. Regression analyses were then carried out for
all available data sets and the resulting equations were plotted together with
the actual prototype observations. Confidence limits and prediction limits were
calculated and plotted for the 95 percent interval.
Upon examination of the plots it was found that Kariba lay far above of the con-
fidence interval and approached the upper limit of the prediction interval. To
increase the statistical significance for the average type high-head site, the
regression analysis was repeated omitting the Kariba data set. The revised
equations are believed to be more representative of the typical high-head plunge
pool site and operations. The equations were again plotted together with the
actual prototype observations. Kariba lies barely within the prediction inter-
val of the revised equations.
Although the development of scour is also affected by geometric, geotechnical
and other par~neters, a comprehensive study of all these factors is not
warranted at the level of conceptual layouts. It is therefore important that the
presented equations are applied with considerable engineering judgment to assess
the extent of scour hole development for each particular design.
4-1
5 -RESULTS AND DISCUSSION
5.1 -Results
The prototype data collected for development of the scour formulas described
herein is shown in Table 5.1. The data is grouped into two head ranges as
described in Section 4. For each group, a best fit equation was determined by
regression analysis. The curvilinear multiple regression was changed by log
transformation into a multiple linear regression which was then solved by the
least squares method. The estimated parameter coefficients are significant at a
minimum confidence level of 95 percent. Confidence intervals and prediction
intervals are based on 95 percent confidence levels. The confidence interval is
defined as the interval within which, with a 95 percent confidence, the mean
scour depth occurs for a given set of independent variables. The prediction
interval is then defined as the interval within which, with a 95 percent
confidence, an individual scour depth (response) is expected for a given set of
independent variables.
The following formulas were developed for heads greater than or
100 ft.
y = 0.24q0.65 H0.32 with R2 = 0.75 ........................
ln Uc, ln Lc = ln Y ± 2.145 s ()/) .............................
where
s2 (Y) = 0.8469 + 0.02537 {ln q)2 + 0.02040 (ln H)2
-0.2045 ln q -0.06722 ln H
equal
(1)
(2)
-0.02323 {ln H) (ln q) ••••••••••.••••••••••••••••• (3)
and
ln Up, ln Lp = ln )I± 2.145 S(d) ..............................
where
s2 {d) = s2 (Y) + o.1o3o ......................................
where
V = Estimated scour depth below tailwater (feet)
q = unit discharge on impact (cubic feet per second per foot)
H = head measured from reservoir level to tailwater level (feet)
R =correlation coefficient
ln Uc, ln Lc = natural log of the upper and lower limits of the
confidence interval
ln Up, ln Lp = natural log of the upper and lower limits of the
prediction interval
5-1
(4)
(5)
to
s2 (Y) = estimated variance of V
s2 (d) = estimated variance of individual response
The following equations were developed for heads less than 100 feet.
Y = 1.48 q0.36 H0.38 with R2 = 0.82 ......................
-+ -ln Uc, ln Lc = ln Y -2.131 S (Y) .............................
where
s2 (Y) = 0.1782 + 0.00503 (ln q)2 + 0.01357 (ln H)2
-0.02466 ln q-0.05778 ln H
(6)
(7)
-0.00787 (ln q) (ln H) •••••••••••••••.•••.••.••••• (8)
and
ln Up, ln Lp -+ = ln Y-2.131 S (d) •••••••••••••••••.••••••.••••• (9)
where
2 - 2 -S (d) -S (Y) + 0.1151 ••.••••••••••••••••••.••••••••••••••••• (10)
The range of applicability of the higher head equation is:
167 ~ q0.65 H0.32 ~ 903
And the range of appicability of the lower head equation is:
6.5 ~ q0.36 H0.38 ~ 57.9
Equations (1) and (6) are sho'lm in Figures 5.1 and 5.2, respectively, together
with the data sets used in each regression. Confidence and prediction intervals
for each data set are shown in Figures 5.3 and 5.4.
The regression estimates and the upper limit of the prediction intervals are
shown for comparison purposes with previously developed simple equations on
Figures 3.2 and 3.3.
5.2 -Discussion
The equations of Section 5.1 were developed on statistical grounds and were
assumed to be dependent on unit discharge and head only. The actual scour depth
for a particular prototype condition may vary greatly from the regression esti-
mate. Such variations, as shown by the prediction limits, may be caused by:
5-2
-the extent of aeration of the discharging jet
-the geometry of spillway chute and flip bucket
-the downstream water levels
-the distribution, variation and duration of the flow
-the geological conditions at the area of impact.
Each of these factors are by themselves complex and often impossible to assess
beforehand. However, a qualitative assessment can generally be made, based on
site observations and spillway design aspects. This study does not provide, nor
quantitatively evaluate such information. Therefore, for the level of study
effort presented herein, the use of sound engineering judgment is strongly
emphasized.
5-3
TABLE 5.1-PROTOTYPE SCOUR DATA
Name of Unit Discharge Head Observed Scour
Number Spillway Country (cfs) ( ft) Depth ( ft) Reference
Maithon India 368 114.0 40.0 (9)
2 Panchet Hill India 270 107.0 26.5 (9)
3 Hirakud India 780 105.0 60.0 (9)
4 Gandhi Sagar India 460 158.0 64.0 (9)
5 Mandira India 230 41.0 64.0 (9)
6 Tilaiya India 39 80.0 34.0 (9)
7 Brazeau Canada 33 98.4 29.5 ( 19)
8 Assekinski Fergan 28 5.9 8.2 (26)
9 Hoschtedt Germany 19 6.2 7.8 (26)
10 Beznau Switzer land 183 20.7 46.9 (26)
11 Unknown Unknown 646 24.0 53.2 (26)
12 Konovingo United States 344 85.3 36.1 (26)
13 Unknown Unknown 538 45.9 59.1 (23)
14 Unknown Unknown 151 29.5 21.0 ( 18)
15 Unknown Unknown 1,830 173.9 180.5 ( 18)
16 Overflow Dam Unknown 646 55.8 55.8 (22)
17 Unknown Unknown 517 62.3 78.7 ( 18)
18 Unknown Unknown 754 62.3 105.0 ( 18)
19 Kariba Rhodesia 754 328.0 230.0 (24)
20 Akosombo Ghana ? ? 137.8 (25)
21 Grand Rapids Canada 975 61.0 90.0 ( 1)
22 Konolopoga Unknown ? 37.4 15.7 ( 17)
5-4
TABLE S.1 -(cont'd)
Name of Unit Discharge Head Observed Scour
Number Spillway Country (cfs) ( ft) Depth ( ft) Reference ---
23 Kindel Unknown 1 S1 37.4 20.S (22)
24 Bakurtsikhe Unknown so 42.7 36.7 (22)
2S Tarbela Pakistan ? ? ? (S)
26 Long Spruce Canada 794 78.0 60.0 (4)
27 Dneproges Unknown 361 123.0 85.3 (24)
28 Far had Unknown S6S 47.0 74.S (24)
29 Bukhtarma USSR S22 220.0 ? (24)
30 Bratsk USSR 42S 313.0 78.4 (24)
31 Sayan-Shushensk USSR 1,216 S4S.O 191.9 (24)
32 Krasnnoyarsk USSR 737 262.0 113.2 (24)
33 Bhakra India 1,416 541.0 ? (24)
34 Picoti Portugal 1 ,47S 162.0 203.4 (24)
3S Inguri USSR 1,023 738.0 264.1 (24)
36 Toktogol USSR 700 600.0 107.0 (24)
37 Kukuan Taiwan 816 278.9 91.9 (8)
38 Wuchieh Taiwan 741 160.8 65.6 (8)
39 Tienlung Taiwan 377 111.6 65.6 (8)
40 Houlung Taiwan 269 101.7 62.3 (8)
41 Shihmen Taiwan 1,025 318.3 98.4 (8)
5-5
300
250
200
1-
IJJ
IJJ u..
J:
ti:l50
IJJ
0
a::
::::>
0
(.)
(/)
100
50
0 I
0 100
63S
6'34 /
/
v
5
/ r< ESTIMATED
SCOUR DEPTH
/v 6
6'37 41
30
64( 6 '38
61
6
2
I I I I
200 300 400 500 600 700 800 900 1000
q0.65 H0.32
LEGEND
6i ith PROTOTYPE OBSERVED SCOUR DEPTH
ESTIMATED SCOUR DEPTH •
HEADS GREATER THAN OR EQUAL TO 100 FEET BIR
5-6 FIGURE 5.1
-.... w w
lL.
J: .... a. w
0
a:: ::>
0
(.)
(/)
110 r-
t:::..
18
100
90
80
70
60
50
40
r-/_
617 v / 1-
t:::..!
k:::
13 t:::..-
26
616
r-
t:::..IO
/ ESTIMATED
SCOUR DEPTH
e:. 612 24 t:::..6
30 1-7
t:::.. A 20 v 14-23
~ 10
0
0 10 20
LEGEND
30
q0.36 H0.38
40
e:-1 ith PROTOTYPE OBSERVED SCOUR DEPTH
ESTIMATED SCOUR DEPTH
HEADS LESS THAN 100 FEET
5-7
50 60
FIGURE 5.2 [i]
1-
LLJ
LLJ u.. -J:
1-a.
LLJ
0
0::
~
0
<..>
U)
600
500
400
300
200
100
0
0 100
p~!5
31
pl!5
p34
P. p41 36
cl!5
~3!5 3!5
p C31
/ p37.o;9
19
p32
634 CCV I-""
p36 36 6
p3 Cu I
p30 C19
~ ~. p4 C31
2 ~~~ c 3 c~ ~ 1 cl!5
P1 :...___c30 '-': C19 -6c34 p31 C~37 C 41 P.p2 6 27 32 36 p3!5
40 c~~ C4 30 c~ p~p34 ~!5 640~ 4 6 6~ p37
c c;;F3 P. ~9 40C2 c C4 3 P 32 ~~ p 38 ~ Cll 39P. P303 4<6~~7 4
~ r~
200 300 400 500 600
q0.65 H0.32
700 800 900 1000
LEGEND
ith PROTOTYPE OBSERVED SCOUR DEPTH
ith PROTOTYPE UPPER AND LOWER LIMITS
OF THE CONFIDENCE INTERVAL
ith PROTOTYPE UPPER AND LOWER LIMITS
OF THE PREDICTION INTERVAL
PREDICTION AND CONFIDENCE LIMITS [i]
HEADS GREATER THAN OR EQUAL TO 100 FEET IPIII
5-8 FIGURE 5.3 ftU [11
180
LEGEND 1"26
p21
170
~, ith PROTOTYPE OBSERVED SCOUR DEPTH
ci ith PROTOTYPE UPPER AND LOWER LIMITS
OF THE CONFIDENCE INTERVAL
160 pi lth PROTOTYPE UPPER AND LOWER LIMITS PIS OF THE PREDICTION INTERVAL
150 !-
pl6
140 17
~2
130 !-p2S
~3
120
110
1-
!-~I c2s
~IS C21
LLJ
LLJ 100 u.
Cis
:I:
1-90 a.
LLJ
0
a:
:::l 80 0 u
CJ)
70
60
50
40
30
20
-p5 / c16
Ctf17
p23 / v
13 -p7
Pa Pl4 Cu
?jo ~5 c26 c 21~=--L ... 1:;1S 26
p24 Al6
!-c5
c c17 c,6
C.,~1o 13c2s 12
C6 C23 / scEJJAM~T~E~
C10 c,4 R
A24 c5 Cu ~12 PIS
p21 26
~ p16
!-
c24 s p17
!7 C23 ~3P2s p12
c14
"' P8 C1o PII
r.-6 A14~--P5 9 c c:., 23
10 kr. 9
8 8
0
0
v 24 R ~0 14 p23 6
p p7
24
10 20 30 40
q 0.36 H0.38
PREDICTION AND CONFIDENCE LIMITS
HEADS LESS THAN 100 FEET
5-9
50 60
FIGURE 5.41 RIR I
6 -CONCLUSIONS AND RECOMMENDATIONS
6.1 -Conclusions and Recommendations
This study resulted in equations which, when applied with sound engineering
judgment, will predict a scour depth with sufficient accuracy for
feasibilitylevel layout designs.
It is recommended that the estimated scour depth curve with the design flood
discharge be used in the preliminary design. Further, it is recommended that
the scour hole be excavated to this depth if excessive excavation in resistive
rock is not required. Should excessive rock excavation be necessary,
consideration should be given to reducing excavation depth consistent with
actual rock levels, resulting in a depth corresponding to flows somewhat lower
than the design flood discharge.
The basic study described herein considers only published information and fails
to quantitatively analyze all known parameters. More detailed studies involving
additional parameters will be valuable and this could reduce the extent of qual-
itative engineering judgments.
Although scale model studies do not easily yield an accurate qualitative assess-
ment of the scouring process, their value is vested in identification of the
effects of changes to the spillway, chute and flip bucket geometry. For this
reason study efforts for more detailed designs should incorporate testing of
scour hole developnent by scale model.
6.2 -Application to the Susitna Project
The previously formulated equations were applied in a general manner to the pro-
posed Watana and Devil Canyon dams of the Susitna Hydroelectric Project. Head
will be relatively constant at both sites, and thus, unit discharge is the only
independent variable. The estimated scour depth together with all 1 imits are
plotted versus unit discharge in Figure 6.1 for Watana, with a 725 foot head.
For a Watana total discharge of 115,000 cfs and a unit discharge of 1,435 cfs,
the scour depth is esimated at 225 feet in the plunge pool below the auxiliary
chute spillway. At Devil Canyon the scour depth will be of the order of 240
feet for the main spillway total discharge of 125,000 cfs and a unit discharge
of 1,925 cfs. The scour relationships for Devil Canyon, with a 565 foot head,
are shown in Figure 6.2. These figures combined with appropriate engineering
judgement are to be employed to determine the feasibility design scour depth for
flip bucket and free overfall schemes.
6-1
700
600
500
...
IJJ 400
IJJ
LL
200
100
0
/
~PPER l.!IMIT-v
PREDICTION INTERVALZ_
L v
/
v
v
/ L_
v fJI"I"t.K L.:IMIT-v
/ ~ONFIDENCE INTV
/
v
/ v
v v v
/ ~STIMATED SCOUR __,.,.,.
/ bEPTH ,.. ......... v v v i-"""'
/ / ...,/ v v v v r-OWER LIMIT-~ --~ONFIDENCE
J / NTERVAL---
;/ / / v ~ ~.~I MIT PREDICTION ---~ -NTERVAL___.,..-r/ / v ~ ~
~ ~ ~ ..,.
~ ~ ...-
0 200 400 600 800 1000 1200 1400 1600 18.00 2000
UNIT DISCHARGE
(CUBIC FOOT PER SECOND PER FOOT)
SCOUR AT WATANA -725 FOOT HEAD
FIGURE 6.1 [i] 6-2
1-
LLI
700
600
500
~ 400
:c:
1-a..
LLI
0
a:: 6 300
(.)
U)
200
100
0
.L
UPPER LIMIT -t--. v PREDICT ION INT RVAL-
L_
L_ v
L
v
/
v
UPPER LIMIT-~ v ~C' '"' v v v
L v /
v
v v ESTlMAT ED SCO ~ v-
/ DEPTH-t::>-"'
1/ v v ~ y LOWER LIMIT C pNADEN ~ / _,.-. INTERvt L----I / v ~ .-
/ v ~
/ ~ LOWER LIMIT P REDICTIO N-
INTERVA L .....,
j/ / """ v ~ -
/ ----~
L.----
~ y ---
0 200 400 600 800 1000 1200 1400 1600 1800 2000
UNIT DISCHARGE
(CUBIC FOOT PER SECOND PER FOOT )
SCOUR AT DEVIL CANYON-565 FOOT HEAD ~
6-3 FIGURE 6.2 •
BIBLIOGRAPHY
1. Acres Consulting Service, "Spillway Scour Hole Model Study" unpublished
report for Grand Rapids Generating Station, Manitoba Hydro, August 1968.
2. Albertson, Dai, Jenson, and Rouse, "Diffusion of Submerged Jets",
Proceedings of the A.S.C.E, No. 10, 1948.
3. Altinbilek and Basmaci, "Localized Scour at the Downstream of Outlet
Structures", paper presented to the Eleventh International Congress on
Large Dams, Madrid, 1973.
4. Carson, R., personal communication.
5. Chao, Paul C., "Tarbela Dam-Problems Solved by Novel Concretes", Civil
Engineering A.S.C.E, December 1980.
6. Chee and Kung, "Stable Profiles of Plunge Basins", Water Resources
Bulletin, Volume 7, No 2, April 1971.
7. Chee and Padigar, "Erosion at the Base of Flipbuckets", Engineering
Journal, Volume 52, No. 11, November 1969.
8. Chian Min Wu, "Scour at Downstream End of Dams in Taiwan", IHAR,
International Symposium of River Mechanics, January 1973, Thailand.
9. Damle, P.M. et al, "Evaluation of Scour Below Ski-Jump Bckets of
Spillways", chapter from Model and Prototype Conformity, Volume 1, Central
Water and Power Research Station, Poona, India, 1966.
10. Doddiah, Altertson, and Thomas, "Scour From Jets", Proceeding, Minnesota
International Convention, September 1953.
11. Franke, P.A., "Uber Kolhildung and Kolhformen", Oesterreichische
Wasserwirtschaft, 12-13, 1960-1961.
12. Gunko et al, "Research on the Hydraulic Regime and Local Scour of Riverbed
Below Spillways of High Head Dams", XI Congress, International Association
for Hydraulic Research, Volume 1, 1965.
13. Johnson, G., "The Effect of Entrained Air on the Scouring Capacity of Water
Jets", XII Congress, IAHR, Volume 3, 1967.
14. Lencastre, A., "Descarregadores de Lamina Livre", L i sboa, LNEC, 1961
(Technical Paper No. 174).
15. Levi, J.J., "Effet Dynamique d• un Courant a Haute Turbulence Sue des
Ouvrages Hydrauliques et Sur le lit des Rivieres", IAHR Ninth Congress,
Dubrovnik, 1961.
16. Loughlin, Mehrotra, Chang, and Kennedy, "Scale Effects in Hydraulic Model
Tests of Rock Protected Structures", Iowa Institute of Hydraulic Research,
Report Number 124, February 1970.
17. Martins, Rui, "Contribution to the Knowledge on the Scour Action of Free
Jets on Rocky Riverbeds .. , paper presented to the Eleventh International
Congress on Large Dams, Madrid, 1973.
18. Martins, Rui, 11 Scouring of Rocky Riverbeds by Free-Jet Spillways", Water
Power and Dam Construction, April 1975.
19. McLean, F. et al, "The Brazean Spillway", The Engineering Journal,
January/February 1971.
20. Meyer-Peter, E. and Muller R., "Affoui llements en Aval des Barrages", IAHR
Second Congress, Volume 1, Stockholm, 1948.
21. Mikhalev, M.A., "Determination of Depth of Scour of Erodable Foundation by
Falling Jet", Hydraulic Engineering Construction, No. 9, September 1960,
(Translation from Russian).
22. Mirtskhulava, Ts. E. et al, "Mechanism and Computation of Local and General
Scour in Non-Cohesive, Cohesive Soils and Rock Beds", IAHR Twelfth
Congress, Volume 3, Fort Collins, 1967.
23. Smol'yaninov, N.T., "Erosion du Radier a I'aval d'un Deversoir",
Gidrotekhnicheski Stroitel'stvo (Soviet Journal), February 1972.
24. Taraimovich, I.I., "Deformations of Channels Below High Head Spillways on
Rock Foundations", Hydrotechnical Construction, r4o. 9, September 1978,
(Translation from Russian).
25. Thomas, H.H., The Engineering of Large Dams, Wiley, 1976.
26. Vyzgo, M.S., "Reducao da Erosao e Qumento du Seccao da Lamina Descarregada
em Escoamentos Rapidos", Gidrotekhnicheski Stroitel 'stvo (Soviet Journal),
No. 7, 194 7.