(PDF) Geotechnical Aspects of Rock Erosion in Emergency Spillway Channels Supplemental Information on Prediction, Control, and Repair of Erosion in Emergency Spillway Channels

Technical

Report

REMR-GT-3

August

1998

US

Army

Corps

of

Engineers

Waterways

Experiment

Station

Repair,

Evaluation,

Maintenance,

and

Rehabilitation

Research

Program

Geotechnical

Aspects

of

Rock

Erosion

in

Emergency

Spillway

Channels

Supplemental

Information

on

Prediction,

Control,

and

Repair

of

Erosion

in

Emergency

Spillway

Channels

by

Christopher

C.

Mathewson,

Kerry

D.

Cato,

Texas

A&M

University

James

H.

May,

WES

19980911

079

Approved

For

Public

Release;

Distribution

Is

Unlimited

rjtiöQi

1

Prepared

for

Headquarters,

U.S.

Army

Corps

of

Engineers

The

following

two

letters

used

as

part

of

the

number

designating

technical

reports

of

research

published

under

the

Repair,

Evaluation,

Maintenance,

and

rehabilitation

(REMR)

Research

Program

identify

the

problem

area

under

which

the

report

was

prepared:

Problem

Area

Problem

Area

Electrical

and

Mechanical

Environmental

Impacts

Operations

Management

cs

Concrete

and

Steel

Structures

EM

GT

Geotechnical

El

HY

Hydraulics

OM

CO

Coastal

The

contents

of

this

report

are

not

to

be

used

for

advertising,

publication,

or

promotional

purposes.

Citation

of

trade

names

does

not

constitute

an

official

endorsement

or

approval

of

the

use

of

such

commercial

products.

The

findings

of

this

report

are

not

to

be

construed

as

an

official

Department

of

the

Army

position,

unless

so

desig-

nated

by

other

authorized

documents.

®

PRINTED

ON

RECYCLED

PAPER

Repair,

Evaluation,

Maintenance

and

Technical

Report

REMR-GT-3

Rehabilitation

Research

Program

August

1998

Geotechnical

Aspects

of

Rock

Erosion

in

Emergency

Spillway

Channels

Supplemental

Information

on

Prediction,

Control,

and

Repair

of

Erosion

in

Emergency

Spillway

Channels

by

Christopher

C.

Mathewson,

Kerry

D.

Cato

Center

for

Engineering

Geosciences

Texas

A&M

University

College

Station,

TX

77843-3115

James

H.

May

U.S.

Army

Corps

of

Engineers

Waterways

Experiment

Station

3909

Halls

Ferry

Road

Vicksburg,

MS

39180-6199

Final

report

Approved

for

public

release;

distribution

is

unlimited

Prepared

for

U.S.

Army

Corps

of

Engineers

Washington,

DC

20314-1000

Under

Work

Unit

32638

US

Army

Corps

of

Engineers

Waterways

Experiment

Station

HEADQUARTERS

BUU3MQ

ENTRANCE

ENVMCfMENTAi.

IABORATORV

FOR

INFORMATION

CONTACT:

PUBLIC

AFFAIRS

OFFICE

U.S.

ARMY

ENGINEER

WATERWAYS

EXPERIMENT

STATION

3909

HALLS

FERRY

ROAD

VICKSBURG,

MISSISSIPPI

39180-6199

PHONE:

(601)

634-2502

STRUCTURES

LABORATORY

AREAOFfiESEKVATOI*

ZTstxtr

Waterways

Experiment

Station

Cataloging-in-Publication

Data

Mathewson,

Christopher

C.

Geotechnical

aspects

of

rock

erosion

in

emergency

spillway

channels:

supplemental

information

on

prediction,

control,

and

repair

of

erosion

in

emergency

spillway

channels

/

by

Christopher

C.

Mathewson,

Kerry

D.

Cato,

James

H.

May;

prepared

for

U.S.

Army

Corps

of

Engineers.

42

p.:

ill.;

28

cm.

—

(Technical

report;

REMR-GT-3)

Includes

bibliographic

references.

1.

Rocks

—

Erosion.

2.

Soils

—

Erosion.

3.

Spillways

—

Repairing

—

Design.

I.

Cato,

Kerry

Don,

1959-

II.

May,

James

H.

III.

United

States.

Army.

Corps

of

Engineers.

IV.

U.S.

Army

Engineer

Waterways

Experiment

Station.

V.

Repair,

Evaluation,

Maintenance

and

Rehabilitation

Research

Program.

VI.

Title.

VII.

Title:

Supplemental

information

on

prediction,

control,

and

repair

of

erosion

in

emergency

spillway

channels.

VIII.

Series:

Technical

report

(U.S.

Army

Engineer

Waterways

Experiment

Station);

REMR-GT-3.

TA7

W34

no.REMR-GT-3

Contents

Preface

v

Conversion

Factors,

Non-SI

to

SI

Units

of

Measure

v

ii

1—Introduction

1

Objectives

and

Approach

of

Study

1

Problem

Statement

2

Purpose

of

an

Emergency

Spillway

3

Erosion

Studies

5

2—Analysis

and

Prediction

of

Spillway

Erosion

9

Classification

of

Erosion

Damage

-9

Spillway

Channel

Erosion

Equation

11

Predicting

Rock

Erosion

in

Spillway

Channels

2

3

3—Repair

and

Remediation

of

Spillway

Erosion

2

8

Channel

Geometry

2

8

Hydrologie

and

Hydraulic

Conditions

2

9

Geologic

Conditions

2

9

References

33

SF298

List

of

Figures

Figure

1.

Plan

view

(A)

and

cross

section

(B)

of

a

typical

emergency

spillway

channel

4

Ml

Figure

2.

Relationship

between

rock

erosion

classification

and

geometric

and

hydraulic

characteristics

of

the

spillway

channel

11

Figure

3.

Relationship

between

flow

depth

and

fall

height

14

Figure

4.

Flood

hydrograph

(A)

of

a

flow

event

that

reaches

peak

design

dis-

charge

showing

erosive

flows

during

vented

(cross

hatched),

unvented,

and

peak

flows.

Erosion

hydrograph

(B)

showing

the

conditions

when

vented

(hatched)

and

unvented

erosion

occurs

...

15

Figure

5.

Rock

erosion

potential

class

based

on

lithology

18

Figure

6.

Rock

erosion

potential

class

based

on

rock

substance

properties.

..

19

Figure

7.

Rock

erosion

potential

class

based

on

rock

genesis

and

environment

of

formation

which

produce

the

first-order

discontinuities

2

0

Figure

8.

Rock

erosion

potential

class

based

on

tectonic

history

of

the

rock

which

produces

the

second-order

discontinuities

2

1

Figure

9.

Schematic

cross

sections

of

model

knickpoints

investigated

by

Clemence

(1988)

2

3

Figure

10.

Rock

erosion

potential

class

based

on

rock

mass

properties

2

4

Figure

11.

Erosion

risk

classification

based

on

slope,

flow

velocity,

and

geometric

anomaly

within

the

spillway

channel

2

5

Figure

12.

Spillway

erosion

assessment

sheet

used

to

compare

erosion

risk

to

erosion

potential

for

each

geologic

and

geometric

section

within

the

spillway

channel

2

6

Figure

13.

Erosion

potential

classification

based

on

lithology,

rock

substance,

material

genesis,

postformational

discontinuities

(tectonics),

and

rock

mass

properties

2

7

Figure

14.

Conceptual

section

of

a

cutoff

wall

at

the

toe

of

a

concrete

discharge

structure

to

resist

undercutting

3

0

Figure

15.

Conceptual

designs

for

the

remediation

of

various

forms

of

spillway

erosion

3

2

IV

Preface

This

study

addresses

the

geologic

factors

that

control

rock

erosion

in

emer-

gency

spillway

channels,

develops

a

technique

to

evaluate

the

risk

and

predict

the

potential

erosion

of

rock

and

soil

exposed

to

hydraulic

attack

during

a

flow

event

in

the

channel,

and

provides

design

concepts

for

the

repair

and

rehabilitation

of

spillway

erosioa

The

study

was

conducted

under

the

Repair,

Evaluation,

Maintenance,

and

Rehabilitation

(REMR)

Research

Program.

The

REMR

Program

Manager

was

Mr.

William

F.

McCleese,

Structures

Laboratory,

WES.

Mr.

Mike

Klosterman

(CECW-E),

Headquarters,

U.S.

Army

Corps

of

Engineers,

was

the

Technical

Monitor.

This

work

was

conducted

under

the

direct

supervision

of

Mr.

J.S.

Huie,

Soil

and

Rock

Mechanics

Division

(SRMD),

and

Dr.

James

H.

May

and

Mr.

John

B.

Palmerton,

the

Principal

Investigators,

Earthquake

Engineering

and

Geophysics

Division

(EEGD),

Geotechnical

Laboratory

(GL),

U.S.

Army

Engineer

Water-

ways

Experiment

Station

(WES).

General

supervision

was

provided

by

Mr.

Joe

L.

Gatz,

Chief,

Engineering

Geology

Branch;

Dr.

A.G.

Franklin,

EEGD;

Dr.

D.C.

Banks,

SRMD;

and

Dr.

W.F.

Marcuson

III,

Director,

GL.

Most

of

the

data

for

this

research

came

from

two

Federal

agencies:

the

U.S.

Army

Corps

of

Engineers

(CE)

and

the

U.S.

Department

of

Agriculture,

Soil

Conservation

Service

(SCS).

Drs.

Dave

Patrick

and

Chris

Cameron,

of

the

University

of

Southern

Mississippi,

contributed

many

hours

of

helpful

discussion.

Messrs.

Dave

Ralston,

Lou

Kirkaldie,

John

Brevard,

John

Moore,

and

Jim

Hyland

of

the

SCS

were

extremely

supportive

by

opening

their

files

and

providing

constructive

feedback

on

ideas

and

written

reports.

Messrs.

Darrel

Temple,

Kerry

Robinson,

and

Gregg

Hanson

of

the

Agricultural

Research

Service

-

Hydraulic

Structures

Laboratory

furnished

many

new

insights

and

observations

concerning

spillway

erosioa

This

report

draws

heavily

from

the

doctoral

dissertations

by

Dr.

May

and

Dr.

Kerry

D.

Cato,

completed

under

the

supervision

of

Dr.

Chris-topher

C.

Mathewson,

Director,

Center

for

Engineering

Geosciences

and

Professor

of

Engineering

Geology

at

Texas

A&M

University,

College

Station,

TX.

At

the

time

of

the

publication

of

this

report,

Dr.

Robert

W.

Whalin

was

Director

of

WES.

The

Commander

was

COL

Robin

R.

Cababa,

EN.

The

contents

of

this

report

are

not

to

be

used

for

advertising,

publication,

or

promotional

purposes.

Citation

of

trade

names

does

not

constitute

an

official

endorsement

or

approval

of

the

use

of

such

commercial

products.

VI

Conversion

Factors,

Non-SI

to

SI

Units

of

Measurement

Non-SI

units

of

measurement

used

in

this

report

can

be

converted

to

SI

units

as

follows:

1

Multiply

By

To

Obtain

cubic

feet

0.02831685

cubic

metres

feet

0.3048

metres

inches

25.4

millimetres

1

pounds

per

cubic

foot

27.6799

grams

per

cubic

centimetre

1

pounds

per

square

inch

6.894757

kilopascals

H

VII

1

Introduction

Objectives

and

Approach

of

Study

The

technology

and

methodology

to

evaluate

erosion

in

emergency

spillway

channels

are

in

their

infancy

with

only

recent

detailed

case

histories

providing

a

foundation;

no

synoptic

study

of

case

histories

or

theoretical

basis

of

soil

and

rock

erosion

exists.

Current

methods

are

not

applicable

to

most

spillways

in

which

material

conditions

range

from

loose

soil,

to

slightly

indurated

soil,

to

weathered

rock,

and

finally

to

massive,

unfractured

unweathered

rock.

Only

erosion

of

loose

sediment

has

been

studied

in

detail.

This

study

is

a

synoptic

approach

that

uses

case

histories

of

emergency

spillway

erosion

to

determine

the

effects

of

wide

ranges

of

geologic

materials

on

the

erosion

process.

The

goal

of

this

study

is

to

determine

which

geologic

factors

are

the

most

useful

predictors

of

emergency

spillway

channel

erosion,

to

use

them

to

develop

a

technique

to

predict

erosion

potential

in

emergency

spillway

channels,

and

to

establish

design

concepts

for

channel

design

and

erosion

repair

projects.

This

objective

was

achieved

through

the

following

subobjectives:

a.

Assessment

of

erosion

damage

at

sites

that

have

received

flows

through

an

analysis

of

the

geometric,

hydraulic,

and

geologic

conditions

affecting

site

performance

using

postflow

surveys

and

observations.

b.

Determination

of

geologic

characteristics

of

the

spillway

channel

foundation

by

reviewing

construction

data

and

through

observation.

c.

Development

of

a

geologic

erosion

equation

by

assessing

the

effect

of

various

geologic

factors

on

spillway

channel

erosion.

d.

Combining

geologic

equation

with

the

geometric

and

hydraulic

factors

to

develop

an

erosion

potential

technique.

e.

Establishment

of

design

concepts

and

recommendations

for

the

design

and

repair

of

erosion

in

emergency

spillway

channels.

These

objectives

were

fulfilled

through

the

following

approach:

Chapter

1

Introduction

a.

Determining

erosion

damage

at

spillways

that

have

experienced

flows

through

the

use

of

observations

and

postflow

surveys

to

identify

where

damage

was

concentrated

in

the

channel

and

describe

the

damage.

b.

Determining

the

geologic

site

conditions

using

construction

data

and

obser-

vations

including

the

construction

of

geologic

cross

sections

parallel

to

the

longitudinal

channel

profiles,

the

identification

of

the

nature

of

each

geo-

logic

erosion

equation

component,

and

the

recording

of

any

geologic

influ-

ence

on

erosion

not

addressed

in

the

equation.

c.

Explaining

the

effect

of

each

geologic

erosion

equation

component

on

the

erosion

performance

for

each

site,

and

if

the

erosion

could

not

be

explained

by

the

existing

factors,

define

new

factors

which

accounted

for

the

erosion.

d.

Synthesizing

the

influence

of

each

component

of

the

geologic

erosion

equation

for

all

sites.

e.

Developing

conceptual

designs

for

the

repair

and

rehabilitation

of

eroded

spillway

channels.

Problem

Statement

Significant

numbers

of

excavated,

earthen-floored

emergency

spillway

channels

are

currently

used

at

a

large

number

of

small

dams

and

at

a

moderate

number

of

large

dams

in

the

United

States.

These

channels

are

excavated

into

all

types

of

rock

and

soil

and,

unlike

service

spillways

which

may

flow

with

greater

frequency

and

hence

have

more

erosion

protection,

are

designed

to

flow

rarely

and

experi-

ence

minor

erosion

(U.S.

Army

Corps

of

Engineers

(US

ACE)

1965).

However,

erosion-producing

flows

over

the

past

10

years

have

resulted

in

a

need

for

mainte-

nance

and

redesign

of

many

emergency

spillway

channels.

These

spillway

perfor-

mances

indicate

that

little

is

known

about

how

to

predict

in

situ

soil

and

rock

resistance

to

hydraulic

stresses.

Emergency

spillway

erosion

will

become

a

greater

problem

as

more

dams

experience

spillway

flows.

The

greatest

probability

of

spillway

flow

appears

to

occur

within

the

first

50

years

of

the

life

of

a

dam.

Relating

the

number

of

operat-

ing

dams

to

spillway

flow

events

indicates

an

alarming

trend.

Consider,

for

example,

the

buildup

of

large

dams

after

World

War

II.

Approximately

2,000

dams

existed

in

1946,

but

by

1986

this

number

almost

tripled

to

5,450

(American

Society

of

Civil

Engineers/US.

Committee

on

Large

Dams

(ASCE/USCLD)

1975,

1988).

It

is

unknown

how

many

large

dams

have

earthen-floored

spillways,

but

it

is

known

that

use

of

excavated

spillways

is

common

practice

with

earth-

or

rock-fill

structures;

83

percent

of

all

dams

operating

in

1986

were

earth-

or

rock-

fill

dams.

In

1946

historical

spillway

flows

totaled

11

events;

however,

51

spillway

flows

occurred

between

1946

and

1986

(ASC/USCLD

1975,1988).

Of

these

51

flows,

the

majority

were

on

dams

constructed

after

1946.

In

the

United

Chapter

1

Introduction

States,

approximately

63

percent

of

large

dams

are

less

than

50

years

old;

yet

fewer

than

2

percent

have

received

spillway

flow.

Large

dams

are

built

more

hydraulically

conservative

than

are

small

dams,

and

their

spillways

are

designed

to

be

seldom,

if

ever,

used.

An

examination

of

dams

designed

and

constructed

by

the

U.S.

Department

of

Agriculture,

Soil

Conservation

Service

(SCS),

predominantly

small-

and

medium-

sized

dams,

provides

an

even

stronger

argument.

The

largest

number

of

dams

was

constructed

during

the

1964-1969

time

period

when

over

5,000

dams

were

completed

(SCS

1989a).

The

surge

of

SCS

dam

building

is

over,

evidenced

by

the

fact

that

only

334

dams

have

been

constructed

during

the

past

5

years.

The

total

number

of

SCS

dams

placed

in

operation

since

1954

is

22,785.

SCS

spillway

erosion

problems

were

first

documented

in

1957;

however,

significant

numbers

of

flows

did

not

occur

until

late

1960s

and

1970s

(Ralston

and

Brevard

1988).

The

peak

of

dam

construction

occurred

in

1964,

but

the

first

spillway

flows

for

most

sites

did

not

transpire

until

20

years

later.

Less

than

5

percent

of

all

SCS

dams

have

experienced

flows.

It

is

probable

that

larger

num-

bers

of

SCS

dams

will

receive

spillway

flows.

These

statistics

emphasize

the

importance

of

understanding

the

erosion

process

and

using

this

knowledge

to

establish

safety

review,

remediation,

new

design

and

redesign,

and

maintenance

policies.

Purpose

of

an

Emergency

Spillway

An

emergency

spillway

conveys

flood

flows

that

exceed

the

designed

storage

space

safely

past

a

dam.

Several

types

of

emergency

spillway

designs

are

cur-

rently

used

and

generally

classified

according

to

their

most

prominent

feature

including:

free

overfall

(straight

drop),

ogee

(overflow),

side

channel,

open

channel

(trough

or

chute),

conduit,

tunnel,

drop

inlet

(shaft

or

morning

glory),

baffled

apron

drop,

culvert,

and

siphon

(Golze

1977).

Emergency

spillways,

sometimes

called

auxiliary

spillways,

may

also

be

classified

as

controlled

or

uncontrolled,

depending

on

whether

they

are

gated

or

ungated.

Open

channel

spillways,

the

focus

of

this

study,

are

used

with

earth-fill

dams

more

often

than

any

other

spillway

type.

Because

83

percent

of

the

dams

currently

in

operation

are

earth

dams,

it

follows

that

a

high

number

of

open

channel

spill-

ways

are

also

operating.

Channel

profile

designs

have

an

upstream

entrance

channel,

a

control

structure,

and

a

downstream

exit

channel.

Control

structures

are

generally

placed

in

line

with

or

upstream

from

the

dam

centerline

(Figure

1).

Downstream

from

the

control,

the

exit

channel

is

constructed

at

minimum

grade

until

it

"daylights"

along

the

valley

Chapter

1

Introduction

EXIT

CHANNEL

-EMERGENCY

SPILLWAY

CONTROL

INLET

CHANNEL

B

•SUBCRITICAL

SUPERCRITICAL-

TTT

HYDRAULIC

JUMP

SUPERCRITICAL

-J

■

INLET

CHANNEL

CONTROL

SECTION

AND

CREST

STEEP

FLOOD-

SECTION

PLAIN

Figure

1.

Plan

view

(A)

and

cross

section

(B)

of

a

typical

emergency

spillway

channel

wall.

To

minimize

excavation,

the

channel

is

stopped

at

this

location;

the

altema

tive

would

be

to

grade

the

channel

a

long

distance

along

the

valley

wall

and

down

into

the

floodplain.

Depending

on

dam

height

and

valley

wall

geometry,

there

is

typically

some

drop-off

or

steeper

gradient

at

the

downstream

end

of

the

exit

channel

which

grades

to

the

floodplain

elevation.

4

Chapter

1

Introduction

Adequate

hydraulic

capacity

is

a

paramount

criterion

for

emergency

spillways

of

earth-fill

and

rock-fill

dams

which

would

potentially

be

destroyed

if

overtopped.

Wide

spillways

are

frequently

constructed

to

meet

this

hydraulic

capacity

and

to

maintain

shallow

flow

depths.

This

design

concept

calls

for

extensive

spillway

channel

excavation,

which

comprises

a

significant

amount

of

the

overall

construc-

tion

cost.

Channel

design

relies

on

an

assessment

of

flow

risk

versus

capitalization

cost.

Emergency

spillways

are

infrequently

used

with

inconsequential

erosion

damage

expected

during

operation

(USACE

1965).

However,

there

is

an

assumption

in

this

philosophy

that

expected

erosion

will

be

minor

and

repairable.

Recent

findings

show

that

numerous

erosion

estimates

were

too

low

and

that

excessive

erosion

occurred

with

flows

below

the

maximum

design

flow

(Cameron

et

al.

1986,

1988).

Erosion

Studies

The

American

Society

of

Civil

Engineers

and

the

United

States

Committee

on

Large

Dams

document

spillway

erosion

by

summarizing

large

dam

failures

and

incidents

dating

to

the

late

1800's

for

United

States

dams

(ASCE/USCLD

1975,

1988).

Where

possible,

factors

such

as

erosion

involving

only

concrete,

flow

damaging

only

the

controlling

gates,

or

dam

overtopping

by

excessive

discharge

are

listed.

However,

a

formidable

number

of

incidents

involve

erosion

of

soil

and

rock.

See

Cato

(1991)

for

a

discussion

of

historical

data.

The

Repair,

Evaluation,

Maintenance,

and

Rehabilitation

(REMR)

Research

Program

spillway

erosion

study

began

in

1984

as

a

result

of

substantial

spillway

erosion

at

the

U.S.

Army

Corps

of

Engineers'

(CE)

Grapevine

Dam,

TX,

in

1981,

and

the

CE-Saylorville

Dam,

IA,

in

1984

(Cameron

et

al.

1986,

1988).

Flow

magnitudes

represented

small

Probable

Maximum

Flood

(PMF)

percentages,

but

large

rock

volumes

were

eroded.

More

importantly,

it

was

felt

that

larger

flows

would

have

produced

spillway

breaches.

The

REMR

study

consisted

of

a

reconnaissance

stage

assessing

the

problem

magnitude

and

a

research

stage

addressing

specific

issues.

The

reconnaissance

study,

performed

by

a

multidisciplinary

team

based

out

of

the

U.S.

Army

Engineer

Waterways

Experiment

Station

in

Vicksburg,

MS,

entailed

contacting

every

CE

Division

to

identify

historic

spillway

flows.

Efforts

were

made

to

visit

each

site,

evaluate

geologic

materials

in

the

spillway

channel,

and

write

up

an

event

case

history.

Further

research

resulted

in

the

following

conclusions

(Cameron

et

al.

1988):

a.

Structural

and

stratigraphic

discontinuities

play

a

major

role

in

the

erosion

of

rock

by

changing

the

erosion

resistance

of

the

bed

material

and

forming

channel

gradient

changes.

Chapter

1

Introduction

b.

It

is

possible

to

rank

erosion

at

sites

by

comparing

volumes

of

material

removed.

c.

Headward

knickpoint

migration

can

be

exacerbated

by

negative

pressures

pulling

the

turbulent

forces

of

the

nappe

toward

the

natural

materials

in

the

headcut.

The

SCS

Emergency

Spillway

Flow

Study

Task

Group

(ESFSTG)

evaluated

the

performance

of

more

than

75

spillway

flows

to

refine

design

criteria

and

guide

repair

of

eroded

sites.

1

Erosion

severity

varies

from

no

damage

to

one

complete

breach.

Observations

show

that

most

eroded

material

consists

of

placed

soil

on

the

exit

channel

floor

and

residual

soil

on

the

natural

hillslope.

However,

the

SCS

is

still

trying

to

define

erodible

versus

non-erodible

rock.

Involvement

with

CE

and

SCS

spillway

studies

led

Cato

(1991)

to

probe

material

performance

case

histories.

Geometric

and

hydraulic

effects

on

erosion

processes

were

analyzed

by

statistically

comparing

erosion

damage

to

geometric

and

hydraulic

variables

for

16

sites;

portions

of

this

work

are

summarized

in

Cameron

et

al.

(1988).

The

analysis

resulted

in

a

method

to

classify

erosion

as

dominantly

downcutting,

transition

or

backcutting.

Conclusions

reached

were

as

follows:

a.

Generally,

there

is

only

minor

statistical

significance

among

the

attempted

correlations.

b.

The

R-squared

values

for

polynomial

regression

analyses

were

higher

than

those

for

linear

regression

analyses.

c.

Although

the

highest

single

R-squared

value

(0.79)

occurred

for

the

compar-

ison

of

volumetric

erosion

ranking

versus

steep

section

length

for

the

SCS

dams,

the

overall

R-squared

values

for

the

combined

SCS-CE

database

were

higher.

d.

The

R-squared

values

for

attempted

correlations

involving

the

geometric

parameters

were

somewhat

higher

than

those

involving

the

hydraulic

param-

eters

including

hydraulic

attack.

The

geologic

and

hydraulic

significance

of

the

regression

analyses

can

be

summarized

as

follows:

a.

Absence

of

overall

significant

statistical

correlation

among

the

sites

in

the

database

may

be

ascribed

to

variable

geological

conditions,

particularly

the

nature

of

structural

or

lithologic

discontinuities.

1

Personal

Communication,

D.C.

Ralston,

1989,

National

Design

Engineer

(retired),

Engineer-

ing

Division,

Soil

Conservation

Service,

U.S.

Department

of

Agriculture,

Washington,

D.C.

Chapter

1

Introduction

b.

Regression

analyses

indicate

that

the

effect

of

hydraulic

parameters

(water

depth

and

velocity)

play

a

minor

role

in

predicting

the

nature

and

extent

of

erosion

of

rock-lined

spillway

channels,

although

these

parameters

should

be

important

for

predicting

erosion

in

noncohesive

soils

and

sediments.

c.

Higher

R-squared

values

geometric

parameters

indicate

that

a

knickpoint

along

the

longitudinal

profile

of

a

spillway

channel

is

important

in

initiating

erosion

and

controlling

the

degree

of

erosion

that

will

occur.

This

initial

study

guided

succeeding

analyses

to

consider

all

three

general

variables

(geometry,

hydraulics,

and

geology).

It

was

found

that

site

geometry

is

a

more

critical

factor

in

controlling

the

initiation

of

erosion

than

flow

hydraulics

and

that

site

geology

appears

to

serve

as

the

dominant

erosion

control

factor.

A

follow-up

statistical

analysis

incorporated

geologic

variables

and

a

more

detailed

database

(Cato

1991).

Improvements

over

the

included

the

following:

a.

The

database

included

only

SCS

dams

in

an

attempt

to

consider

dams

of

similar

size

and

hydraulic

conditions.

b.

Each

channel

was

divided

into

segments

of

similar

gradients;

then,

only

damages

from

that

segment

were

compared

to

hydraulic,

geometric,

and

geologic

variables.

c.

Performance

of

soil

and

rock

materials

were

compared

separately

using

linear

and

multiple

linear

regression

techniques

along

with

graphical

dis-

plays

of

all

comparisons.

d.

Comparisons

of

variables

to

the

following

measures

of

erosion:

(1)

Damage

classificatioa

(2)

Area

eroded

in

channel.

(3)

Total

volume

of

soil

and

rock

eroded

in

channel

(and

unit

volume).

(4)

Volume

of

soil

eroded

in

channel

(and

unit

volume).

(5)

Volume

of

rock

and

transition

material

eroded

(and

unit

volume).

(6)

Gully

morphology

(knickpoint

shape

and

depth

and

hydraulic

radius).

Rock

mass

classification

systems,

such

as

those

by

Kirsten

(1988),

Barton

(1988),

and

Bieniawski

(1988)

developed

for

ripping

and

tunneling,

had

poor

correlations.

Rock

Quality

Designation

(RQD),

reported

by

Woodward

(1985)

to

be

a

good

indicator

of

erodibility,

proved

an

ineffective

indicator

for

the

sites

Chapter

1

Introduction

analyzed

in

this

study.

The

more

descriptive

system,

the

Unified

Rock

Classifica-

tion

System

(URCS),

and

comparison

of

its

components

have

merit.

Overall,

the

linear

correlations

are

low,

trends

are

not

strong,

and

some

apparent

correlations,

such

as

Standard

Penetration

Test

(SPT)

blow

counts,

are

not

valid

because

of

sparse

data.

Multiple

linear

correlations

also

show

low

correlation

coefficients,

generally

below

0.50,

and

show

only

moderate

increases

over

linear

coefficients.

Observations

at

numerous

sites

attesting

to

geologic

control

cannot

be

ignored;

therefore,

statistical

approaches

with

this

amount

or

type

of

data

are

not

adequate

to

show

geologic

effects.

8

Chapter

1

Introduction

2

Analysis

and

Prediction

of

Spillway

Erosion

Classification

of

Erosion

Damage

Investigated

spillways

came

from

two

sources,

the

CE

and

SCS.

SCS

sites

were

generally

more

completely

documented

and

were

more

similar

in

their

size

and

characteristics

than

CE

sites.

The

CE

sites

provide

good

geologic

and

hydrau-

lic

variation

and

represent

larger

structures.

Erosion

classification

enabled

damage

comparisons

between

sites.

This

was

performed

by

developing

an

emergency

spillway

incident

classification

and

classifying

each

channel

segment.

Erosion

can

be

classified

in

many

ways,

including:

a.

Erosion

effect

on

dam

operation.

b.

Erosion

process.

c.

Volume

of

material

eroded.

d.

Area

of

exit

channel

scoured.

e.

Repair

cost.

The

incident

classification

developed

in

this

study

is

based

upon

the

following

hierarchy:

a.

Damage

to

spillway

structure

whether

it

be

a

breach,

severe

erosion,

or

only

minor

erosion.

b.

Type

of

materials

eroded

whether

it

is

placed

topsoil

or

in

situ

rock

or

soil.

The

erosion

damage

classification

system

developed

by

Cato

(1991)

is

pre-

sented

in

the

following:

Chapter

2

Analysis

and

Prediction

of

Spillway

Erosion

a.

Failure

Type

2

(F2).

Applies

only

if

breach

occurs.

A

major

failure

of

an

operating

dam

which

involved

complete

abandonment

of

the

dam.

b.

Failure

Type

1

(Fl).

Applies

only

if

breach

occurs.

A

failure

of

an

operat-

ing

dam

which

at

the

time

may

have

been

severe,

but

was

of

a

nature

and

extent

that

permitted

successful

damage

repair

and

continued

operation.

For

example,

an

Fl

failure

could

be

produced

by

breach

of

spillway

control

section

with

uncontrolled

release

of

reservoir

waters.

c.

Erosion

Type

2

(E2).

Erosion

in

excavated

spillway

channel

and/or

erosion

at

the

downstream

end

of

the

excavated

spillway

that

consisted

of

a

great

deal

of

natural

in

situ

material

in

addition

to

placed

topsoil.

This

type

of

erosion

would

cause

severe

damage

to

the

spillway,

but

would

not

cause

breach

and

release

of

reservoir

waters.

This

type

of

erosion

involves

the

same

processes

as

in

El,

but

damage

severity,

and

hence

cost

to

repair,

is

greater.

d.

Erosion

Type

1

(El).

Erosion

in

excavated

spillway

channel

and/or

erosion

at

the

downstream

end

of

the

excavated

spillway

that

consisted

primarily

of

the

removal

of

soil

and

possibly

some

rock,

material.

This

type

of

erosion

would

cause

minor

to

moderate

damage

to

the

spillway

exit

channel

or

downstream

reaches.

This

type

of

erosion

would

also

include

headward

migration

of

gullies

if

the

erosion

is

removing

natural

in

situ

rock

or

a

residual

soil

and/or

formation

of

scour

holes

where

the

spillway

waters

enter

the

floodplaia

e.

Scour

Type

1

(SI).

Erosion

in

the

excavated

spillway

that

consists

of

removal

of

only

topsoil.

A

technical

breach,

one

in

which

erosion

progresses

through

the

control

section,

could

be

S1

if

the

depth

of

erosion

does

not

involve

in

situ

rock

or

residual

soil.

/.

Vegetation

Removal

(VR).

Removal

of

some

vegetation

and

only

minor

amounts

of

soil.

g.

No

Damage

(ND).

Emergency

spillway

suffered

no

soil

erosion

or

vegeta-

tion

removal.

The

observed

damages

range

from

almost

no

erosion

in

any

channel

segments

at

site

SCS

AR

-

West

Fork

Point

Remove

10

(SCS

AR-WFPR-10);

to

severe

erosion

and

significant

repair

costs

at

CE-Grapevine,

TX,

CE-Saylorville,

IA,

and

SCS

AR-WFPR-3;

and

finally

a

complete

breach

at

SCS

MS-BC-53.

See

Cato

(1991)

for

a

detailed

discussion

of

the

erosion

damage

and

characteristics

of

the

spillway

channels

evaluated

to

develop

his

classification

system.

Based

on

the

performance

of

various

geologic

units

subjected

to

erosive

stresses,

the

spillway

incident

classification

was

used

to

define

a

spillway

erosion

potential

classification

system.

This

system

separates

material

into

four

classes:

10

Chapter

2

Analysis

and

Prediction

of

Spillway

Erosion

AAAA

=

Erosion-resistant

rock

AAA

=

Moderately

erosion-resistant

rock

AA

=

Moderately

erodible

material

A

=

Erodible

soil

Stable

conditions

for

each

class

are

given

in

the

form

of

slope,

maximum

flow

velocity,

expected

erosion,

and

effect

of

anomaly

on

erosion

(Figure

2).

Expected

erosion

is

based

upon

the

incident

classification

established

by

Cato

(1991).

Class

A

is

for

nonvegetated

soil

because

vegetation

can

offer

protection

up

to

7

ft/sec.

Anomalies

refer

to

breaks

in

surface

cover

produced

by

roads

or

lineations

in

the

vegetation

and

pilot

channels,

knickpoints,

or

other

abrupt

slope

changes.

CHANNEL

STABILITY

EROSION

RISK

CLASS

AAAA

AAA

AA

A

Slope

(percent)

30-45

15-30

4-15

<

4

Flow

Velocity

(ft/sec)

10-15

7-10

.

4-7

<

4

Anomaly

Effect

Minor

Moderate

Major

Severe

Figure

2.

Relationship

between

rock-erosion

classification

and

geometric

and

hydraulic

characteristics

of

the

spillway

channel.

For

example,

a

rock

classified

as

an

AAAA

is

predicted

to

be

able

to

withstand

flows

in

a

channel

of

from

10

to

15

ft/sec

Spillway

Channel

Erosion

Equation

The

to

assess

the

erosion

potential

of

geologic

materials,

both

descriptively

or

in

an

application,

inadequately

assesses

erodibility

behavior.

Multidisciplinary

studies

of

spillway

channel

erosion

have

determined

that

three

major

factors

control

erosion

(Cameron

et

al.

1986,

1988;

SCS

1987b).

These

factors

can

be

expressed

qualitatively

as:

E

=

f(CJi,G)

(D

where

E

=

Erosion

C

=

Channel

geometry

H

=

Hydraulics

G

=

Geology

1

A

table

of

factors

for

converting

non-SI

units

of

measurement

to

SI

units

can

be

found

on

p.

vii.

Chapter

2

Analysis

and

Prediction

of

Spillway

Erosion

11

The

precedence

and

usefulness

of

a

qualitative

equation

come

from

kindred

approaches

investigating

soil

formation

and

soil

erosion.

In

1941,

Jenny

(1941)

proposed

the

soil

forming

factor

equation,

today

popularly

known

as

the

"CLORPT"

formula:

SF=f(CL,OJi,P,T)

(2)

where

SF

=

Soil

formation

CL

=

Climate

O

=

Organics

R

=

Relief

P

=

Parent

material

Jenny

felt

that

quantification

of

these

factors

was

impossible,

but

explicit

identification

of

these

factors

in

an

equation

made

more

detailed

studies

possible.

The

Universal

Soil

Loss

Equation

also

exemplifies

usefulness

of

this

approach

(Smith

and

Wischmeier

1962).

In

this

case,

soil

loss

is

a

function

of:

SL

=

f(LS,P,CJi,K)

(3)

where

SL

=

Soil

loss

LS

=

Length

slope

P

=

Conservation

practices

C

=

Cropping

or

vegetation

type

R

=

Rainfall

K

=

Material

type

Both

of

these

equations

list

all

factors

affecting

the

stated

process

as

a

guide

for

other

workers,

but

are

not

a

"cookbook

approach"

because

resulting

accuracy

depends

totally

upon

the

user's

judgment.

This

concept

of

a

qualitative

factorial

equation

to

predict

emergency

spillway

erosion

is

the

basis

of

the

erosion

prediction

technique

developed

in

this

study.

Equation

1,

the

spillway

channel

erosion

equation

addresses,

channel

geometry

(C),

hydrology/hydraulics

(H),

and

geology

(G).

Channel

geometry

(C)

Channel

geometry,

C,

can

be

used

as

a

first

approximation

in

erosion

predic-

tion.

An

important

design

factor

is

the

vertical

drop

that

takes

place

along

the

spillway

channel

length;

the

drop

represents

potential

energy

to

erode.

Other

geometric

factors

include:

12

Chapter

2

Analysis

and

Prediction

of

Spillway

Erosion

a.

Channel

width.

b.

Excavated

channel

length.

c.

Excavated

channel

gradient.

d.

Length

of

steep

section.

e.

Steep

section

gradient.

/.

Channel

form

anomalies,

such

as

pilot

channels,

knickpoints,

and

flow

concentrators

Channel

geometry

factors

greatly

influence

flow

regime

and

erosion

processes

acting

in

the

spillway

channel

(Chow

1959).

Field

inspections

of

emergency

spillways

that

experienced

erosion

during

flow

events

revealed

that

the

initiation

of

erosion

was

often

associated

with

a

geomorphic

anomaly.

For

example,

the

existence

of

a

local

road

and

fill

within

the

spillway

channel

at

CE

Grapevine

Lake

acted

like

an

overtopping

dam

and

concentrated

erosive

energy

at

the

downstream

toe

of

the

fill.

This

resulted

in

the

development

of

deep

channel

"gully"

erosion.

Other

observed

anomalies

that

concentrate

flow

and

therefore

establish

the

point

of

initial

erosion

include

pilot

channels,

trees

and

clump

grasses,

and

abrupt

changes

in

channel

gradient,

such

as

at

the

interface

between

the

channel

and

the

original

floodplain.

Hydrology/hydraulics

(H)

Channel

geometry

combined

with

site

hydrology

establish

hydraulic

forces

operating

in

the

spillway

channel.

Hydraulic

factors,

H,

include:

a.

Instantaneous

peak

flow.

b.

Cumulative

flow.

c.

Flow

duration.

d.

Flow

depth.

e.

Flow

velocity.

The

hydrology

of

the

drainage

basin

above

the

spillway

combined

with

the

operational

procedures

at

the

particular

structure

are

the

controlling

factors

on

the

flow

conditions

in

the

spillway.

Postflow

field

inspections

of

eroded

spillway

channels

revealed

that

once

erosion

was

initiated

at

some

geometric

anomaly,

the

primary

erosion

mechanism

was

either

boundary

shear,

as

at

CE

Lewisville,

or

back

cutting

of

knickpoints,

as

at

CE

Grapevine

and

CE

Saylorville.

Chapter

2

Analysis

and

Prediction

of

Spillway

Erosion

13

The

controlling

hydraulic

factor

in

boundary

shear

erosion

appears

to

be

flow

velocity,

while

in

knickpoint

erosion

the

controlling

factor

is

a

combination

of

geometry

and

flow

depth.

May

(1988)

showed

in

fixed

bed

experiments

on

knick-

point

hydromechanics

that

the

greatest

knickpoint

migration

rates

may

not

corre-

spond

to

highest

velocity

or

discharge.

May

identified

two

significant

conditions:

(a)

the

relative

depth

of

flow

compared

to

the

height

of

the

fall

(Figure

3)

and

(b)

the

degree

of

hydraulic

venting

below

the

nappe

of

the

overfall

(Figure

4).

In

the

first

condition,

as

flow

depth

increased

compared

to

the

height

of

the

fall,

the

influence

of

the

geometric

anomaly

decreased

until

it

eventually

became

a

part

of

the

channel

roughness.

Unvented

knickpoints

accelerate

headcutting

by

orders

of

magnitude

greater

than

vented

knickpoints.

Analyses

of

flow

events

determined

that

unvented

conditions

commonly

occur

in

excavated

spillway

channels

during

conditions

of

below

design

peak

flows.

Robinson

(1988,

1989)

showed

that

low

tailwater

depths

allow

greater

stress

impacts

on

the

knickpoint

face

and

in

the

plunge

pool

than

do

high

tailwater

levels

and

that

the

rate

of

knickpoint

migration

can

be

slowed

tremendously

by

increasing

water

depth

and

submerging

the

overfall.

Geology

(G)

Geologic

factors,

G,

are

the

least

understood

of

the

three

major

variables;

are

independent

of

site

hydraulics

or

channel

geometry;

often

control

spillway

design;

and,

most

significantly,

govern

all

erosion

processes.

For

example,

a

spillway

excavated

into

massive

unfractured

granite

can

be

much

narrower

and

can

be

excavated

at

a

steeper

gradient

than

one

with

a

loess

foundation

because

of

the

greater

erosion

resistance

of

the

granite.

However,

predicting

erosion

resistance

of

natural

materials

is

seldom

this

clear-cut

because

a

great

number

of

geologic

variables

create

immense

numbers

of

situations

possible.

To

provide

analytical

structure,

a

geologic

erosion

equation

is

proposed

to

concisely

express

critical

geologic

variables

influencing

erosion.

The

equation

•

Y

1

I

--

\

36.

32.

.

\

'

\

21.

24.

z

>

20.

ie.

EROSION

ZONE

-~.

#7s/

ANGLE

OF

IMPACT

7

12.

a.

/*v=

8

/t

M

_]

L

C7

vo

4.

0

10

20

30

40

SO

60

70

eo

so

e

14

Figure

3.

Relationship

between

flow

depth

and

fall

height

(after

May

1988)

Chapter

2

Analysis

and

Prediction

of

Spillway

Erosion

HYDROGRAPH

TIME

>~

UNVENTED

EROSION

B

X

\

VENTED

EROSION

Figure

4.

Flood

hydrograph

(A)

of

a

flow

event

that

reaches

peak

design

discharge

showing

erosive

flows

during

vented

(cross-hatched),

unvented,

and

peak

flows.

Erosion

hydrograph

(B)

showing

the

conditions

when

vented

(hatched)

and

unvented

erosion

occurs.

Note

the

significant

increase

in

erosion

during

unvented

conditions

(after

May

1988)

reads:

where

GR

L

SP

G

ST

MP

GR

=

f(L,SP,G,STMP)

Geologic

resistance

to

hydraulic

erosion

Lithology

Rock

substance

properties

Genesis

of

the

material

Structure

and

tectonic

history

Rock

mass

properties

(4)

Some

lithologies

unceasingly

erode

while

others

have

never

been

observed

to

erode

within

a

human

time

frame.

Rock

substance

properties

define

whether

material

behaves

as

soil

or

rock,

that

is,

whether

it

erodes

in

a

grain-by-grain

fashion

or

detaches

along

discontinuities

as

blocks.

Defining

soil

versus

rock

with

accepted

definitions,

such

as

strength,

is

not

justifiable

because

strength

alone

does

not

appear

to

control

erodibility.

Genesis

of

the

material

determines

lateral

continuity,

bedding

thickness,

and

types

of

interbeds.

Structural

and

tectonic

Chapter

2

Analysis

and

Prediction

of

Spillway

Erosion

15

history

determine

rock

unit

orientation.

Unit

orientation

can

channalize

and

shift

erosion

toward

the

dam

if

units

strike

parallel

to

flow

or

can

significantly

reduce

erosion

if

units

dip

upstream

into

the

flow

direction.

Rock

discontinuities,

such

as

fractures,

function

as

detachment

surfaces

that

define

rock

blocks.

The

following

sections

describe

the

role

of

each

of

the

geologic

factors

and

define

an

erodibility

classification

for

each.

Lithology.

Different

erosion

processes

observed

for

various

rock

types

indicate

the

first

and

most

significant

erosion

criterion

should

be

lithology.

Lithology

is

the

basic

classification

of

natural

material

for

a

geologist

and

relates

to

material

genesis.

The

three

general

lithologic

types

are

igneous,

sedimentary,

and

metamor-

phic.

Spillway

flows

have

been

documented

on

two

of

the

three

lithologic

types,

but

by

far

the

greatest

frequency

of

flows

has

occurred

on

sedimentary

rock.

Observed

rock

types

include:

a.

Extrusive

igneous.

b.

Intrusive

igneous.

c.

Clastic

sedimentary

(sandstone

and

shale).

d.

Organic

sedimentary

(limestone).

Clastic

sedimentary

rocks

make

up

most

observed

sites;

this

is

expected

because

clastic

sedimentary

rocks

comprise

approximately

80

percent

of

the

U.S.

surface

area.

Metamorphic

rocks

and

chemical

sedimentary

rocks

were

not

evaluated.

Igneous

and

metamorphic

rocks

have

a

wide

range

in

their

resistance

to

erosion,

but

not

as

great

as

that

observed

for

sedimentary

rocks.

Relative

to

sedimentary

rocks,

igneous

and

metamorphic

rocks

tend

to

have

higher

strengths

and

densities

and

display

competent

mass

properties.

Igneous

and

metamorphic

rocks

display

block-by-block

detachment

processes

rather

than

single-grain

erosion

processes.

No

examples

exist

of

severe

threatening

spillway

erosion

in

these

types

of

lithologies.

Sedimentary

rocks

have

an

extensive

range

of

physical

properties,

and

most

examples

of

spillway

erosion

are

from

clastic

sedimentary

rocks.

This

is

caused

by

the

fact

that

most

examples

of

spillway

erosion

came

from

structures

built

by

the

SCS

and

their

projects

are

predominantly

located

on

sedimentary

bedrock.

Lithologies,

discharge,

and

effect

of

lithology

on

erosion

are

given

in

Cato

(1991).

Selected

case

histories

for

various

lithologies

are

presented

below:

a.

Sandstone.

A

resistant

sandstone

unit

provided

ample

resistance

to

the

1984

flow

at

CE-Saylorville,

IA.

The

ogee

weir

control

structure

is

founded

on

this

unit

as

well

as

is

the

exit

channel.

A

drop-off

occurs,

however,

16

Chapter

2

Analysis

and

Prediction

of

Spillway

Erosion

downstream

of

the

exit

channel

along

with

channel

narrowing

and

flow

concentratioa

Shales

with

some

interbedded

sandstones

and

siltstones

comprise

the

stratigraphy

of

the

steep

section.

The

flow

produced

a

series

of

overfalls

(stairsteps)

which

migrated

moderate

amounts

upstream.

The

sandstone

unit

itself,

comprising

the

exit

channel

floor,

was

quite

resistant.

Dislodgment

of

large

sandstone

blocks

did

occur

because

the

shale

founda-

tion

for

each

block

could

not

resist

the

flow

attack.

b.

Siltstone.

The

CE-Grapevine,

TX,

spillway

erosion

in

1981

was

the

biggest

factor

persuading

the

USACE

to

establish

its

rock

erosion

study

(Cameron

et

al.

1986).

Severe

gullying

and

downcutting

of

the

excavated

channel

threatened

site

integrity

and

were

produced

by

a

9,000

cfs

flow.

Initially,

it

was

thought

that

a

resistant

sandstone

unit

would

provide

adequate

erosion

resistance

to

prevent

downcutting

and

headward

gully

migration.

However,

the

sandstone

unit,

part

of

an

old

fluvial

and

deltaic

sequence,

was

laterally

discontinuous

and

underlain

by

highly

erosive,

interbedded

siltstones,

shales,

and

sandstones.

The

rock

substance

properties

of

the

siltstones,

low

plasticity

indices,

indicated

the

material

was

erodible.

c.

Shale.

SCS

WV-SF-17

shows

how

a

shale

can

be

very

resistant.

The

exit

channel

discharges

onto

a

41-percent

slope

which

drops

70

to

80

ft

down

to

the

floodplaia

Row

removed

waste

rock

placed

on

the

steep

slope

during

construction

and

uncovered

a

shale

bedding

plane.

Blocks

of

shale

were

removed

by

mass

wasting

processes

as

the

toe

of

the

slope

was

undercut;

however,

very

little

shale

was

removed

during

this

extreme

attack.

CE-

Lewisville,

TX,

experienced

spillway

flow

at

the

same

time

as

the

Grape-

vine

flow

(Cameron

et

al.

1986).

In

this

case,

attack

only

removed

a

surfi-

cial

veneer

of

weathered

and

desiccated

shale.

Unweathered

Eagleford

shale

was

quite

resistant

to

erosion.

d.

Cohesive

soil.

A

good

example

of

soil

material

resistance

is

a

January

1989

spillway

flow

in

Kentucky

at

an

SCS

spillway,

KY-UT-8.

A

gradient

change

at

the

downstream

end

of

the

excavated

emergency

spillway

channel

from

2.5

to

20

percent

produced

eroding

velocities.

Colluvium

blanketed

dipping

sandstones

and

shales

in

the

area

where

erosion

was

concentrated.

The

colluvium,

classified

as

a

soil

GM

by

the

Unified

Soil

Classification

System,

was

resistant

to

erosion

and

performed

almost

as

well

as

the

rock.

The

colluvium

that

did

erode

did

not

do

so

in

a

grain-by-grain

detachment

of

gravel,

sand,

and

silt

particles.

It

detached

along

soil

mass

discontinuities

producing

blocks

of

colluvial

material,

some

as

much

as

1

ft

in

diameter.

e.

Granular

soil.

Mississippi

Black

Creek,

MS-BC-53,

was

a

soil

spillway

that

breached

in

1983

(Temple

1989).

The

highly

erosive

loess,

sand,

and

gravel

offered

very

little

erosion

resistance

to

the

flow

once

a

gully

began

migrating

headward.

The

lithology

of

this

site

indicated

that

it

was

highly

susceptible

to

erosion.

The

sediment

was

very

poorly

cemented

and

had

,a

very

low

density.

It

is

thought

that

both

grain-by-grain

detachment

and

Chapter

2

Analysis

and

Prediction

of

Spillway

Erosion

17

block

detachment

processes

operated

during

this

flow

event.

The

loess

detached

along

soil

mass

discontinuities

and

because

of

its

low

density

was

able

to

be

transported

downstream

as

discrete

soil

particles.

The

erosion

classification

for

lithology

(Figure

5)

is

a

guide

to

expected

performance

range

and

is

not

used

to

place

material

into

an

erosion

class;

this

equips

the

user

with

an

expected

performance

range.

Determination

of

subsurface

lithology

requires

more

interpretation

and

hence

more

subjectivity

than

the

other

Geotechnical Aspects of Rock Erosion in Emergency Spillway Channels Supplemental Information on Prediction, Control, and Repair of Erosion in Emergency Spillway Channels

Recommended publications

Supplemental Information 2

Supplemental Information 8

Supplemental Information 4

Supplemental Information 3