A method for predicting local structure-induced sediment scour based on near bottom flow measurements

UFL/COEL-93/003

A METHOD FOR PREDICTING LOCAL
STRUCTURE-INDUCED SEDIMENT SCOUR BASED
ON NEAR BOTTOM FLOW MEASUREMENTS

by

Maximo Ramon Castillo Ramos III

Thesis

1993

A METHOD FOR PREDICTING
LOCAL STRUCTURE-INDUCED SEDIMENT SCOUR
BASED ON NEAR BOTTOM FLOW MEASUREMENTS

By

MAXIMO RAMON CASTILLO RAMOS III

A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE

UNIVERSITY OF FLORIDA

1993

I

ACKNOWLEDGEMENTS

I would like to express my sincere appreciation and gratitude to my advisor and

supervisory committee chairman, Prof. D. Max Sheppard, for his continuous support and

guidance throughout my study at the University of Florida. My thanks are also extended

to Dr. Alan Wm. Neidoroda and Dr. Ashish J. Mehta for serving as members on my

supervisory committee.

Many thanks go to Sidney Schofield, Jim Joiner, Chuck Broward, Vernon

Sparkman, Danny Brown and the other members of the Coastal and Oceanographic

Laboratory for their cooperation and assistance on this project. This work would not

have been accomplished without the assistance of Mike Dombrowski, Suresh, Prashant

and Kailash during the hydrodynamic and scour experiments.

I am also grateful to the Florida Department of Transportation for providing

financial support for this research and to the Department of Civil Engineering at the

University of Florida for allowing this research to be conducted in their Hydraulic

Research Flume.

Thanks go to my fellow Coastals; Mark, Mike Del, Phil, Phil and Eric. A big

thank you goes to my two best friends, Cherryl and Ica, for their love, patience and

support. Finally, I am most grateful to my parents who showed me how important and

enjoyable pursuing an education can be.

TABLE OF CONTENTS

ACKNOWLEDGEMENTS ................................. ii

LIST OF TABLES ................... .................... v

LIST OF FIGURES ...................................... vii

KEY TO SYMBOLS ...................................... xii

ABSTRACT ................... ....................... xiv

CHAPTER 1 INTRODUCTION ................................ 1

CHAPTER 2 LITERATURE REVIEW ......................... 5
2.1 Overview ..................................... 5
2.2 Scour Measurements Around Multiple Pile Structures .......... 6
2.3 Flow and Scour Depth Measurements Inside a Scour Hole ....... 8
2.4 Flow Measurements Using Constant Temperature Anemometers ..... 9
2.5 Sediment Transport Formulas ........................ 10

CHAPTER 3 PROBLEM STATEMENT AND METHODOLOGY ........ 12
3.1 Problem Statement ................................ 12
3.2 Solution M ethod ................................. 13
3.3 Selection of Parameters and Scale Modeling ................ 13
3.4 Experimental Plan and Test Facility .................. ... 17

CHAPTER 4 HYDRODYNAMIC STUDY ...................... 22
4.1 Background .................................... 22
4.2 Experimental Procedures ........ ........ ........... 27
4.3 Data Acquisition ................................. 32
4.3.1 X Sensor Probe ............................ 32
4.3.2 Temperature Probe .......................... 33
4.4 Data Reduction .................................. 38

CHAPTER 5 SCOUR STUDY .............................. 40
5.1 Background .................................... 40
5.2 Sediment ..................................... 41
5.3 Instrumentation .................................. 44
5.4 Test Procedures ................................. 45
5.5 Data Acquisition ................................. 48
5.5.1 Echo Sounder ............................. 48
5.5.2 Potentiometer Readings ...................... 49
5.6 Data Reduction ................................. 49
5.7 Time Rate of Scour .............................. 57

CHAPTER 6 ANALYSIS OF HYDRODYNAMIC AND SCOUR
RESULTS .................................. 63
6.1 Mathematical Model ............................. 63
6.2 Mathematical Model Results ......................... 66

CHAPTER 7 RESULTS AND CONCLUSIONS ................... 80
7.1 Results ...................................... 80
7.2 Conclusions .................................... 85

APPENDIX A HYDRODYNAMIC STUDY ...................... 89
A. 1 Instrumentation ................................. 90
A.2 Temperature Correction ............................ 91
A.3 Vortex Shedding from Probe Holder .................... 92
A.4 Preliminary Flow Measurements ...................... 93

APPENDIX B SCOUR STUDY .............................. 99
B. 1 Instrumentation ................................ 100
B.2 Echo Sounder Head Angle Calibration ................... 101
B.3 Sediment .................................... 105

APPENDIX C DETAILED RESULTS .......................... 107

REFERENCE LIST ...................................... 129

BIOGRAPHICAL SKETCH ................................. 131

LIST OF TABLES

2.1 Some formulas for bed load and total load .........

3.1 Reynolds numbers for hydrodynamic and scour studies .

3.2 Sequence of hydrodynamic tests ................

3.3 Sequence of scour tests .....................

A. 1 Constant temperature anemometer technical data ......

X sensor probe technical data ....................

Temperature probe technical data ..................

Variation of mean velocity with time at the same location ...

Variation of mean velocity across the tank in front of and away
influence of the structure .......................

X sensor probe cosine response probe vertical .........

X sensor probe cosine response probe horizontal ....... .

Echo sounder technical data .....................

Values of x and y at grid points of hydrodynamic study ....

Results of hydrodynamic Test H-1 .................

Results of hydrodynamic Test H-2 . . . . .

Results of hydrodynamic Test H-3 .................

Predicted and measured bottom elevations for scour Test S-1
(no pile cap structure) .........................

....... 90

....... 91

....... 97

from the
....... 97

....... 98

....... 98

. ..... 100

....... 108

....... 110

A.2

A.3

A.4

A.5

A.6

A.7

B.1

C.1

C.2

C.3

C.4

C.5

. . .. 114

....... 116

C.6 Predicted and measured bottom elevations for scour Test S-2
(70 pile cap structure top position) . . . . ... 122

C.7 Predicted and measured bottom elevations for scour Test S-3
(70 pile cap structure bottom position) . . . ..... 128

LIST OF FIGURES

3.1 Definition sketch of pile cap.............................. 20

4.1 Shear stress profiles in front of and away from the influence of the
structure ................... ..................... 25

4.2 Velocity profile corresponding to shear stress profiles in Figure 4.1 .... 25

4.3 Grid points for hydrodynamic study (1:1 scale) . . . ..... 26

4.4 Grid points for hydrodynamic study (distorted 1:4.5 scale) . ... 26

4.5 Placement of test structure prior to hydrodynamic experiment ........ 28

4.6 Experimental set-up for hydrodynamic study . . . ..... 28

4.7 Overall set-up of experimental apparatus and instrumentation for
hydrodynamic study . . . ..... . . ..... 29

4.8 Calibration curve for the temperature probe . . . ..... 30

4.9 Calibration curve for the X sensor probe . . . ... ....... 31

4.10 Unfiltered U and V velocity components for hydrodynamic Test H-1 at
grid point B2 ..................................... 34

4.11 Filtered U and V velocity components for hydrodynamic Test H-1 at
grid point B2 ..................................... 35

4.12 Power density spectra of U velocity components in
Figures 4.10 and 4.11 ................................ 36

4.13 Power density spectra of V velocity components in
Figures 4.10 and 4.11 ................................ 37

4.14 Mean velocity in front of structure vs. time . . . ..... 38

I

4.15 Turbulent energy/unit mass in front of structure vs. time . . 39

5.1 Overall set-up of experimental apparatus and instrumentation
for scour study .................................... 42

5.2 Experimental set-up for scour study . . . ... ........ 43

5.3 Test area of flume showing no pile cap structure and central region
near structure ...................................... 43

5.4 Calibration curve for the echo sounder output signal . . .. 45

5.5 Modified calibration curve for echo sounder angle. . . ... 46

5.6 Unfiltered output signals after completion of Test S-1 . . .... 50

5.7 Filtered output signals after completion of Test S-1 . . .... 51

5.8 Power density spectra of echo sounder output (Channel CHOO) in
Figures 5.6 and 5.7 ................................. 52

5.9 Power density spectra of y position potentiometer (Channel CH03) in
Figures 5.6 and 5.7 ........... ........................... 53

5.10 Definition sketch showing measurements needed for conversion of raw data
to x-y coordinates .................................. 55

5.11 Data points before start of scour Test S-1 (No pile cap structure) . 56

5.12 Data points after completion of scour Test S-1 (No pile cap structure) 56

5.13 Time rate of scour of piles 1,2 and 6 of no pile cap structure ........ 58

5.14 Time rate of scour of piles 1,2 and 6 of 70 pile cap
(top position) structure ................... ............. 59

5.15 Timerate of scour of pile 1 ............................ 60

5.16 Time rate of scour of pile 2 ............................ 61

5.17 Time rate of scour of pile 6 ............................ 62

6.1 Comparison of measured bottom elevation, predicted equilibrium bottom
elevation and dimensionless shear stress for the no pile cap structure
(row 2 : x = 10 cm) ................... .............. 67

6.2 Comparison of measured bottom elevation, predicted equilibrium bottom
elevation and dimensionless shear stress for the no pile cap structure
(row 3 x = 40 cm) ................................. 68

6.3 Comparison of measured bottom elevation, predicted equilibrium bottom
elevation and dimensionless shear stress for the no pile cap structure
(row 4 : x = 70 cm) ................................. 69

6.4 Comparison of measured bottom elevation, predicted equilibrium bottom
elevation and dimensionless shear stress for the no pile cap structure
(row 5 : x = 100 cm) ................... ............. 70

6.5 Comparison of measured bottom elevation, predicted equilibrium bottom
elevation and dimensionless shear stress for the no pile cap structure
(row 6 : x = 130 cm) ................... ............. 71

6.6 Comparison of measured bottom elevation, predicted equilibrium bottom
elevation and dimensionless shear stress for the no pile cap structure
(row 7 : x = 160 cm) ................... ............. 72

6.7 Comparison of measured bottom elevation, predicted equilibrium bottom
elevation and dimensionless shear stress for the 70 pile cap (top) structure
(row 2 : x = 10 cm) ................................. 73

6.8 Comparison of measured bottom elevation, estimated and predicted
equilibrium bottom elevations and dimensionless shear stress for the
700 pile cap (top) structure (row 3 : x = 40 cm) . . ..... 74

6.9 Comparison of measured bottom elevation, predicted equilibrium bottom
elevation and dimensionless shear stress for the 70 pile cap (top) structure
(row 4 : x = 70 cm) ................................. 75

6.10 Comparison of measured bottom elevation, predicted equilibrium bottom
elevation and dimensionless shear stress for the 700 pile cap (top) structure
(row 5 : x = 100 cm) ................... ............. 76

6.11 Comparison of measured bottom elevation, predicted equilibrium bottom
elevation and dimensionless shear stress for the 70 pile cap (top) structure
(row 6: x = 130 cm) ................... ............. 77

6.12 Comparison of measured bottom elevation, predicted equilibrium bottom
elevation and dimensionless shear stress for the 700 pile cap (top) structure
(row 7 : x = 160 cm) ............................... 78

6.13 Comparison of measured bottom elevation, predicted equilibrium bottom
elevation and dimensionless shear stress for the 700 pile cap (bottom)
structure (row 1 : x = 10 cm) ........................... 79

7.1 Contours of dimensionless shear stress, Td, around no pile cap structure .82

7.2 Contours of dimensionless shear stress, rd, around 700 pile cap
(top position) structure ................................ .. 82

7.3 Contours of measured bottom elevation around no pile cap structure after
completion of the scour test ............................ ..83

7.4 Contours of predicted equilibrium bottom elevation around no pile case
structure .............................. ........ 83

7.5 Contours of measured bottom elevation around 700 pile cap (top position)
structure after completion of the scour test .. ........... 84

7.6 Contours of predicted equilibrium bottom elevation around 70 pile cap
(top position) structure ........ ........................ 84

7.7 700 pile cap (bottom position) structure before the scour test ....... 87

7.8 700 pile cap (bottom position) structure after completion of the scour test 87

7.9 700 pile cap (top position) structure before the scour test ........... 88

7.10 700 pile cap (top position) structure after completion of the scour test ... 88

A. 1 Mean velocity vs number of data points sampled . ....... 95

A.2 Variation of mean velocity parallel to the 70 pile cap
(top position) structure .................. ............. 96

A.3 Variation of mean velocity normal to the 700 pile cap
(top position) structure ................................ 96

B. 1 Calibration curve for the echo sounder head angle . .. ...... 101

B.2 Profile of flat bottom using original calibration curve . ....... 103

B.3 Modified calibration curves of the echo sounder head angle ... ... 104

B.4 Profile of flat bottom using modified calibration curves ............ .104

B.5 Sediment size distribution for Category 1 sand (away from structure). .. 105

B.6 Sediment size distribution for Category 2 sand (anchoring layer) ..... 106

B.7 Sediment size distribution for Category 3 sand (around structure) .... 106

KEY TO SYMBOLS

a dimensionless constant relating -u'/w and u'2 + v'2

b dimensional empirical coefficient in de=f (r)

c dimensionless empirical coefficient in de=f (r)

de scour depth

dep predicted equilibrium scour depth

dee estimated equilibrium scour depth

dem measured scour depth after 28 hours

d5o median sediment diameter (0.60 mm)

d diameter of probe holder

D diameter (or width) of piles

f, vortex shedding frequency of probe holder

g acceleration of gravity

h water depth

k dimensionless constant relating u'2 + v'2 and u'2 + w2

St Strouhal Number; fvd/U

iT water temperature

u' fluctuating velocity component in x direction

U velocity component in x direction

U, value of U where motion of sediment with sediment diameter d5o is

initiated on a flat bottom away from the structure

v' fluctuating velocity component in y direction

V velocity component in y direction

w' fluctuating velocity component in z direction

W velocity component in z direction

x position along the tank (in line to flow)

y position across the tank (transverse to flow)

z distance from the bottom (positive upward)

0 angle of echo sounder head with vertical

0 sediment angle of repose (320)

p water density

Ps sediment density

rd dimensionless shear stress, ratio of r and ro

7 near-bottom shear stress

T, viscous near-bottom shear stress

rt turbulent near-bottom shear stress

To near-bottom shear stress away from influence of structure

Tvo viscous near-bottom shear stress away from influence of structure

7to turbulent near-bottom shear stress away from influence of structure

fi dynamic (absolute) viscosity

Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science

A METHOD FOR PREDICTING
LOCAL STRUCTURE-INDUCED SEDIMENT SCOUR
BASED ON NEAR BOTTOM FLOW MEASUREMENTS

By

Maximo Ramon Castillo Ramos II

May 1993

Chairperson: Dr. D. Max Sheppard
Major Department: Department of Coastal and Oceanographic Engineering

A method of predicting equilibrium scour depths around multiple pile structures

based on pre-scoured bottom shear stress was developed in this study. It was

hypothesized that a relationship exists between the pre-scoured bottom shear stress and

the equilibrium scour depth. A series of hydrodynamic tests were conducted in which

near-bottom flow measurements were made in the vicinity of a variety of multiple pile

structures. The distribution of bottom shear stress was estimated from these flow

measurements. Scour tests were then made in the same flume using the same structures.

A simple relationship between the equilibrium scour depth and the pre-scoured bottom

shear stress was formulated and the data from the two sets of experiments were used to

calibrate and test the formulation. The formulation gives reasonable predictions for the

range of conditions tested. The approach appears promising as an alternative way of

estimating equilibrium local scour depths for complex multiple pile structures. In

addition, a number of interesting and useful findings were made regarding the rate at

which a local scour hole forms near complex structures.

CHAPTER 1
INTRODUCTION

Most bridge piers are supported by piles. The friction between these piles and

the surrounding sediment is essential to the structural integrity of the bridge. If the

stream bed elevation drops such that more of the piles are exposed, the structure is

significantly weakened.

There are several ways in which the stream bed elevation can change. One

mechanism is aggradation or degradation of the stream bottom. This results from

changes in the flow conditions and would occur whether the bridge was present or not.

Although the aggradation/degradation of the stream is not caused by the presence of the

bridge, this must be considered during the design of the bridge and its piers. Another

cause of stream bed elevation change is contraction scour. If the cross-sectional area of

the flow is decreased, there is a corresponding increase in flow velocity. Depending on

the size of the bridge piers relative to the stream cross-sectional area, a significant

amount of contraction scour can be caused by the presence of the bridge. The third

mechanism is local scour. This is due to the individual elements of the bridge and is,

in general, the most important factor regarding stream bed elevation changes around a

pier (Shen et al., 1966). In addition to local scour, if the structure foundation is

composed of multiple piles, this may result in global or dishpan scour. Jones (1989)

found that when the piles are in close proximity to each other, a/D < 3, where a is the

L

2
distance between pile centers and D is the pile diameter, the characteristic width to be

used in the pier scour equation is the sum of the widths of the piles. For 3 < a/D <

11, both local and global scour are present, and for a/D > 11, the piles act as individual

structures (Hannah, 1978).

As mentioned above, local scour is considered to be the most important factor of

stream bed elevation change around a bridge pier. The main scour mechanism for steady

flow around blunt vertical structures is thought to be the horseshoe vortex. This vortex

is caused by a vertical gradient in the stagnation pressure along the leading edge of the

structure resulting from velocity gradients in the bottom boundary layer (see e.g.

Sheppard and Niedoroda, 1990). The presence of the bridge pier also tends to accelerate

the flow around the structure which leads to an increase in the bottom shear stress.

During the past few years, several bridges in the United States have failed due

to sediment scour. These accidents have resulted in fatalities and significant loss of

property. Among the more recent bridge failures, are the New York Thruway bridge

over Schoharie Creek in 1987 and a bridge on US 51 over the Hatchie River in

Tennessee in 1989. These cases served to raise the level of concern among engineers

responsible for the design and maintenance of bridges. Over the last few decades,

various researchers have developed methods for predicting scour depths around bridge

piers under extreme flow conditions.

All of the sediment scour mechanisms listed above are complex and difficult if

not impossible to treat analytically. Computational approaches using large, high speed

computers have not to date been successful in predicting sediment scour even for simple

structures. Consequently, researchers have resorted to empirical equations based

primarily on laboratory data. A large number of experiments have been performed with

simple structures such as circular and rectangular cylinders (see Shen et al., 1966).

Fewer tests have been made with multiple pile structures (Hannah, 1978 and Sheppard

and Niedoroda, 1992). There are many flow, sediment and structure variables important

to the scour process. A large number of experiments are needed to examine the effects

of each quantity on the scour process. Most laboratory tests have used prototype

sediment and water while scaling many. of the other parameters. This always raises

questions regarding scale effects when relating model results to prototype conditions.

These problems motivated the approach taken in this thesis, which was an attempt

to predict scour depths from near-bottom flow measurements over a fixed bed. The

bottom shear stresses were estimated from flow measurements taken in the region

surrounding the structure where local scour was anticipated. If the scour depths around

a structure can be determined from these flow measurements, it would reduce the need

for moveable bed scour tests. One of the problems with this approach, is that the flow

pattern around the structure can be significantly altered as the scour hole deepens. The

question then is whether there is sufficient information in the pre-scoured bottom shear

stress to predict equilibrium scour depths.

A similar approach has been used in fixed bed models for sediment transport in

coastal areas. A coastal structure like a jetty may need to be lengthened or have its

shape altered. A (usually distorted) fixed bed scale model of the area of interest is

constructed in a basin. Waves and/or currents are generated and the flow around the

various structure configurations being considered is measured. The sediment transport

throughout the region is then estimated using the measured velocities.

The objectives of this study were to 1) estimate the bottom shear stress in the

vicinity of a complex pier structure from near-bottom flow measurements and 2) estimate

the equilibrium local scour depth near the structure using the pre-scoured bottom shear

stress. The prediction scheme was calibrated using scour measurements from tests made

with the same multiple pile models. It was assumed that the relation between the bottom

shear stress and the equilibrium scour depths will hold if the structure configuration

exposed to the flow does not change significantly as the scour hole develops. The

structure configurations considered included a multiple pile pier without a pile cap and

the same structure with a pile cap at two different vertical locations. The flow conditions

(water depth and flow velocity) and mean sediment size were maintained constant

throughout both the hydrodynamic and scour tests.

CHAPTER 2
LITERATURE REVIEW

The following chapter is a literature review of several fields of study and

their approaches to solving flow-structure interaction problems. The review will cover:

1) scour experiments around multiple pile structures, 2) flow and scour depth

measurements around and inside the scour hole, 3) fluid measurements using constant

temperature anemometers and 4) sediment transport formulas. The approaches used in

these fields can be combined to develop a method of relating near bottom hydrodynamic

flow measurements to bottom shear stress. The shear stress measurements can then be

used to predict scour depths around piers with pile caps or footings.

2.1 Overview

Most bridge piers are designed with footings and piles and it is not clear how the

location and type of footing affects the scour depth around the pier. The presence of the

footing makes it difficult to apply the scour depth equations developed for simple shapes.

In order to determine the effect of the footing, it is usually necessary to conduct scale

model tests to determine the scour depths. The limitation of this method would be the

amount of time and expense needed to cover a large number of pier and footing

configurations. If a relation between equilibrium scour depths and prescoured bottom

shear stress can be developed, this would allow more pier and footing combinations to

be studied.

6
2.2 Scour Measurements Around Multiple Pile Structures

There have been numerous authors such as Shen (1966), Melville and Raudkivi

(1977) and Sheppard and Niedoroda (1992) who have conducted scour experiments

around simple structures such as circular and rectangular cylinders. As a result of these

experiments, a number of predictive equations have been developed. On the other hand,

very little scour data exists for multiple pile structures. A brief discussion of the

multiple pile structure papers reviewed as part of this study is given below.

Jones. Kilgore and Mistichelli (1992). Jones investigated the effect of the vertical

location of a footing on local scour. He conducted scour experiments on a 1:50 scale

model of a pier from the Baldwin Bridge in Connecticut. The experiments were

performed in a 21.3 m long and 1.8 m wide tilting flume. Since the flume was unable

to recirculate sediment, Jones was limited to clear water experiments with the flow

slightly below incipient motion for the sediment in the test area. The sediment used in

the tests had a d50 of 0.38 mm.

The model was a 9.7 cm wide pier with a 19.7 cm wide footing. The footing was

designed such that it could be moved up or down in the flow field. The footing was

modified to add an extension to the upstream face because Jones believed that without

the extension, the original footing would have little effect on the diving currents

associated with pier scour. The position of the footing ranged from slightly below the

bed to the full flow depth.

1 I

The duration of the tests were 4 hours. At 4 hours, the rate of change in scour

depth appeared to be negligible. Although other researchers have performed longer tests,

Jones pointed out that those tests were for larger grain sizes.

The results indicate that local pier scour was reduced if the footing was flush or

slightly below the bed. The scour hole depth and width increased as a larger percent of

the footing was exposed to the flow.

Jones. Stein and Kilgore (1990). These researchers investigated the effects of

varying the footing location relative to the stream bed. Clear water experiments were

conducted on a 1:50 model of a pier of the Acosta Bridge in Jacksonville, Florida. At

the start of the test, the flow was first set to that of incipient sediment motion. After

reaching this flow, a pier was placed into the sand and allowed to scour for 4 hours.

The footing of the model tested was 0.51 m long, 0.46 m wide and 0.09 m thick

and the pier was 0.46 m long and 0.17 m wide. The entire model was then moved up

or down relative to the bed such that the footer would be at different flow levels. The

results indicate that the scour depth is not greatly influenced by the footing location if the

footing is at or above the bed. The pier in this study did not extend all the way to the

bottom. Below the footing were piles that extended through the bed. It should be noted

that the footer could not be moved up or down relative to the model. This resulted in

different pile lengths for each test case and led the researchers to consider the results

inconclusive.

2.3 Flow and Scour Depth Measurements Inside a Scour Hole

It is difficult to relate hydrodynamic flow measurements before scour begins

around the structure to the actual scour depths because as the scour hole develops, the

flow changes. The flow in the equilibrium scour hole may not resemble that of the pre-

scour flow measurements. This section reviews experiments on cylinders which compare

hydrodynamic flow measurements in and around a scour hole and the resulting scour

depths.

Melville and Raudkivi (1977). Melville conducted clear water experiments on

a 5.08 cm diameter cylinder in a flume 19 m long, 0.46 wide and 0.44 deep. The

sediment used in these experiments had a d5o of 0.385 mm. The tests covered three

stages of scour (initial flat bed, intermediate scour hole after 30 minutes from start and

equilibrium scour hole). After each stage, a concrete (fixed bed) model was made of the

scour hole using plaster-of-paris. The sand in the live bed experiments was used to coat

the concrete model which was then placed in the tank. The researchers found the static

angle of repose of the sand under water to be 320.

The velocity and turbulence intensities in and around the scour hole were

measured with a constant temperature anemometer using a single component conical

probe. The mean bed shear stress was estimated using the mean velocity measurements

at 2 mm above the bed such that

K =KAu (2.1)
Ay

where

A = local mean velocity at 2 mm from the bed,

Ay = 2 mm and

K = 2 kg*mm/m2 (a calibration constant).

Melville found that the location of the maximum bed shear stress was where the

scour hole first developed. The results also show that the locations of large bed shear

stress correspond to regions of low turbulent intensity.

Carstens and Sharma (1975). The researchers found that the scour depths they

measured near a 11.5 cm diameter cylinder were similar to maps of shear stress

distribution of Hjorth (1974) around a 7.5 cm diameter cylinder. Hjorth conducted

experiments on cylinders and measured bed shear stresses on a fixed bed up to 12 times

higher than that of the undisturbed flow.

Carstens and Sharma noted that the scour hole covered a larger area than the zone

of high shear stress. Studies have also found the bed shear stress in a scour hole to be

75% of the theoretical critical value for the sediment. Raudkivi (1976) attributed this to

increased turbulence with strong intermittent eddies present.

2.4 Flow Measurements Using Constant Temperature Anemometers

Constant temperature anemometers measure fluid velocity by sensing the changes

in heat transfer from a sensor exposed to fluid motion. Constant temperature

anemometers have several features that make them suitable for turbulence studies

(Goldstein, 1983). These anemometers have a high frequency response which allows it

to measure the fluctuating components of the flow. Measurements up to several hundred

10
Khz are easily performed. Constant temperature anemometers are generally small in size

which allows point measurement. A hot-wire sensor is typically 5 /m in diameter and

about 2 mm long.

2.5 Sediment Transport Formulas

Sediment can be transported as bed load or suspended load near the bed. Bed

load transport occurs when sediment next to the bed moves by rolling and sliding.

During this time, the sediment remains close to the bed. Another method of transport,

called saltation, is when the sediment performs jumps. The sediment is temporarily

carried in the flow and then returns back to the bed. Total load refers to the sum of the

bed load and the part of the suspended load that can also be found in the bed (Sleath,

1984). A list of some of the sediment transport formulas are shown in Table 2.1

Table 2.1 Some formulas for bed load and total load (Sleath, 1984).

Investigator Formula Comments

Du Boys (1879) Qs = Aro(ro-r) Bed Load
O'Brien and Qs = A(ro-rc)" Bed Load
Rindlaub (1934)
Shields (1936) Qs = AQS(ro-Tr)/ Bed Load
((p,-p)gD)
Brown (1950) Qs = AD32(r,/ Bed Load
(p,-p)gD)3
Garde and Qs = Q(D/d)f(D) Total Load
Albertson (1958) (u.d/y)__
Chang et al. (1967) Qs = AU(ro-rT) Total Load

where

Q, = bed load transport rate,

Tr = critical shear stress on the bed for initial motion,

To = bed shear stress,

ii = shear velocity, (r/p)1/2 and

A = constant or dependent on sediment characteristics.

Table 2.1 lists just a few of the sediment transport formulas that are based on

bottom shear stresses. Note that the powers that the shear stress are raised to vary from

0.33 to 3.0.

CHAPTER 3
PROBLEM STATEMENT AND METHODOLOGY

3.1 Problem Statement

Structure-induced sediment scour has been studied by various researchers

over the past few decades. Most experiments have been conducted on simple structures

such as circular and rectangular cylinders and several predictive equations have been

developed based on these experiments. Structures with multiple piles such as bridge

piers and offshore oil platforms are also of interest. The flow near these structures is

more complex and the scour more difficult to predict than the single element structure.

The offshore industry divides the scour that occurs near structures into two categories,

local and dishpan (or global). In this case, local refers to the scour adjacent to each

piling and dishpan to the depression around the entire structure. Hydraulics engineers

working with bridge pier scour combine the two and call it local scour.

Bridge piers come in many shapes and sizes. Some bridge piers are designed

with a footing or pile cap. The presence of the pile cap usually complicates the flow

field due to the enhanced turbulence and the size and complexity of the wake region. In

general, the equations developed for simple structures cannot be used for these multiple

pile structures. Thus, there is a need for innovative ways of dealing with this problem.

3.2 Solution Method

Due to the limited data for multiple pile structures, the scour prediction equations

that exist are by necessity conservative to allow for uncertainties in both the data and the

equations. Properly conducted scour tests are expensive and time consuming. The time

required to reach equilibrium scour depths increases with the size and complexity of the

structure. The results of this study clearly show that portions of the scour hole are far

away from equilibrium after 28 hours of steady flow in the flume. This provided

motivation for the approach taken in this thesis.

It was hypothesized that the equilibrium scour depths near a structure are related

to the bottom shear stresses on the pre-scoured bed. To test this hypothesis, near-bottom

flow measurements (from which bottom shear stresses could be estimated) were made

near complex multiple pile structures. Sediment scour tests were then made in the same

flume with the same structures and flow conditions. A simple relationship between

equilibrium scour depth and bottom shear stress was formulated and data from the two

sets of experiments were used to calibrate and test the formulation.

3.3 Selection of Parameters and Scale Modeling

The phenomenon of local structure-induced sediment scour depends on several

flow and structure parameters (Sheppard and Niedoroda, 1990). These include the aspect

ratio ( h I D ), the sediment regime number ( U/Uc 1), the Froude Number

( U / vg'h ) and to a lesser extent, the Reynolds Number ( UDp / I' ), where

h = water depth,

14
D = diameter (or width) of pile,

U = depth (and time) average flow velocity,

U, = value of U where motion of sediment with diameter d50 is initiated

on a flat bottom away from the structure,

p = mass density of water,

1 = dynamic viscosity of water and

g = acceleration of gravity.

Similitude modeling in open channel flow experiments can be performed using

Froude or Reynolds scale modeling. If Froude scale modeling is used to achieve

similarity between the model and prototype, the ratio of the inertia and gravity forces for

the model should be equal to the ratio for the prototype, i.e.

( )= ( I), (3.1)

where F denotes force, i inertia, g gravity, m model and p prototype. Substituting the

expression for these forces in terms of the physical properties, length scales and flow

kinematics results in

p( V2 L2 L3 P g2 L2 (3.2)
pg L' i pg L'

15

= ( 2 (3.3)
g L [gL ,

or

(Vm Vpi)2 =1. (3.4)
(8. p,) (L.L,)

Note that

8. = gp, (3.5)

thus

V2 L
(3.6)
V2 L
p p

or

Y_ LP (3.7)
V \

Let

L. (3.8)

then Equation 3.7 becomes

-VP" (3.9)
V

On the other hand, if Reynolds scale modeling is used, the ratio of inertia to

viscous forces for the model should be equal to the ratio for the prototype, i.e.

F, (3.10)

where F denotes force, i inertia, v cinematic viscosity, m model and p prototype.

Substituting the expressions for these forces in terms of the physical properties, length

scales and flow kinematics results in

p V2 L2 p V2 L2 (3.11)
p v VL pvVL ),

In these experiments, water of similar mass density and viscosity to that of the prototype

are used. Therefore

p, = pp (3.12)

and

VX = Vp (

(3.13)

17
Substituting Equations 3.12 and 3.13 into Equation 3.11 results in

(VL)m =(VL)p (3.14)

or

v. L,

Combining Equations 3.15 and 3.8 gives

V_ 1 (3.16)

By comparing Equations 3.9 and 3.16, it is evident that both Froude and Reynolds

scaling cannot be achieved for the problem under consideration without considerable

effort and expense. Both the flow and sediment transport are less dependent on the

Reynolds Number in the range of the flow parameters of interest in these tests (see Table

3.1) than the Froude Number. Table 3.1 lists the Reynolds Numbers based on the pile

cap widths and the individual pile widths for the hydrodynamic and scour studies. Based

on this, Froude scale modeling was used for the hydrodynamic and scour experiments.

3.4 Experimental Plan and Test Facility

Two sets of laboratory experiments were conducted on a 1:15 scale model of a

bridge pier. The first set of experiments consisted of hydrodynamic tests on a bridge

pier with and without a pile cap. Table 3.2 lists the hydrodynamic tests that were

Table 3.1 Reynolds numbers for hydrodynamic and scour studies.

Hydrodynamic Study Scour Study
prototype model prototype model
Water depth (m) 5.33 0.36 5.85 0.39
Design Velocity 1.12 0.29 1.22 0.31
(m/sec)
Re (based on pile 1.02 x 107 1.75 x 105 1.11 x 107 1.87 x 105
cap width)
Re (based on pile 0.98 x 106 1.69 x 104 1.06 x 106 1.80 x 104
width)

conducted. The type and location of the pile cap relative to the bottom was varied and

near-bottom flow measurements were made over a portion of the area where scour was

anticipated. A pile cap with 70 o sides was used in these experiments (see Figure 3.1).

Experiments were performed without the pile cap and with the pile cap at two different

positions relative to the bottom. The no pile cap case represents the situation where the

pile cap is positioned above the water surface. For the second position, the top of the

12.7 cm thick pile cap was 1.3 cm above the water surface. In the third position, the

bottom of the pile cap was resting on the flume bottom. The second set of experiments

consisted of sediment scour tests with the same multiple pile structures. A list of the

scour tests that were conducted is given in Table 3.3. The hydrodynamic and scour tests

had a flow skew angle of 00, i.e. the flow was in line with the pier. During the sediment

scour tests, an echo sounder was used to map the bottom in the vicinity of the pier

19
Table 3.2 Sequence of hydrodynamic tests.

Test Number Pile Cap Type Pile Cap
Location
H-1 No Pile Cap NA
H-2 700 Pile Cap Top
H-3 700 Pile Cap Bottom

Table 3.3 Sequence of scour tests.

Test Number Pile Cap Type Pile Cap
Location
S-1 No Pile Cap NA
S-2 700 Pile Cap Top
S-3 700 Pile Cap Bottom

before, during and after the 28 hour duration tests. Mechanical (point gauge)

measurements were also made at the end of each test.

These experiments were conducted in the Hydraulics Research Flume in the Civil

Engineering Department at the University of Florida. This recirculating flume has a flat,

fixed bottom and the main flume is 30 m long, 2.44 m wide and 0.75 m deep. The flow

is generated by a 100 hp pump with a maximum discharge of 1100 liters/sec. The flow

rate and depth are controlled by a V-notch weir at the entrance and a tail gate at the

downstream end of the flume. The water returns to the pump in a return channel 34 m

15.2cm
5.1cm x5.1cm Piles -.
1 2 3 4 \5 6 7 8 9 10 11 12

PLAN VIEW 11.3

X^_______________

Pile Cap 8
Rectangular 900
Sloping 700
15.2cm

c
cm

13.7cm

/

-< 190.0cm
SIDE VIEW

END VIEW
END VIEW

Figure 3.1 Definition sketch of pile cap.

Flow c

B
52.7cm
A

E 3 E E 0 a B m a

a a a a a 0 0n a []a a

/

|

\

21
long, 1.22 m wide and 1.00 m deep. Since the pump was not designed to recirculate

sediment, a "sediment trap" was built prior to the sediment scour tests and placed at the

downstream end to prevent sediment removed from the test area from reaching the pump.

The flume also has a 6.10 m long, 2.44 m wide and 0.33 m deep recess located

approximately 13 m downstream from the entrance. During the hydrodynamic flow

measurements, the recess was covered with metal plates. To obtain a uniform roughness,

a layer of sand was glued to the plates. The plates were removed and the recess filled

with sand for the scour tests.

CHAPTER 4
HYDRODYNAMIC STUDY

4.1 Background

As stated in the introduction, the primary objective of this study was to determine

if equilibrium local scour depths can be predicted from near-bottom flow measurements

over the pre-scoured bed near a multiple pile structure. The hydrodynamic study with

the fixed bed is discussed in this chapter and the scour study is discussed in Chapter 5.

Sediment transport is a complex process but it is generally believed that the

movement of sediment is related to the shear stress at the bed. For this reason, an

attempt was made to estimate the bottom shear stress from flow measurements at a fixed

distance above the bed (3 mm). In a fully developed open channel flow, estimation of

the bottom shear stress is relatively straight forward. This is not the case for the

complex "wake region" near a multiple pile structure and certain assumptions regarding

the stress distributions had to be made. To further complicate the situation, the mean

flow direction near the structure was found to vary with time as a result of the vortices

being alternately shed by the piles and pile cap.

A two component Dantec 56C01 constant temperature anemometer was used for

the flow measurements. The response of the Dantec R61 X sensor probe used was such

that the high frequency fluctuating components of the velocity could be measured. It was

23

determined that when the X sensor probe was mounted such that the filaments lie in

parallel vertical planes, variations in the mean velocity direction in a horizontal plane

could not be resolved. That is, when the probe is orientated to measure velocities in the

x (horizontal) and z (vertical) directions, changes in the direction of the horizontal

velocity were not detected. This is to be expected since the probe is not designed for this

flow situation. This means that the turbulent component of the shear stress, pu'v',

could not be accurately measured near the structure where the horizontal component of

velocity undergoes fluctuations in direction as large as 150. Ordinarily, the horizontal

and vertical velocities would be measured and the mean horizontal velocity, U, and time

average of the product u'w' computed in order to determine the total shear stress

aU (
t = puv' (4.1)

To avoid the problem of the probe not being able to resolve variations in flow

direction, it was decided to rotate the probe 90 so that the filaments were in parallel,

horizontal planes. In this position, the probe measures the two horizontal components

of velocity, U and V. It was assumed that the turbulent shear stress is proportional to

p(u'2 +v'2). The total shear stress would then be

U + ap(u'+ v'2) (4.2)

I

24

where "a" is a constant to be determined from measurements in front of the structure at

x = -150.0 cm and z = 3 mm. The analysis resulted in a value of 0.10 for "a".

The value of "a" in front of the structure at different heights (z = 23.5, 105.5

and 187.5 mm) above the bottom ranged from 0.08 to 0.12. The value of "a" next to

the structure ranged from 0.04 to 0.09. The lower values of "a" next to the structure

were probably due to the inability of the X sensor probe to resolve variations in the mean

flow direction. Based on the above, a value of 0.10 was used throughout the flow field.

Figure 4.1 compares the shear stress profiles in front of and away from the

influence of the structure computed using Equations 4.1 and 4.2. The shear stress

profiles are similar and at z = 3 mm both give the same value for r (0.22 N/m2). The

velocity profile corresponding to the shear stress profiles in Figure 4.1 is shown in

Figure 4.2. A dimensionless constant k, was also computed (k = 1.08) to relate the

turbulent energy obtained by using the probe in a horizontal position versus the two

component turbulent energy using the probe in a vertical position such that

u'12 + v12
k = (4.3)
12 + /2

A grid was laid out over a portion of the area expected to scour. The region

directly in front of the structure was excluded due to the complexity of the flow direction

resulting from horseshoe vortices. The grid is shown in Figures 4.3 and 4.4. In Figure

4.3, the x and y scales are equal. The y scale has been expanded 4.5 times in Figure

4.4. This distorted scale is used to present both hydrodynamic and scour results.

I

S30-
C -

m:

20

a '

.
0

10-

U I I, i.T. . -
0.0 0.1 0.2 0.3 0.4 C
Shear Stress (N/m**2)

Figure 4.1 Shear stress profiles
the structure at x =

10-

0

0.1 ., .o .

in front of and
-150.0 cm and

away from the influence of
y = 11.1 cm.

Velocity U (cm/sec)

Velocity profile corresponding to shear stress profiles in Figure 4.1.

" Probe horizontal
*** Probe vertical

1
1\

" !

Figure 4.2

-)
FLOW

60.71 2 3 4 5 7 E
D
C
B

I------------------------------------------------------I
1.1Figure 4.3 Grid points for hydrodynamic study (1:1 scale). I
-150.0
OUTER EDGE OF PUIS

Figure 4.3 Grid points for hydrodynamic study (1:1 scale).

3 4

5 6

E

FLOW
Flolw

-D

C

B

A
II--------------II -- I -- I -- II -- I -- 11

II---------------I -- II -- I -- II -- I -- c

I I---------------I: -- "" 1 1 -- 1 1 -- 1 -- ||

1 1----------------I -- I I -- I I -- I --- I -- II

x (am)

--------------------1.

PILE CAP

OUTER EDGE OF PILES

Figure 4.4 Grid points for hydrodynamic study (distorted 1:4.5 scale).

60.7

^\

R

11.1
-1

50.0

27
4.2 Experimental Procedures

Prior to conducting the actual tests, the instrumentation had to be set up and

calibrated and the structure centered in the test area as shown in Figure 4.5. Figure 4.6

shows the experimental set-up for the hydrodynamic study. Preliminary tests were then

conducted in the flume to determine the optimum location to place the model. It was

determined that the flow over the V-notch weir had a jetting effect such that the velocity

across the tank was not uniform as the flow approached the test area, so a fine mesh

screen was constructed and placed directly downstream of the weir. This significantly

decreased the jetting effect. A floating wooden lattice was also placed at the entrance

of the flume to dampen the waves generated by the weir and a plastic wire mesh wave

absorber placed at the downstream tailgate minimized wave reflection. The anemometer

measurements were sensitive to water temperature variations so a temperature probe was

placed near the anemometer probe as shown in Figure 4.7. The calibration curve for the

temperature probe is shown in Figure 4.8. The hydrodynamic tests were conducted at

water temperatures ranging from 25 to 29 OC.

The mean flow direction at each grid point had to be determined for each

structure configuration. A vane was used to establish the mean flow direction at each

grid point. Knowing the mean flow direction, the X sensor probe could be directed in

that mean direction to minimize the probe skew angle.

The hydrodynamic tests were conducted at night (starting around midnight) when

the air temperature was on the decline, to minimize the change in water temperature

during the test. Even with the decline in air temperature, the water temperature increased

Figure 4.5 Placement of test structure prior to hydrodynamic experiment.

Experimental set-up for hydrodynamic study.

Figure 4.6

Flume Bottom

Overall set-up of experimental apparatus and
instrumentation for hydrodynamic study.

Figure 4.7

40.0

S35.0

;~25.0
E-

20.0 . .
5.50 5.75 6.00 6.25 6.50
Voltage Reading (volts)

Figure 4.8 Calibration curve for the temperature probe
T (OC) = 445.89 +102.22V 4.05V2.

at a rate of approximately 0.3 OC/hour due to the pump and recirculation. Prior to

starting the experiment, the anemometer probe was towed at several velocities through

the still water in the flume to obtain a calibration curve for the velocity. A typical

velocity versus probe output voltage curve for the X sensor probe is shown in Figure

4.9. After calibration, the pump was started and the flow allowed to stabilize for

approximately 45 minutes prior to starting the flow measurements.

The probe was first moved in front of and away from the influence of the

structure and the velocity measured at depths ranging from 3 to 187 mm above the

bottom. This was used to obtain the undisturbed velocity profile (Figure 4.2). The

carriage was then moved along the structure and measurements were taken at the

100.0

90.0

80.0

70.0

60.0
o
50.0

q 40.0

30.0

20.0

10.0

0.0
0.0 0.8 1.6 2.4 3.2 4.0 4.8
Voltage Reading (volts)

Figure 4.9 Calibration curve for the X sensor probe
Vel = 1.86 0.33V + 2.05V2 1.01V3 + 0.15V4.

specified grid points. The grid points are in rows and columns with the rows having

constant x and the columns having constant y values. Measurements were made at all

grid points in one column prior to moving to the next column, starting with points in

front of the structure. Because the temperature in the flume increased at a rate of

approximately 0.3 OC/hour, the velocity measurements in front of the structure were used

to determine variations in the flow measurements due to changes in the velocity

calibration. After completing the measurements at the grid points, the velocities directly

behind the structure were measured.

32
4.3 Data Acquisition

A Metrabyte 16F computer board and ACQWIRE data acquisition software were

used to digitize and record the analog signals from the X sensor and temperature probes.

Scales were placed along and across the flume and the readings were recorded manually.

The vertical position traversing mechanism was controlled by a lap top computer. In

addition, a cut off switch attached to the probe support was used for the lowest (3 mm)

position (see Figure 4.7). This switch was used to insure that the bottom velocity

measurements were made at the same fixed distance above the bed at all locations.

4.3.1 X Sensor Probe

The X sensor probe has 2 perpendicular wires in parallel planes that sense the

normal components of the flow in the plane of the wires. The output voltages that are

proportional to the velocity components are processed by the ACQWIRE data acquisition

software to obtain U (velocity along the tank) and V (velocity across the tank) based on

a velocity calibration curve such as that shown in Figure 4.9. Flow and temperature

measurements were sampled at a rate of 50 hertz. Due to the change in the average

temperature of the water in the flume, the X sensor probe had to be calibrated prior to

each set of tests. The two velocity components, U and V can be calculated from the

sum and difference of the hot film anemometer output voltages, Volt1 and Volt2, as

follows:

U (Volt+Volt, ) (4.4)
V2

and

V = ( Volt-Volt ). (4.5)

A typical unfiltered signal is shown in Figure 4.10 and the same signal after

filtering is shown in Figure 4.11. Plotted in Figures 4.12 and 4.13 are the power density

spectra of U and V for the time series shown in Figures 4.10 and 4.11. The power

density spectra of the unfiltered U signal shows a large spike of energy at 10 hertz. This

was caused by vortex shedding from the probe holder. Although the noise at 10 hertz

for the V signal was not as large as for the U signal, a bandstop, fourth-order

Butterworth filter at 10 hertz was applied to both signals.

4.3.2 Temperature Probe

The temperature probe was attached downstream of the X sensor probe to

measure the instantaneous temperature fluctuations. Ideally, the temperature probe would

be located at the X sensor probe but since the temperature probe was rather large

(approximately 4.76 mm in diameter and 101.6 mm long), it was placed far enough

downstream so as not to disturb the flow at the X sensor probe. It was located

approximately 30.5 cm downstream of the X sensor probe. These water temperature

measurements were used in the ACQWIRE data acquisition software to apply a

correction to the velocity data after measurement. This correction formula is valid only

for small temperature changes and is discussed in Appendix A.

U (velocity along tank)

2 3 4 5 6 7 I 9 10 11 12 13 14
[sc]
chOOa

V (velocity across tank)

2 a 4 5 6 7 8 9 10 11 12 13 14
chOl]
ch01o

Figure 4.10

Unfiltered U and V velocity components for hydrodynamic
Test H-1 at grid point B2.

35

U (velocity along tank)

[sec]
fchOOo
V (velocity across tank)

2 3 4 5 6

[sec]
fch01o

Figure 4.11

Filtered U and V velocity components for hydrodynamic
Test H-1 at grid point B2.

Power Density Spectrum of Unfiltered U (velocity along tank)

2.4

2.2

2

1.8

1.6

14

1.2

1

0.8

0.5

0.4

0.2

0

2.4

2.2

2

1.8

1.6

1 4

1.2

0.8*

0.6

0.4

0.2

[Hz]
pmfchOOo

Figure 4.12 Power density spectra of U velocity components in

Figures 4.10 and 4.11.

[Hz]
pmch00a

Power Density Spectrum of Filtered U (velocity along tank)

Power Density Spectrum of Unfiltered V (velocity across tank)

[Hz]
pmch01o

Power Density Spectrum of Filtered V (velocity across tank)

12
[Hz]
pmfchOlo

Figure 4.13

Power density spectra of V velocity components in
Figures 4.10 and 4.11.

38

4.4 Data Reduction

In order to compensate for the change in water temperature during the course of

the test, curves of the mean velocity and turbulent energy in front of the structure versus

time were constructed as shown in Figures 4.14 and 4.15. Since the flow velocity did

not vary significantly across the tank, the mean velocity and turbulent energy curves

could be applied to the entire test area. This allowed the mean velocity and turbulent

energy at each grid point to be non-dimensionalized by the mean velocity and turbulent

energy in front of the structure.

25.0

20.0

15.0

8 10.0

5.0

0.0

Figure 4.14

1 2 3 4 5 6 7
Time (hours)

Mean velocity in front of structure vs. time
at x = 150.0 cm and z = 3 mm (test H-1)
Vel = 21.551 0.863T 0.150T2.

*1

,25.0

U

*20.0
U
1 20.0

S15.0

|10.0 -

0.0
0o 1 2 3 4 5 6 7
Time (hours)

Figure 4.15 Turbulent energy/unit mass in front of structure vs. time
at x = 150 .0 cm and z = 3 mm (Test H-1)
Turb. Energy/Unit Mass = 22.402 1.461T 0.350T2.

CHAPTER 5
SCOUR STUDY

5.1 Background

Sediment scour experiments were performed in the same flume and with the same

model piers as the hydrodynamic study. The objectives of the scour study were: 1) to

provide scour data for calibrating and testing the scour prediction relationship developed

as part of the hydrodynamic study and 2) to provide laboratory scour data for multiple

pile structures with and without pile caps. Another important aspect of this study was

the measurement of the rates of scour.

Instrumentation with the ability to make in situ scour experiments (i.e.

measurements of the scour depth as it occurs) were developed. In the past, scour

measurements have been very laborious and time consuming. Recent developments in

underwater acoustics technology now allow techniques, formerly used only for bottom

profiling at prototype depths- to be used in the laboratory. The main component of the

system used was the Simrad Mesotech 810 echo sounder.

The measurement system that utilizes the Mesotech echo sounder was developed,

constructed and tested in the Coastal Engineering Laboratory at the University of Florida.

An acoustic pulse sent by the echo sounder reflects off the bottom and the return pulse

is sensed by the echo sounder. The time required for the signal to travel to and from the

bottom is measured and the distance computed. Returns from particles in suspension also

40

41

show up in the signal and must be dealt with as noise. A schematic drawing of the

system is shown in Figure 5.1. The bowl which sits just below the surface of the water

when measurements are being made serves to minimize the disturbance to the flow (see

Figure 5.2).

To allow the echo sounder to send acoustic signals to the bottom, a small window

was cut out of the bowl and a vinyl sheet glued over the slot. The acoustic reflection

from the vinyl sheet was minimal. The level of water in the bowl was such that the

sounder head was always submerged.

The echo sounder was mounted on a frame as shown in Figure 5.1 and a motor

was used to rotate the echo sounder and thus the beam towards and away from the

structure. The entire depth measuring system was mounted on a carriage that traversed

the width of the flume. A small wheel in a track attached to a potentiometer was used

to measure the position across the flume and a similar arrangement was used to measure

the position of the carriage along the flume. Scales placed along and across the flume

were used to calibrate the position potentiometers. The start and ending points for each

traverse was recorded in the log book. A data reduction program was then used to

transform the output from the echo sounder signal and the position potentiometers to xyz

coordinates.

5.2 Sediment

The metal plates covering the recess in the flume were removed and the multiple

pile structure was centered in the test area. Sand was then placed around the structure

as shown in Figure 5.3. Three different sediment sizes were used in the flume. The

Traverse Across The Tank

Flume Bottom

Overall set-up of experimental apparatus and
instrumentation for scour study.

Figure 5.1

Figure 5.2 Experimental set-up for scour study.

Test area of flume showing no pile cap structure
and central region near structure.

Figure 5.3

~

-~C~II~~F~F~;1~1Cna"~ ----~-.`, ~4r
--I "1

~amPa*AI J

ri8~

44
largest quantity of sand (Category 1) was placed in the regions of the test area where no

scour was anticipated. This category had sediment sizes ranging from 0.2 to 0.8 mm.

Category 2 consisted of sand that was used to anchor the sand in the first category and

had a range of sediment sizes from 0.84 to 2.00 mm. This minimized the amount of fine

sand transported down the flume and through the pump. The third category of sand was

placed in the region around the structure where scour was anticipated. This category had

sediment sizes ranging from 0.42 to 0.84 mm and a median grain size, d5o, of 0.60 mm.

5.3 Instrumentation

Before the scour experiments were started, the instrumentation had to be set up

and calibrated. The acoustic bed profiling system is based upon the use of an echo

sounder operating at 2.25 Mhz. The 2.25 Mhz operating frequency was chosen to give

reasonable accuracies at short ranges, as well as to ensure sufficient energy in the

reflection from the bed. A 10 /sec pulse is emitted at a 100 Hz rate. The returned

pulse is an analog time/amplitude history of the water column. The travel time of the

signal is proportional to the distance to a reflecting object. The amplitude of the pulse

at any point in time is a measure of the number, size and density of reflecting particles

in the water column.The acoustic signal from the echo sounder was calibrated by

positioning a flat metal plate at different elevations below the sounder and the output

recorded. A calibration curve for the echo sounder output is given in Figure 5.4.

The angle the echo sounder head makes with the vertical was measured using a

ten turn precision potentiometer. A protractor mounted on the side of the unit was used

45

40.0

35.0

0
0
o

A 30.0

25.0

20.0
20.0 i i . .
0.0 0.5 1.0 1.5 2.0
Voltage Reading (volts)

Figure 5.4 Calibration curve for the echo sounder output signal
Distance (cm) = 0.112 + 23.385V.

in the calibration process. The head was rotated to various angles and the output voltage

from the potentiometer recorded. Because of some play in the gears moving the sounder

head, a lag between the actual angle and the angle measured was observed. To

compensate for this, a lag was added or subtracted from the calibration curve depending

on the direction the sounder head was swinging. A plot of the modified calibration curve

used is shown in Figure 5.5 along with the actual readings during the calibration process.

More details on this subject are given in Appendix B.

5.4 Test Procedures

Prior to each scour test, the sand was compacted with a tamper, smoothed and

leveled. After the bottom was leveled, the flume was slowly filled with water to prevent

scour during the fill process. Before starting the flow, the bottom was mapped with the

40.0
-0 -7.0 dog to 0.0 deg
**** 0.0 deg to 32.0 deg
----- 32.0 deg to 0.0 deg
0.0 deg to -7.0 deg
*** Calibration Points
30.0

o 20.0

%. %
0.0 '\
\

-10.0-
0.0 1.0 2.0 3.0 4.0 5.0
Voltage Reading (volts)

Figure 5.5 Modified calibration curve for echo sounder angle.

echo sounder. This allowed the scour measurements to be compared with the original

bottom. After the pump was started, minor flow adjustments had to be made during the

first 30 minutes of the test to obtain the desired velocities. Mapping of the bottom began

a few minutes after the pump was started and this continued throughout the duration of

the test (- 28 hours). The interval between the scans increased as the rate of scour

decreased (from 10 minutes at the start to 90 minutes at the end). After the flow

was stopped, the bottom was mapped once more. The water was then drained from the

flume and test area and point gauge measurements of the bottom were made.

A local maximum (if not the absolute maximum) in structure-induced scour occurs

near the transition from clearwater to live bed conditions. For this reason, the scour

(and hydrodynamic) tests were performed just below transition conditions. This was also

47
done to minimize the amount of suspended sediment transported through the pump. As

a result, the amount of sand in suspension was small except immediately adjacent to the

piles.

After the scour tests on the multiple pile structures were completed, a decision

to conduct two additional experiments was made. These two test were for a 4 inch

cylinder under the same sediment, water depth and flow conditions (U = 0.31 m/sec)

and at an increased flow rate (U = 0.37 m/sec). The purpose of these tests were to

determine if the conditions of the multiple pile tests were close enough to transition (clear

water to live bed) that maximum scour was occurring. The first cylinder test (conducted

at U = 0.31 m/sec) produced and ultimate scour depth ratio (de/D) of 1.68. The second

cylinder test (conducted at U = 0.37 m/sec) produced an ultimate scour depth ratio of

1.91, 14% larger than that of the first test. The dimensionless scour ratio of 1.91

matches the value obtained by Hannah (1978) for single circular cylinder. This indicates

that the flow rates for the pier tests were a bit low. Sand ripples formed away from the

structure and sand was transported down the flume for every test conducted indicating

a live bed condition, at least for a portion of the sand grains. However, more ripples

formed and more sand was transported during the second test with the cylinder were

closer to transition for the d50 (0.60 mm) sediment. To determine if the 14% increase

in scour depths found for the single cylinder could be applied to the multiple pile

structure, an additional test was made on the structure with no pile cap at the higher flow

rate. As with the cylinder, the scour depth increased approximately 14%. There was

some contraction scour during this test making these measurements not as precise as for

48

the other tests. Thus, it seems appropriate to apply a scour depth correction to the

multiple pile test conducted at the lower rate.

Section 5.7 shows that the scour depths are not close to equilibrium except for the

first few rows of the structure. At 28 hours, the scour depths at these front rows are

approaching equilibrium. Ettema (1976) found that the scour depths at 24 hours are 90%

of the equilibrium scour depths Because of this the scour depth correction and the

equilibrium correction is only applied to the front of the structure such that

1.14
de = de. (5. 1)
0.9

where

dee = estimated equilibrium scour depth and

dem = measured scour depth after 28 hours.

5.5 Data Acquisition

A Data Translation DT2801 computer board and GLOBAL LAB data acquisition

software were used to digitize and record the analog signals from the echo sounder output

and the potentiometer readings at a sampling frequency of 40 hertz.

5.5.1 Echo Sounder

Noise in the acoustic data (Channel CHOO) came from reflections from the pile

cap and/or piles and suspended sediment. Since the main mechanism for sediment

movement was bed load transport, the noise caused by suspended sediment was minimal.

The output from the echo sounder was filtered using a lowpass, sixth-order Butterworth

49
filter with a cutoff frequency of 3 hertz. A plot of the echo sounder output signal before

filtering is shown in Figure 5.6 Figure 5.7 shows the same signal after filtering and

power spectra for both signals are shown in Figure 5.8.

5.5.2 Potentiometer Readings

Channel CHO1 recorded the output from the potentiometer measuring the position

along the flume (x direction) while Channel CH03 was used for the position of the

carriage across the flume (y direction). Both output signals were filtered using lowpass,

sixth-order Butterworth filters with a cutoff frequency of 0.5 hertz. The noise in

Channel CH03 was caused by sudden movements of the manually operated traverse

across the flume. The carriage was moved along the flume with a variable speed motor

at a slow speed (approximately 2 cm/sec). The smooth motion along the flume

minimized the noise in Channel CH01. Figure 5.9 compares the power density spectra

of Channel CH03 before and after filtering. As mentioned before, the start and end

points of the x and y traverses were recorded in the log book.

The echo sounder angle signal was recorded on Channel CH02 and was filtered

using a lowpass, second-order Butterworth filter with a cutoff frequency of 20 hertz.

The sounder head was designed to swing from -90 to 350 but only data converted in the

range from -70 to 320 were used because of reflection and attenuation of the echo

sounder signal at the ends of the swing.

5.6 Data Reduction

A data reduction program was written to convert sediment scour data obtained

from the echo sounder and position potentiometers to bottom coordinates in rectangular

Echo Sounder Output

2

1

0 O.I 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 1.2
[min]
CHOO

Position Along Tank
5-

0.1 0.2 0.3 0.4 0.5 0.8
[min]
CHOl

0.7 0.8 0.8 I 1.1 12

Echo Sounder Angle

0.1 0.2 0.3 0.4 0.5 0.6
[min]
CHO2

0.7 0.8 0.9

Position Across Tank
0

-1

-2

-3

-4

-,
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 1.2

[min]
CH03

Figure 5.6

Unfiltered output signals after completion of Test S-1.

1.1 1.2

i

Echo Sounder Output

4

-- 3

0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0 0.9 1 1.1 1.2
[min]
fch00

Position Along Tank

4

_3

2

0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.0 0.9 1 1.1 1.2
[min]
fchOl

Echo Sounder Angle
5

4

-3

2

0
0 0.1 0.2 0.3 0. 0. 0.0 0.7 O0 0.9 1 1.1 1.2
[min]
fch02

Position Across Tank
0

-1

-2
-4

0 0.1 0.2 0.3 0.4 0.6 0.6 0.7 0.8 0. 1 1.1 1.2
[min]
fch03

Figure 5.7 Filtered output signals after completion of Test S-1.

Power Density Spectrum of Unfiltered Sonar Output

[Hz]
pmchOO

Power Density Spectrum of Filtered Sonar Output

0.5 1 1.5 2 2.5
[Hz]
pmfchOO

3 3.5 4 4.5 5

Figure 5.8 Power density spectra of echo sounder output (Channel CHOO)

in Figures 5.6 and 5.7.

0.26

0 24

0.22

0.2

0.18

0.16

a0.14
^-O.

0.26

024

0.22

0.2

0.18

0 16

0 14

0 12

0.1

0.01

006

0.04

0.02

0

Power Density Spectrum of Unfiltered Position Across Tank

0 0.1 0.2 0.3 0.4 0.5 0.6 07 0.8 09
[Hz]
pmch03

Power Density Spectrum of Filtered Position Across Tank

0 0.1 02 0.3 0.4 0.S
[Hz]
pmfch03

0.6 0.7 0.8 0.9

Figure 5.9 Power density spectra of y position potentiometer (Channel CH03)
in Figures 5.6 and 5.7.

54
coordinates (x,y and z). The raw data consisted of 1) voltage proportional to the distance

from the echo sounder head to the bottom, 2) voltage proportional to the position of the

depth measuring system along the tank (x position), 3) voltage proportional to the angle

of inclination of the sounder head (measured from a vertical line with the angle varying

from -70 to 320) and 4) voltage proportional to the position of the depth measuring

system across the tank (y position). A detailed explanation of the data reduction process

is given in Appendix B. Figure 5.10 is a definition sketch of the measurements needed

to convert the raw data to x-y coordinates. The equations used are

z = (R + R) cos O zoand

d = (R + R) sin 9,

where

R = distance from echo sounder face to bottom,

d = radial distance from echo sounder to bottom measurement,

zo = vertical distance from echo sounder pivot to original bottom,

Ro = distance from echo sounder pivot to its face and

0 = angle of echo sounder with vertical axis,

(-70 to +320).

After the data reduction process, the data were gridded using the surface fit

program, SURFER, which was used to interpolate and map the bottom. Figure 5.11

shows the data points used to map out the bottom prior to the test while Figure 5.12 is

Echo
Sounder

z Scoured
t4x Bottom

The origin of the right hand coordinate system is located at the
original bottom on the front outer corner of pile C1.
Positive x is in the direction of flow, Positive z is upward and
Positive y is normal to the flow.

Figure 5.10 Definition sketch showing measurements needed for
conversion of raw data to x-y coordinates

100.0

75.0

50.0

25.0

0.0

-25.0 OUTER EDGE OF PLES

-50.0
n1w
-75.0

-100.0 ,- ,
-100.0 -50.0 0.0 50.0 100.0 150.0 200.0 250.0
x (cm)

Figure 5.11 Data points before start of scour Test S-1
(No pile cap structure).

100.0

75.0 I

50.0

25.0

0.0 -

-25.0 OUTER EDGE OF PIUS

-50.0

-75.0

-100.0 .... .... .... .... .... .... ....
-100.0 -50.0 0.0 50.0 100.0 150.0 200.0 250.0
x (cm)

Figure 5.12 Data points after completion of scour Test S-1
(No pile cap structure).

57
a plot of the data points after the test was completed. The number of data points used

to map the bottom was approximately 9,000.

5.7 Time Rate of Scour

Since these experiments were conducted under clear water scour conditions, the

scour hole development could be observed visually. A soon as the flow was started,

sediment along the structure was put into motion. Due to the clear water scour

conditions, there was very little sediment being transported into the control volume

surrounding the front row of piles, thus the rate of erosion in this region is large.

Further downstream in the structure, there was initially (and for some time) more

sediment being transported into the region than was being scoured, thus accretion

occurred. Accretion continued until the transport into the region fell below the scour

rate. The sediment transport processes were more rapid at first but decreased with time

making it very difficult to judge the "maturity" of the scour hole strictly by observation.

Plots of the scour depths versus time at different locations along the structure

were, however, very informative (see Figures 5.13 and 5.14). From these figures, it is

evident that the local scour depths at rows 1 and 2 were approaching equilibrium after

28 hours. On the other hand, a net scour at the piles in row 6 was just beginning 25

hours into the flow. Figures 5.14 to 5.17 compare the time rate of scour of piles 1, 2

and 6 for the no pile cap structure and the 700 pile cap (top position) structure.

58

25
**** No Pile Cap Pile 1
No Pile Cap Pile 2
No Pile Cap Pile 6
20

o
15

u 10-
0

5-

r o 7

-5-

0 5 10 15 20 25 30
Time (hours)

Figure 5.13 Time rate of scour of piles 1,2 and 6
of no pile cap structure.

15-

5-
. 10- / ^. ,,'-
0 I

-10 7-li----II -II II--I------I I l l i-

0 5 10 15 20 25
Time (hours)

Figure 5.14 Time rate of scour of piles 1,2 and 6
of 70' pile cap (top position) structure.

Time (hours)

Figure 5.15

Time rate of scour of pile 1.

Time (hours)

Figure 5.16 Time rate of scour of pile 2.

15
Time (hours)

Figure 5.17 Time rate of scour of pile 6.

CHAPTER 6
ANALYSIS OF HYDRODYNAMIC AND SCOUR RESULTS

6.1 Mathematical Model

Preliminary tests indicated that the X sensor probe had to be rotated 900 (such that

the filaments were in parallel, horizontal planes) in order for the probe to resolve

horizontal variations in the flow direction (see Chapter 4). In this position, the probe

measured the two horizontal components of velocity, U and V. It was assumed that the

turbulent shear stress is proportional to p(u'2 +v'2). The total shear stress would then

be

r = + ap(u/+ v'2) (6.1)

where "a" is a dimensionless constant to be determined from measurements in front of

(and not influenced by) the structure. The analysis resulted in a value of 0.10 for "a".

The dimensionless shear stress, rd, was then computed at each grid point where

flow measurements were made. The dimensionless shear stress is the ratio of the shear

stress at any point, r, and the shear stress of the undisturbed flow in front of the

structure, ro, at z = 3 mm. The surface fit program SURFER, was used to interpolate

and map the dimesionless shear stress over the study area. SURFER was also used to

fit the scour data. The grid laid out over a portion of the area expected to scour is

63

64
shown in Figure 4.4. With the existing probe support, hydrodynamic measurements

closer than 11.1 cm to the piles could not be made due to the pile cap. This made it

necessary to extrapolate the shear stress from the closest grid point to the piles.

In attempting to obtain a functional relationship between the equilibrium scour

depth and the shear stress, the following guidelines were used. The relationship should

be as simple as possible and be directed at the specific problem at hand. That is, it

should be directed at equilibrium scour depths at the transition from clear water to live

bed conditions thus eliminating the need to introduce sediment properties, etc. The

thought being that this relationship could be extended to include a broader range of

conditions at a later date. In light of the above, the following form was selected,

de =f () (6.2)

where

de = equilibrium scour depth,

TO ( rd 1(6.3)

d (6.4)
0

and b and c are empirical coefficients.

65

Since the increase in bottom shear stress caused by the different multiple pile

structures was the only parameter changing during the hydrodynamic experiments, the

coefficients b and c were determined by a least squares best fit between the predicted and

estimated equilibrium scour depths at a section at the front of the pier. The estimated

equilibrium bottom elevation is the measured bottom elevation at 28 hours extrapolated

to equilibrium and adjusted for the velocity being below the transition between live bed

and clear water conditions. The coefficients are considered constant for the range of

conditions considered here. For a more general formulation, the coefficients will

undoubtedly be dependent on some of the quantities held constant during these

experiments. Various sediment transport equations have been developed using the critical

shear stress, rc and/or bed shear stress, r,. The shear stresses in these equations were

raised to powers varying from 0.33 to 3 (see Table 2.1).

The first step in evaluating the coefficients was to assign a value for c. Using the

dimensionless shear stress, rd, and the measured scour depth at grid point A3, a value

for b was computed. Grid point A3 ( see Figure 4.4) was selected because the time rate

of scour curves in Figures 5.8 and 5.9 indicate that this area is close to its equilibrium

scour depth. Grid point A2 was not chosen because rd is lower there than at A3. The

reason being that point A2 is far enough forward that the flow at that point is less

influenced by the structure. The values obtained for b and c are 4.62 cm and 0.5

respectively.

SURFER was then used to create a finer grid of the estimated shear stresses and

the corresponding scour depths at these grid points were computed using Equation 6.3.

L

66
One constraint on the computed equilibrium scour depths was that the scour depth

between two adjacent grid points could not change more than that allowed by the

sediment angle of repose, 320 (Melville and Raudkivi, 1977). The predicted equilibrium

scour depth is the larger between the scour depth based on the dimensionless shear stress

and the scour depth which satisfies the sediment angle of repose.

6.2 Mathematical Model Results

To simplify the presentation of the results, bottom elevations instead of predicted

and measured scour depths are presented. Bottom elevation, z, and scour depth, de, are

related by z = -de. Figures 6.1 to 6.12 show Td, the measured bottom elevation after

a 28 hour scour test, the estimated and predicted equilibrium bottom elevations for the

no pile cap and the 700 pile cap (top position) structures. The results indicate that the

predicted equilibrium bottom elevations are closer to the measured bottom elevations

towards the front of the structure. Figure 6.13 shows the results at row 3 (x = 40 cm)

for the 70 pile cap (bottom position) structure. The predicted equilibrium bottom

elevation is actually less than that measured. A possible explanation for this is that the

flow-structure configuration which was used for the hydrodynamic measurements

significantly changed as the scour hole developed. As the scour hole developed, a new

turbulence generating mechanism (the piles) came into effect. That is, the computed

bottom elevation in Figure 6.13 is for a structure with a pile cap that extends down

through the bottom.

- 30.0

0 25.0

S20.0

E 15.0
.4.1
0
0
m 10.0

3 5.0

. 0.0

-5.0

-10.0

. -15.0

E -20.0

-25.0

y (cm)

Comparison of measured bottom elevation, predicted equilibrium
bottom elevation and dimensionless shear stress for the
no pile cap structure (row 2 : x = 10 cm).

Figure 6.1

30.0

25.0

20.0

15.0

10.0

5.0

0.0

-5.0

-10.0

-15.0

-20.0

-25.0

measured bottom elevation
**** predicted equilibrium bottom elevation
-**-* dimensionless shear stress

~* -
-~

-_____^ ^ ^ ^ ^ ^ ___

S I I I l I I I I I I I 1 1 1 1 1 1 1 1 I I I I 1
0 10 20 30 40 50 60 7
y (cm)

Figure 6.2 Comparison of measured bottom elevation, predicted equilibrium
bottom elevation and dimensionless shear stress for the
no pile cap structure (row 3 : x = 40 cm).

S30.0

g 25.0

J 20.0

S 15.0
0

M 10.0

a 5.0

.r 0.0

-5.0

g -10.0

-15.0

g -20.0

-25.0

Figure 6.3

T

S measured bottom elevation
predicted equilibrium bottom elevation
-* dimensionless shear stress

--

0 10 20 30 40 50 60
y (cm)

Comparison of measured bottom elevation, predicted equilibrium
bottom elevation and dimensionless shear stress for the
no pile cap structure (row 4 : x = 70 cm).

" 30.0

25.0

| 20.0

S15.0
0

m 10.0

a 5.0

-b 0.0

-5.0

M -10.0

. -15.0

S-20.0
Qm

, I I I I iI i
10 20 30 40
y (cm)

50 60

Comparison of measured bottom elevation, predicted equilibrium
bottom elevation and dimensionless shear stress for the
no pile cap structure (row 5 : x = 100 cm).

-- measured bottom elevation
predicted equilibrium bottom elevation
*-** dimensionless shear stress

.2"
*" -----c--
-^ ^ __ ~ - ___

-25.0

-I.

Figure 6.4

-

-- measured bottom elevation
Predicted equilibrium bottom elevation
*-* dimensionless shear stress

^" tr --. ** _- .
- ___ ^ ---- --

" 30.0

S25.0

a 20.0

E 15.0
0
M 10.0

5.0

S0.0

-5.0

-10.0

. -15.0

S-20.0

-25.0

Figure

Comparison of measured bottom elevation, predicted equilibrium
bottom elevation and dimensionless shear stress for the
no pile cap structure (row 6 : x = 130 cm).

I I I I I I I I i I i i I I [ I I I I I i I I I
0 10 20 30 40 50 60
y (cm)

6.5

-

72

i 30.0

25.0- measured bottom elevation
computed equilibrium bottom elevation
*-**-. dimensionless shear stress
e 20.0

g 15.0
0o
0
m 10.0

S 5.0

-5.0

g -10.0 -

. -15.0

. -20.0

25.0 i i1 1 i ,,
0 10 20 30 40 50 60 70
y (cm)

Figure 6.6 Comparison of measured bottom elevation, predicted equilibrium
bottom elevation and dimensionless shear stress for the
no pile cap structure (row 7 : x = 160 cm).

S40
y (cm)

Comparison of measured bottom elevation, predicted equilibrium
bottom elevation and dimensionless shear stress for the
700 pile cap structure (row 2 : x = 10 cm).

' 30.0

S25.0

v 20.0

g 15.0
0
a 10.0

0 5.0
a
4 o 0.0

-5.0

a -10.0
a
. -15.0

. -20.0

-25.0

Figure 6.7

30.0

0 25.0

S20.0

g 15.0
0
m 10.0

5.0
m
0.0-

-5.0

S-10.0

0 -15.0

0 -20.0

-25.0 -
0

Figure 6.8

y (cm)

Comparison of measured bottom elevation, estimated and predicted
equilibrium bottom elevations and dimensionless shear stress for
the 700 pile cap (top) structure (row 3 : x = 40 cm).

" 30.0
o

0 25.0 --- measured bottom elevation
3 -predicted equilibrium bottom elevation
-c** dimensionless shear stress
S20.0

g 15.0
0
m 10.0

S 5.0

S 0.0

-5.0

-10.0

.0 -15.0

| -20.0

-25.0
0 10 20 30 40 50 60 70
y (cm)

Figure 6.9 Comparison of measured bottom elevation, predicted equilibrium
bottom elevation and dimensionless shear stress for the
700 pile cap (top) structure (row 4 : x = 70 cm).

- 30.0

S25.0

S20.0

g 15.0
0
0
m 10.0

a 5.0

. 0.0

-5.0

S-10.0
a
1. -15.0

. -20.0

-25.0

y (cm)

Figure 6.10

Comparison of measured bottom elevation, predicted equilibrium
bottom elevation and dimensionless shear stress for the
700 pile cap (top) structure (row 5 : x = 100 cm).

T

" 30.0

g 25.0

, 20.0

S15.0
o
0
m 10.0

a 5.0

.~ 0.0

-5.0

M -10.0

.o -15.0

e -20.0

-25.0

Figure 6.11

Comparison of measured bottom elevation, predicted equilibrium
bottom elevation and dimensionless shear stress for the
700 pile cap (top) structure (row 6 : x = 130 cm).

S measured bottom elevation
: predicted equilibrium bottom elevation
*-**- dimensionless shear stress

-4 -- --4...

0 10 20 30 40 50 60
y (cm)

-

30.0

25.0

20.0

15.0

10.0

5.0

0.0

-5.0

-10.0

-15.0

-20.0

-25.0

Figure 6.12

measured bottom elevation
predicted equilibrium bottom elevation
*-*-* dimensionless shear stress

10 20 30 40 50 60 70
y (cm)

Comparison of measured bottom elevation, predicted equilibrium
bottom elevation and dimensionless shear stress for the
700 pile cap (top) structure (row 7 : x = 160 cm).

1

-

- 30.0

S25.0

M 20.0

8 15.0

m 10.0

- 5.0

So.o
~-

-5.0

M -10.0

-15.0-

. -20.0

-25.0
0

Figure 6.13

y (cm)

Comparison of measured bottom elevation, predicted equilibrium
bottom elevation and dimensionless shear stress for the
700 pile cap bottom (row 3 : x = 40 cm).

CHAPTER 7
RESULTS AND CONCLUSIONS

7.1 Results

The primary objective of this study was to determine if equilibrium local scour

depths can be predicted from near-bottom flow measurements over the fixed bed near a

multiple pile structure. The first step was to measure flow parameters over a portion of

the area where scour was anticipated. These measurements were then used to estimate

the bottom shear stress over this area. A simple relationship between equilibrium local

scour and pre-scoured bottom shear stress was postulated. The coefficients in the

formulation were evaluated using measured scour depths extrapolated to equilibrium. In

its present form, this relationship is restricted to predicting equilibrium scour depths at

the transition from clear water to live bed conditions. Future plans include extending this

formula to a broader range of conditions.

The scour depths predicted by this formula seem reasonable and are consistent

with extrapolations of scour time series plots. Both the predicted scour depths and the

time history plots for various pile rows indicate that the time to reach equilibrium

increases dramatically as the distance from the front of the structure increases. The rate

of scour around the piles on the leading edge appears to be close to that for a single pile

structure, i.e. after 24 hours the scour depth is approximately 90% of the ultimate

81
(equilibrium) depth. Further downstream, the scour appears to be much less mature at

the end of 24 hours and the rates seem to be decreasing.

Figures 5.8 and 5.9 show that piles 1 and 2 of the no pile cap and 700 pile cap

(top position) structures approached equilibrium local scour depths at the end of the scour

tests (i.e. 28 hours after the start). On the other hand, pile 6, which was located

approximately halfway down the structure, was just beginning to scour below the original

bottom after 25 hours.

If the relationship developed here (or a similar expression) proves to be valid for

a wide range of conditions, the time and cost savings in evaluating design scour depths

for complex bridge piers could be significant.

Figures 7.1 and 7.2 are contours of dimensionless shear stress around the no pile

cap and 700 pile cap (top position) structures. Figures 7.3 and 7.5 are the measured

bottom elevations over the same area. Figures 7.4 and 7.6 are contours of the computed

equilibrium bottom elevations (of the area where hydrodynamic flow measurements were

made) based on the dimensionless shear stresses in Figures 7.1 and 7.2. These figures

show that the predicted equilibrium bottom elevations are deeper than the measured

values throughout the area but the difference is greater toward the back of the structure.

This is one indication that the scour is much closer to equilibrium near the front of the

structure.

The predicted equilibrium bottom elevations do not match the measured bottom

elevations for the cases where the pile cap was initially resting on the bottom. This

55.0

50.0

45.0 -

40.0

35.0 -

S30.0

25.0

20.0

15.0

10.0
10.0

Figure 7.1

30.0 50.0 70.0 90.0
x (cm)

110.0 130.0 150.0

Contours of dimensionless shear stress, rd, around

no pile cap structure.

55.0

50.0

45.0

40.0

35.0

30.0

10.0
10.0 30.0 50.0 70.0 90.0 110.0 130.0 150.0
x (cm)

Figure 7.2 Contours of dimensionless shear stress, Td, around

700 pile cap (top position) structure.

50.0
55.0

45.0

40.0

35.0

S30.0

25.0

20.0

15.0

10.0
10.0 30.0 50.0 70.0 90.0 110.0 130.0 150.0
x (cm)

Figure 7.3 Contours of measured bottom elevation around no pile cap

structure after completion of the scour test.

45.0

40.0

30.0 -4.0

25.0

20.0 .2-.

10.0
10.0 30.0 50.0 70.0 90.0 110.0 130.0 150.0
x (cm)

Figure 7.4 Contours of predicted equilibrium bottom elevation around

no pile cap structure.

55.0

50.0

45.0

40.0

35.0

> 30.0

25.0

20.0

15.0

10.0
10.0

Uu.u
x (cm)

Contours of measured bottom elevation around 700 pile cap (top position)

structure after completion of the scour test.

55.0

50.0

45.0

40.0

S35.0

30.0

25.0

20.0

15.0

10.0 -
10.0

x (cm)

Figure 7.6 Contours of predicted equilibrium bottom elevation around

700 pile cap (top position) structure.

Figure 7.5

MILLISECOND	CLASS.METHOD	MESSAGE
0	sobekcm_page_globals.constructor
0	sobekcm_page_globals.constructor	Application State validated or built
0	sobekcm_database.verify_item_lookup_object
0	sobekcm_page_globals.constructor	Navigation Object created from URI query string
0	sobekcm_database.verify_item_lookup_object
0	sobekcm_page_globals.display_item	Retrieving item or group information
0	sobekcm_page_globals.get_entire_collection_hierarchy	Retrieving hierarchy information
0	sobekcm_assistant.get_entire_collection_hierarchy
0	cached_data_manager.retrieve_item_aggregation
0	cached_data_manager.retrieve_item_aggregation	Found item aggregation on local cache
0	item_aggregation_builder.get_item_aggregation	Found 'all' item aggregation in cache
0	system.web.ui.page.page_load (ufdc.page_load)
0	sobekcm_page_globals.constructor.on_page_load
0	html_echo_mainwriter.add_style_references	Adding style references to HTML
0	html_echo_mainwriter.add_text_to_page	Reading the text from the file and echoing back to the output stream
39	html_echo_mainwriter.add_text_to_page	Finished reading and writing the file