Citation
Wave generated bedforms in the near-shore sand environment

Material Information

Title:
Wave generated bedforms in the near-shore sand environment
Series Title:
UFLCOEL
Creator:
Jettâe, Christopher D., 1969- ( Dissertant )
Hanes, Daniel M. ( Thesis advisor )
Dean, Robert G. ( Reviewer )
Mossa, Joann ( Reviewer )
Sheppard, Donald M. ( Reviewer )
Thicke, Robert J. ( Reviewer )
University of Florida -- Coastal and Oceanographic Engineering Dept
Place of Publication:
Gainesville Fla
Publisher:
Coastal & Oceanographic Engineering Dept., University of Florida
Publication Date:
Copyright Date:
1997
Language:
English
Physical Description:
xii, 132 leaves : ill. ; 29 cm.

Subjects

Subjects / Keywords:
Sand waves ( lcsh )
Ocean bottom ( lcsh )
Marine sediments ( lcsh )
Coastal and Oceanographic Engineering thesis, Ph. D ( local )
Dissertations, Academic -- Coastal and Oceanographic Engineering -- UF ( local )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Abstract:
To measure bedform geometry, a high-resolution sea-bed profiling system has been developed. The multiple transducer array (MTA) consists of an array of ultra-sonic (5MHz) transducers. The first prototype MTA described herein measures two-dimensional bedform profiles over a length of 45 cm. Under ideal conditions, the instrument is capable of 1 mm vertical resolution and 2 cm horizontal resolution. Laboratory and field tests of the MTA show the system’s capability to accurately measure bedforms of known geometry and the ability to work under conditions with relatively high suspended sediment concentrations. Previous high-resolution profiling systems have either had moving parts, or have been unable to obtain the resolution of the system described herein. A multiple transducer array (MTA) was used to measure seabed geometry in a near-shore wave dominated environment in Duck, NC. The instruments were developed using a sensor insertion system (SIS) installed on the research pier. Bedforms were measured under a variety of wave conditions and at several cross-shore locations. Two existing empirical ripple prediction models are compared to the measured field data. A new model is also empirically determined from these measurements. This new model describes vortex ripple geometry strictly in terms of sediment and flow conditions and does not require the calculation of shear stress. When all available field measurements of small scale ripples are compared to these predictive models, the new model has lower errors in predicting ripple height, steepness, and length than any of the other models. Both small scale ripples with lengths of less than 40 cm, and large scale megaripples were measured during this experiment. Observations indicate that ripple flattening and reformation is a function of wave mobility number. Ripple reformation was commonly observed to occur in less than one minute after flattening and under certain conditions reformation was observed to occur within a few wave periods. Observations of megaripple migration indicate that cross-shore sediment transport due to ripple migration may be a significant process. Estimated cross-shore bed load transport rates are a good indicator of small scale ripple migration direction. Measurements indicate that megaripple lengths are proportional to near-bed orbital diameter, and that these types of bedforms can exist at small and large values of mobility number and orbital diameter.
Thesis:
Thesis (Ph. D.)--University of Florida, 1997.
Bibliography:
Includes bibliographical references (leaves 129-131).
Funding:
This publication is being made available as part of the report series written by the faculty, staff, and students of the Coastal and Oceanographic Program of the Department of Civil and Coastal Engineering.
Statement of Responsibility:
by Christopher D. Jettâe.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact Digital Services (UFDC@uflib.ufl.edu) with any additional information they can provide.
Resource Identifier:
41567372 ( OCLC )

Full Text
UFL/COEL-TR/115

WAVE GENERATED BEDFORMS IN THE NEAR-SHORE SAND ENVIRONMENT
by
Christopher D. Jetti

Dissertation

1997




WAVE GENERATED BEDFORMS IN THE NEAR-SHORE SAND ENVIRONMENT

By
CHRISTOPHER D. JETTED
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA

1997




ACKNOWLEDGMENTS

I wish to express a great deal of thanks to my advisor and supervisory committee chairman, Dr. Daniel M. Hanes, for his support and guidance during my graduate research. I would also like to thank the members of my supervisory committee, Dr. Robert G. Dean, Dr. Joann Mossa, Dr. Donald M. Sheppard, and Dr. Robert J. Thieke. I would also like to thank the ONR Coastal Sciences Program for funding this research.
I would like to thank my parents, Sandie and Bill Kline and Don Jett6, for their love and support over the years. I would also like to recognize my grandparents, Alma and Leon Jett6, for their unending encouragement. Thanks to my wife, Monica, for her support and for making my life wonderful. Thanks to Pierre and Helene for their support. Thanks to Eric Thosteson for help in data collection and reduction, and for being a great friend. Thanks to Mike and Craig for their aid in the collection of field data.
Thanks to everyone at the Coastal Engineering Lab, without their help this research never would have been possible. Thanks to Chuck, Sidney, Vernon, Vic, Danny, and J.J.. Thanks also to the clerical staff, Becky, Sandra, Cynthia, and Lucy, for having patience with my general failure to meet deadlines. Thanks to everyone at the Army Corps of Engineers Field Research Facility in Duck, North Carolina for allowing us to use their facility and for their assistance when it was needed.
Thanks also to my friends for the great times and sanity checks; thanks big Paul, little Paul, Phil, Pedro, Darwin, Kenny, Mark, Tim, Al, Mike, Rob, Cindy, Marc, Louis, and Christina. Thanks for entertainment go to Iron Wood Golf Course, Market Street Pub, Butler Beach, S. R. 441 bike rides, and 97X. I would like to thank God for my family, friends, and health, and also for giving us yeast. Much thanks to King Neptune for the awesome waves he has sent my way.




TABLE OF CONTENTS
pne
ACKNOW LEDGM ENTS ................................................................................................... ii
LIST OF TABLES .............................................................................................................. v
LIST OF FIGURES ............................................................................................................ vi
LIST OF SYM BOLS .......................................................................................................... ix
ABSTRACT ......................................................................................... xi
INTRODUCTION ................................................................. I .............................................. I
PREVIOUS W ORK ............................................................................................................ 4
M echanics of Ripple Form ation ....................................................................................... 4
Rolling Grain Ripples ................................................................................. I .................. 5
Vortex Ripples .............................................................................................................. 6
Classification Schem es .................................................................................................... 6
Bagnold (1946) ............................................................................................................. 6
Dingler and Inm an (1976) ........................................................................................... 7
Clifton (1976) ............................................................................................................... 7
Nielsen (198 1) .............................................................................................................. 9
Osbourne and Vincent (1993) ....................................................................................... 9
Recent Bedform Observations ....................................... I ................................................ 10
Hayand W ilson (1994) ........................ ............ 10
Hay, Craig and W ilson (1996) .................................................................................... 11
THE M ULTIPLE TRANSDUCER ARRAY .................................................................... 13
Previous M easurem ent Techniques ............................................................................... 13
The M TA ....................................................................................................................... 14
Testing of the M TA .................................................................................................... 17
A field test .................................................................................................................. 20
Analysis of m easurements .......................................................................................... 21
Conclusions ................................................................................................................ 24




FIELD MEASUREMENTS OF BEDFORMS AND A COMPARISON OF
PREDICTION TECHNIQUES .................................................................................. 26
Introduction .................................................................................................................... 26
Instrum entation and experim ental setup ........................................................................ 27
D ata Analysis ................................................................................................................. 30
Ripple M odel Descriptions ............................................................................................ 35
Nielsen (198 1) field m odel ......................................................................................... 35
W iberg and H arris (1994) ........................................................................................... 36
M odel Com parisons ....................................................................................................... 38
Nielsen (1981) ............................................................................................................ 38
W iberg and Harris (1994) ........................................................................................... 39
Using near bottom hydrodynamics in predicting ripple geometries .............................. 42
New M odel Developm ent .............................................................................................. 48
Comparison of new model with previous field ripple measurements ........................ 50
Conclusions .................................................................................................................... 53
OBSERVATION S OF BEDFORM DYNAM ICS ............................................................ 55
Introduction .................................................................................................................... 55
Experim ental Setup ........................................................................................................ 56
Ripple Flattening and Reform ation ................................................................................ 59
Ripple m igration ............................................................................................................ 67
Large Scale M igration ................................................................................................ 67
Sm all Scale M igration ................................................................................................ 74
M odel Com parisons ....................................................................................................... 82
Conclusions .................................................................................................................... 86
DISCUSSION AND CONCLU SIONS ............................................................................. 89
APPENDIX A: SIS95 AND SIS96 DATA AND PLOTS ............................................ ... 93
Key to term s in appendix A .......................................................................................... 93
Table A l. SIS96 Hydrodynam ic data ........................................................................... 94
Table A2. SIS96 Ripple D ata ........................................................................................ 95
Table A3. SIS95 Hydrodynam ic D ata ........................................................................... 96
Table A4. SIS95 Ripple D ata ........................................................................................ 97
APPENDIX B: SIS95 AND SIS96 PROGRAMS AND DATA FILE DESCRIPTIONS ... 98
Description of files ......................................................................................................... 98
Listing of program s ...................................................................................................... 100
BIOGRAPHICAL SKETCH ........................................................................................... 132




LIST OF TABLES

Table RMe
I Wiberg and Harris (1994) ripple classification .......................................................... 36
2. Relative error between measured and predicted ripple geometries ............................ 41
3. Relative error between measured and predicted ripple geometries when
significant near-bottom conditions are used as input ........................................... 48
4. Relative error between measured and predicted ripple geometries for all
previous sm all scale field data ............................................................................. 52
5. Relative error between measured and predicted ripple geometries for all available
small scale field data; including the SIS95 and SIS96 data sets .......................... 53




LIST OF FIGURES

Figure n
1. Schematic of the multiple transducer array ripple measurement system ............. 15
2. Laboratory test of MTA with sand covered corrugated fiberglass target.............. 17
3. MTA profiles measured in a laboratory wave tank...................................... 20
4. MTA profiles measured during a field experiment in Duck, NC...................... 24
5. Cross-shore beach profiles measured during SIS95 and SIS96 experiments......... 28
6. SeaTek 64 element MTA used in 51S96 experiment ................................... 29
7. SIS96 Instrument setup as viewed from the offshore direction....................... 29
8. Time series of mobility number and ripple profiles from SIS96 run 14.............. 31
9. Time series of mobility number and ripple profiles from SIS96 run 8.............. 32
10. Time series of mobility number and ripple profiles from SIS96 run 26............. 33
11. Measured nondimensional ripple height (I/A) and nondimensional ripple
length (V/A) versus Mobility Number, and (c) ripple steepness versus
Shields Parameter for SIS95 and SIS96 data with Nielsen (198 1) model
curves .................................................................................... 38
12. Nielsen (1981) predicted versus measured SIS95 and SIS96 ripple height,
length, and steepness.................................................................... 39
13. Measured SIS95 and SIS96 ripple data with Wiberg and Harris (1994)
model curves............................................................................ 40
14. Wiberg and Harris (1994) predicted versus measured SIS95 and SIS96
ripple height, length, and steepness ................................................... 41




15. Power spectrums for corrected surface elevation and near-bottom orbital
excursion for run 6 of the SIS95 data set ........................................................ 43
16. Orbital diameter versus run number and mobility number versus run
number for near-bottom significant values and Hmo derived values for
the SIS95 and SIS96 data sets .......................................................................... 45
17. Measured nondimensional ripple height (T/A), nondimensional ripple
length (X/A), and ripple steepness for SIS95 and SIS96 data with Nielsen
(1981) m odel curves ......................................................................................... 46
18. Measured SIS95 and SIS96 ripple data with Wiberg and Harris (1994)
m odel curves ................................................................................................... 47
19. Measured nondimensional ripple height (rj/A) and ripple steepness for
SIS95 and SIS96 data with the new model curves .......................................... 50
20. Nondimensional ripple height/ orbital semi-excursion and ripple steepness
versus Mobility Number with new model curves and all available field
ripple data ......................................................................................................... 5 1
21. Time series of mobility number and ripple height for run 2 ................................ 60
22. Small scale bedform profiles for run 2 .................................................................. 61
23. Time series of mobility number and ripple height for run 14 .............................. 63
24. Small scale bedform profiles for run 14 ............................................................... 63
25. Time series of mobility number and ripple height for run 20 .............................. 65
26. Small scale bedform profiles for run 20 ............................................................... 65
27. Under-water video images from Run20 ............................................................. 66
28. Time series of mobility number for runs 19, 20, and 21 ..................................... 68
29. Bedform profiles for runs 19, 20, and 21 ............................................................ 69
30. Rotating scanning sonar (RSS) image from the beginning of run 19 .................. 70
31. Mobility number time-series for run 9, and bedform time-series showing
m egaripple m igration ........................................................................................ 73
32. Time series of mobility number and ripple height for run 15 .............................. 75




33. Small scale bedform profiles for run 15 ............................................................... 75
34. Time series of mobility number and ripple height for run 16 .............................. 76
35. Small scale bedform profiles for run 16 ............................................................... 77
36. Time series of mobility number and ripple height for run 3 ................................ 78
37. Small scale bedform profiles for run 3 ............................................................... 78
38. Small scale bedform profiles for run 19 ............................................................... 79
39. Underwater video image from the beginning of run 19 showing brickpattern ripples .................................................................................................... 80
40. Measured SIS96 small scale and large scale ripple data versus
nondimensional orbital diameter for nondimensional ripple height,
length, and steepness ........................................................................................ 84
41. Megaripple steepness versus nondimensional orbital diameter, do/d50,
mobility number, and mean current for the SIS96 data set ............................... 85




LIST OF SYMBOLS

A [L] orbital amplitude of fluid just above boundary layer
Ar [L2] cross-sectional area of megaripple
c [LT-'] megaripple migration rate
dh/dx [-] bedslope
d50 [L] median grain diameter
do [L] wave orbital diameter
f [T-'] wave frequency
fp [L] diameter of acoustic foot print
f2.5 [-] grain roughness friction factor
g [LT-2] acceleration due to gravity
h [L] water depth
Hmo [L] wave height determined from surface elevation spectrum
k [L-1] wave number, 21t/L
Qb(t) [L2T-1] bedload transport rate
Qs [MT-'L-'] sediment transport rate
s [-] sediment relative density
TP [T] peak wave period from surface elevation spectrum
u [LTl] cross-shore velocity
um [LT-1] maximum near bottom orbital velocity
z [L] vertical distance above bed
A [-] relative error
8, [L] wave boundary layer thickness




[-] sediment porosity
F(t) [-] non-dimensional bedload transport rate
n [L] ripple height
TI ano [L] anorbital ripple height
Tlorb [L] orbital ripple height
ilsub [L] suborbital ripple height
k [L] ripple length
kano [L] anorbital ripple length
Xorb [L] orbital ripple length
Xsub [L] suborbital ripple length
02.5 [-1 grain roughness Shield's parameter
p [ML-3] fluid density
o [T-'] radian frequency, 2rt/T
[-] mobility number
dh
[-] local bed slope
dx




Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy
WAVE GENERATED BEDFORMS IN THE NEAR-SHORE SAND ENVIRONMENT By
Christopher D. Jett6
December 1997
Chairman: Daniel M. Hanes
Major Department: Coastal and Oceanographic Engineering
To measure bedform geometry, a high-resolution sea-bed profiling system has been developed. The multiple transducer array (MTA) consists of an array of ultra-sonic (5MHz) transducers. The first prototype MTA described herein measures twodimensional bedform profiles over a length of 45 cm. Under ideal conditions, the instrument is capable of 1 mm vertical resolution and 2 cm horizontal resolution. Laboratory and field tests of the MTA show the system's capability to accurately measure bedforms of known geometry and the ability to work under conditions with relatively high suspended sediment concentrations. Previous high-resolution profiling systems have either had moving parts, or have been unable to obtain the resolution of the system described herein.
A multiple transducer array (MTA) was used to measure seabed geometry in a nearshore wave dominated environment in Duck, NC. The instruments were deployed using a sensor insertion system (SIS) installed on the research pier. Bedforms were measured under a variety of wave conditions and at several cross-shore locations. Two existing empirical ripple prediction models are compared to the measured field data. A new model




is also empirically determined from these measurements. This new model describes vortex ripple geometry strictly in terms of sediment and flow conditions and does not require the calculation of shear stress. When all available field measurements of small scale ripples are compared to these predictive models, the new model has lower errors in predicting ripple height, steepness, and length than any of the other models.
Both small scale ripples, with lengths of less than 40 cm, and large scale megaripples were measured during this experiment. Observations indicate that ripple flattening and reformation is a function of wave mobility number. Ripple reformation was commonly observed to occur in less than one minute after flattening, and under certain conditions reformation was observed to occur within a few wave periods. Observations of megaripple migration indicate that cross-shore sediment transport due to ripple migration may be a significant process. Estimated cross-shore bedload transport rates are a good indicator of small scale ripple migration direction. Measurements indicate that megaripple lengths are proportional to near-bed orbital diameter, and that these types of bedforms can exist at small and large values of mobility number and orbital diameter.




CHAPTER 1
INTRODUCTION
In the coastal environment the seabed is rarely flat. In contrast, it is usually covered with bedforms such as bars, dunes, sand ripples, and biological matter. In modeling sediment transport it is very important to be able to predict the geometry of the bedforms. The geometry of bedforms provides a bed roughness that influences the shear stress at the bed and the profile of the eddy viscosity or diffusion coefficient. Thus the bedform geometry is required to determine the concentration profile or the rate of sediment transport (Vongvisessomjai, 1987).
The modeling capability of bathymetric change is also dependent on the modeling of bedform geometry. Vincent and Osbourne (1992) found that under similar wave conditions the suspended sediment over a flat bed was restricted to the lowest 5 to 10 centimeters above the bed, whereas when bedforms were present suspensions to 30 and 40 centimeters were common. They also found bedform migration to be a significant part of the total cross-shore transport when the bedforms present had low steepness. Thus the ability to model sediment transport and bathymetric change is dependent on the ability to model bedform geometry and bedform migration.
Most ripple models currently in use are empirically based. Nielsen (1981) developed separate models for bedform geometry based on the type of waves present, i.e., one model for laboratory (regular wave) and a different model for field (irregular wave) conditions. These models are functions of the mobility number and the grain roughness Shield's parameter and can be applied to sediments of different densities. Wiberg (1994) developed a set of empirical models that are meant to work in both laboratory and field wave




conditions. These models use the wave orbital diameter and sediment size to predict the geometry of bedforms under oscillatory flows for a quartz sand bed.
One of the major problems in modeling bedform geometries under waves has been the lack of reliable field data. Nielsen's (1981) field ripple models were based on 126 data points, of which 54 were measured by scuba divers and 72 were measured using acoustic instrumentation. When Wiberg and Harris (1994) developed their models they took all of the data used by Nielsen plus 59 points recorded after 1981. All reliable existing field data sets were used to construct this model. Thus there were no data with which to check the model.
In the past, the most common method for measuring bedforms was for scuba divers to use some sort of manual measurement device. With the use of a meter stick, divers can accurately determine ripple lengths. Ripple heights, however, tend to be more difficult to measure. When measuring ripple heights, the weight of the meter stick tends to flatten the crests of the ripples. In addition, the water motion due to the waves tends to complicate matters (Dingler and Inman, 1972). In the past divers have also used a piece of clear plastic and a grease pencil to try to measure sand ripples in the field (Inman, 1957). Also divers have tried to use a large "comb" covered with grease so that once it is pushed down into the bed the sand sticks to the grease and thus the bed profile is recorded (Kawata et al., 1992).
More recently underwater video has been used to measure ripples. When water visibility allows, videography can be used to accurately measure ripple length. However, even using available image processing techniques, ripple height is still very difficult to determine in the field using underwater video.
The latest trend in measuring bedform dimensions is with the use of ultra-sonics. Dingler and Inman (1974) were one of the first groups to measure bedforms with ultrasound. They had a movable frame with a manually translated transducer which the divers moved back and forth over the rippled seabed recording the ripple profile. Today




there are similar automated systems such as the HRRTS (High Resolution Remote Tracking Sonar). With the use of a stepper motor, the HRRTS tracks an ultrasonic transducer back and forth across a stationary frame recording the bottom profile (Greenwood et al.(1993)). Also transducers have been used in side-scan mode to measure ripple lengths in conjunction with downward aimed transducers to measure ripple heights as ripples migrate underneath (Vincent and Osbourne, 1992). Because there have been inherent problems involved in all of these methods, a new approach to measuring bedform geometries has been developed. This technique involves no moving parts and is capable of remotely measuring bedform profiles with millimeter vertical resolution.




CHAPTER 2
PREVIOUS WORK
Mechanics of Ripple Formation
Once the wave shear stress acting on a movable sand bed reaches a strength strong enough for incipient sediment motion, the sediment begins to be carried as bed load. As the bed shear stress increases, the amount of suspended sediment increases and the flow becomes more vigorous. Then, once another critical value of shear stress is surpassed, the sediment begins to be carried as sheet flow. Bedforms may be active anytime the bed shear stress is above that required for incipient motion.
One non-dimensional parameter used to classify these different regimes is the grain roughness Shield's parameter (02.5 ),
0 2.5 = 0 .5 f2.1 Vf equation 2.1
where "f2.5" is the wave friction factor, given by Swart (1974) as:
f25= e xp (5.2 13 (2.5 d,50/A ))0194 5.977 equation 2.2
and Iis the mobility number, defined as:
(Ac)2 euto .
=(s 1)gd 50eqai23 Where "d50" is the median grain diameter, "A" is the water semi-excursion determined from the H.o wave height and peak period, and 'W' is the peak angular frequency of the




waves as determined from the surface elevation spectrum. The mobility number is a nondimensional term representing essentially the ratio between the water velocity amplitude and the settling velocity of the sediment for typical wave conditions (A of the order 0.1 m2m) and sand sizes (Nielsen, 1981).
Values of the grain roughness Shield's parameter are given to classify the different regimes of flow. For 02.5 less than or equal to 0.05, the flow is considered to be too weak for sediment motion. For a 02.5 greater than 0.05 and less than or equal to 1.0 the bed will be active and covered with bedforms that will be approximately in equilibrium with the flow conditions. When 02.5 is greater than 1.0, the ripples are flattened and sheet flow occurs (Dingier, 1974).
Rolling Grain Ripples
Bedforms produced by the action of progressive waves on a horizontal bed were first classified into two groups by Bagnold (1946). Bedforms with a large enough height to length ratio to allow vortex formation in the lee of the crest he called "vortex ripples," and bedforms with too small of a height to length ratio to allow vortex formation he called "rolling-grain" ripples. On an initially flat bed without obstructions capable of forming vortices, rolling-grain ripples are the first bedforms to appear.
Sleath (1984) investigated rolling-grain ripple formation on a flat bed with a small disturbance present; the required disturbance could even be as small as a single grain of sediment projecting from the bed. He modeled the streamlines around this disturbance and found that the steady drifts, which form in the vicinity of the bed, tend to be directed toward the disturbance. This would carry sediment from the trough of the ripple to the crest. Such flow would promote ripple growth. Gravity acts as an opposing force to ripple formation, therefore limiting the maximum height attainable of the ripples under given wave conditions.




Vortex Ripples
If the flow is strong enough, once the height to length ratio of the rolling grain ripples reaches a critical value, vortex formation will occur in the lee of the ripple crests twice every wave period. Vortex ripples are commonly found at low to moderate transport rates (Sleath, 1984). Once separation occurs in the lee of the ripple crest both the mainstream flow on the upstream face, and the reverse flow on the lee side face of the crest tend to carry sediment from the trough to the crest of the ripple. The forces acting in opposition to vortex ripple formation are gravity and the erosional effects on the ripple crest by the main stream flow. Thus for given sediment and wave conditions, an equilibrium bedform geometry will exist where there is no net transport across any cross-section of the bedform profile.
Classification Schemes
Several previous investigators have classified bedforms in different ways. Most of the classifications used in sediment transport are based on the genesis of the bedforms. This section divides classification schemes up by authors and lists them in chronological order. Bagnold (1946)
As mentioned in the above section, Bagnold (1946) was the first to develop a classification scheme for bedform ripples. He called ripples with a height to length ratio large enough for vortex formation in the lee of the ripple crest "vortex ripples," and ripples without this separation "rolling grain ripples." Bagnold's classification system is still widely used today. Rolling grain ripples appear to be stable at velocities smaller than twice that required for incipient motion (Bagnold, 1946). Because of this, rolling grain ripples are present when sediment transport is low; thus, most of the bedform research has been done in the vortex ripple regime where sediment transport rates are greater. The vortices formed by the flow over vortex ripples are strong; Tunsdall and Inman (1975) found that




the velocities within the vortices are of the same magnitude as that of the main stream flow.
Dingier and Inman (1976)
Dingier and Inman (1976) found that three distinct ripple types occur on fine sand beds in coastal areas. These are relict ripples, vortex ripples, and transition ripples. Relict ripples are found when wave parameters are not sufficient to initiate sediment motion on the bed. Relict ripples are the remains of ripples that were formed under a previous event when flow conditions were strong enough for sediment motion. Vortex ripples are defined as ripples that have a relatively constant steepness; even though the ripple heights and lengths change as the wave intensity becomes larger. Once the steepness of the ripples begins to decrease due to increased flow intensity, the ripples are called transition ripples. Transition ripples are found when the flow intensity is too large to maintain the relatively constant vortex ripple steepness of 0.15, but not large enough for sheet flow. Dingier and Inman (1976) used mobility number to classify flow intensity.
Values of mobility numbers (y) were determined for each of the ripple classes. When
was less than 40 but great enough for sediment motion, the bedforms were found to be in the vortex ripple class. When the flow intensity increased and XV was greater than 40 but less than 240 the bedforms were found to be in the transition ripple class. When the flow intensity was greater yet (Wj greater than 240) no bedforms were present and sheet flow was observed (Inman and Dingier, 1976).
Clifton (1976)
Clifton (1976) found that when vortex ripple lengths were compared to the ratio of orbital diameter to median grain diameter (djd50) the ripples could be classified into three categories. First, when ddd50 is relatively small, less than 2000, the ripple length (k) is proportional to orbital diameter, Clifton (1976) classified these ripples as orbital ripples. For large values of djd5o, greater than 5000, the ripple length appears to be a function of




grain diameter and nearly independent of orbital diameter. Clifton (1976) refers to these ripples as anorbital ripples. There exists a transition range for intermediate values of dJdo where both orbital and anorbital ripple formations appear possible, as well as ripples with intermediate lengths. Clifton (1976) called ripples in this transition range suborbital ripples.
Wiberg and Harris (1994) used the above ripple classification system, however they quantified values of flow parameters and bedform geometry for each of the different classes. Wiberg and Harris (1994) found orbital ripples to dominate in laboratory environments because of the limits on the maximum wave orbital diameter that can be generated in the laboratory with most wave flumes. Wiberg and Harris (1994) defined orbital ripples as ripples with heights twice as large as the thickness of the wave boundary layer, or 5Jrl < 0.5 Since for this flow the thickness of the wave boundary layer (8.) is roughly proportional to the wave orbital diameter (d,), and d. is more readily attainable in most instances, Wiberg and Harris (1994) defined orbital ripples as ripples with the ratio do/I<10. The Wiberg and Harris (1994) ripple model classifies the ripple regimes according to the ratio of djdlano, where Tlano is the predicted anorbital ripple height. When predicted anorbital ripple height is used, the definition of the orbital ripple regime becomes do/Tlano<20, which was determined empirically. Wiberg and Harris (1994) classify anorbital ripples as ripples with heights several times smaller than the wave boundary layer. By definition anorbital ripples have heights less than one quarter of the wave boundary layer thickness, or 8JTiano > 4.0. For ease of calculation, this value was also put in terms of orbital diameter; giving doflano > 100.
Wiberg and Harris (1994) then classify the transitional suborbital regime for conditions where the ratio JTi.ano is greater than 0.5 and less than 4.0, or in terms of the wave orbital diameter, 20 < do/hano < 100. When laboratory and field data are compared for a given value of djd50 in the suborbital regime, the ripple wavelengths can differ by more than a factor of three. Anorbital ripples dominate on the continental shelf, and on




medium to coarse sand beds suborbital ripples can also be present. Under limited conditions on coarse sand beds orbital ripples can also be present on the continental shelf, however it is not common (Wiberg and Harris, 1994). Nielsen (1981)
Nielsen (198 1) studied vortex ripples and classified them by the type of waves present as either laboratory (regular wave) or field (irregular wave) ripples. He found that under field conditions, where the waves are often irregular and far from being sinusoidal, the steepness of the ripples tend to be less than those formed under laboratory conditions. In the past, laboratory studies could not generate prototype field wave conditions. However, with new large wave flumes operating under spectral conditions, prototype field wave conditions can be generated in the laboratory. This makes using the field and laboratory classification system obsolete since it is difficult to draw a fine line between the two conditions. Wiberg and Harris (1994) suggest that the differences Nielsen (1981) found between lab and field data are probably due to most of the data falling into the orbital ripple regime for the laboratory data, and into the suborbital and anorbital regime for the field data.
Osbourne and Vincent (1993)
Upon completion of a field study that included diver observations of bedforms, Osbourne and Vincent (1993) classified observed bedforms into two major categories, small-scale bedforms (rj<4 cm, X<20 cm) and large scale bedforms (11>4 cm, X>20 cm). These bed types were then divided into subtypes depending on the number of crest dimensions, type of vortex shedding/suspension, and symmetry. The number of crest dimensions were defined as two dimensional if the bifurcation density was low (> 10 cm between bifurcations) and as three dimensional if the bifurcation density was high (<10 cm between bifurcations). The vortex shedding/suspension classification named pre-vortex ripples as ripples with no vortex shedding and no suspension. These types of ripples have




previously been referred to as rolling grain ripples. Vortex ripples were the same as defined previously, with sediment suspension and vortex shedding from crests at regular intervals under large waves. Post-vortex ripples were classified as ripples where sediment suspension and vortex shedding occurs at irregular bursts. The ripples were then characterized as being either symmetric or asymmetric in profile.
Osbourne and Vincent (1993) found that under non-breaking wave conditions smallscale bedforms dominated, while under breaking wave conditions large-scale bedforms dominated over the sea-bed. These large-scale bedforms were generally crescentic in planform and migrated on-shore, they have been called lunate-shaped megaripples by Clifton (1976). In their field study Osbourne and Vincent (1993) found that two dimensional small-scale bedforms (TI=0.5 to 3 cm, X=8 to 20 cm) had migration rates of up to 5 cm./min., and for well defined lunate megaripples (rj=3 8 cm, X,=20 80 cm) migration rates of up to 3 cm/min were documented.
Recent Bedform Observations
Hay and Wilson (1994)
A rotating sidescan sonar (RSS) was used to measure bedforms within a 10 meter diameter field on the crest of a nearshore bar. Hay and Wilson (1994) documented planform images of three-dimensional bedforms present throughout a storm on Lake Huron in Burley Beach, Ontario. Such a sonar system proved capable of measuring the orientation and length of small and large-scale bedforms under certain conditions. Their measurements of bedforms indicate that the seabed can be highly three-dimensional.
Transitions ,of bedform types were observed during the progression of a storm. Clifton (1976) discussed such transitions in a cross-shore sense; however, Hay and Wilson (1994) observed such transitions in a single location as the wave and current conditions changed. The progression was observed during decay in the wave velocities. The




progression was from a 'flat' bed with no vortex ripples, to a bed populated with longcrested shore-parallel ripples with occasional megaripples. The seabed was then populated with oblique cross ripples combined with patchy shore-parallel ripples and occasional megaripples. Finally, the bed was covered with irregular short crested ripples. The time for this progression was approximately six hours. During the onset of the storm, a similar progression was observed, but in reverse sequence, and over the course of only 2 hours.
Hay and Wilson (1994) demonstrate that planform images of the seabed can be recorded with a rotating side-scan sonar under certain wave and current conditions. They also indicate that information on bedform amplitudes is contained in the images as shadow length, however it is not as readily accessible as bedform patterns. The major finding of this paper is that the local evolution of the bed involves transformations among twodimensional and three-dimensional bedform fields, and that this transformation can occur on time scales of a few hours or less.
Hay, Craig and Wilson (1996)
Bedform measurements from Duck, NC are presented. A rotating side-scan sonar and a rotating pencil beam sonar were used to measure bedforms. The rotating pencil beam sonar consisted of a 2.25 MHz sonar with a 2 degree farfield beam-width that rotated about a horizontal axis. Vertical slice images recorded with this instrument were used to measure bedform height. The main focus of this investigation was to measure megaripple dynamics.
These observations were similar to the Hay and Wilson (1994) observations in the sequence of bedform patterns from megaripples, through cross-ripples, to irregular ripples as wave energy decreased. Also, in both circumstances the low energy end-state of the seabed was short-crested irregular ripples. They propose that such a progression suggests the existence of general relationships between fluid forcing and bedform patterns.




Their observations indicate that lunate megaripples migrate in the direction of their horns, and that megaripple migration rates appear to be a function of the bedform length scale. Both long-shore and cross-shore megaripple migrations were observed. These measurements indicate that the direction of migration depends on the strength of the longshore current. Migration rates reached 50-160 cm/hr longshore, and 20 cm/hr onshore during this experiment. The sediment transport rate resulting from the onshore migration of megaripples was in agreement with a cross-shore bedload transport formula based on laboratory experiments. This supports the suggestion of Hay and Bowen (1993) that the cross-shore sediment transport due to the migration of meter-scale bedforms during a storm could be predicted with laboratory-based bedload transport formula.




CHAPTER 3
THE MULTIPLE TRANSDUCER ARRAY
Previous Measurement Techniques
During the past forty years bedform measurement techniques in the field have evolved from scuba divers using meter sticks to remotely operated acoustic profilers (e.g. Inman 1957; Dingler and Inman, 1976; Osbourne and Vincent, 1993; Greenwood et al., 1993). Most of the previous acoustic systems used a single element sonar head, which was either translated over a track near the bed or was oriented in a side-scan mode. There are problems that exist with both of these arrangements.
Several systems have been developed using a single downward aimed transducer that is translated across a track supported by a frame near the sea bed (Dingler and Inman, 1976; Greenwood et al., 1993). These systems are able to measure bedform profiles, but there are some short-comings with such arrangements. The most difficult problem to overcome when using mechanical parts in the nearshore ocean environment is vibration of the sonar head during travel across the track. Also, the size of the complete system makes deployment difficult and could possibly influence the local hydrodynamics. An advantage to this system is that the measured profile length can span several meters.
In side-scan mode the sonar head is oriented slightly downward from horizontal in order to receive returns from the upslope portion of the bedforms. It is the authors' experience with such instruments that under times of relatively high flow intensity it is difficult, if not impossible, to separate the returns from the upslopes of the bedforms from the returns due to suspended sediment. Also, with this type of instrument only ripple




lengths can be measured. There are unacceptable errors when estimating small scale ripple heights from a single sonar with such orientation.
Recently, rotating side-scan sonars that can be used in sea-floor imaging have become commercially available. These systems operate similar to the side-scan system described above except that the transducer rotates about its vertical axis in order to give a 360 degree image of the sea-floor. These systems can be used to measure small-scale ripple lengths, however they have some of the same shortcomings as the stationary side-scan sonar described above. It is difficult under intense flows to separate suspended sediment returns from actual returns from the sea-floor. Also, small scale ripple heights have yet to be accurately measured with such a system. An advantage to this system is that under certain conditions ripple orientation and length can be recorded (Hay and Wilson, 1994).
The MTA
In this chapter an acoustic multiple transducer array (MTA) is described, which is capable of measuring bedform geometries with one millimeter vertical resolution and two centimeter horizontal resolution. This system uses an array of stationary downward aimed transducers, thus eliminating any moving parts and the necessity for a large support frame and track. The transducers are fired sequentially across the linear array. The bottom locations are recorded for each transducer, thus recording a two-dimensional sea-bed profile.
The first prototype multiple transducer array consists of 37 transducers. The center to center spacing of the 10 mm disk shaped transducers is 1.2 cm, which gives a profile length of 45 cm. The operating frequency of these transducers is 5 MHz, resulting in a wave length of 0.3 mm. The -3 dB half beam angle is 0.9 degrees and the mechanical quality factor (Qm) for the transducers is 500. The length, width, and height of the array are 50 cm, 8.5 cm, and 5 cm respectively.




bottom profile

The transducer array is connected to a dedicated under-water electronics package. This package contains the driver and receiver circuitry, a multiplexer board, a data-logger, and the timing circuitry necessary to run the transducers. A multiplexer is used so that a single driver and receiver can be used for all of the transducers.

multiplexer, driver& receiver circuitry, and data locicier (_

cross section of IATA

transducers

transmitted 5 MHz ultrasonic signal

recorded points along bottom profile

Figure 1. Schematic of the multiple transducer array ripple measurement system
A threshold method is used for bottom detection. Once the transmitted pulse is sent from the transducer a counter is started. When the amplitude of the 5 MHz return signal exceeds a certain threshold voltage the counter is stopped and the value on the counter is stored in memory. This value represents the amount of time taken for the sound to travel from the transducer to the target and back. Once this time is found, with knowledge of the speed of sound, the distance from the transducer to the target is determined.




Because the system is microprocessor controlled, many settings can be changed from a remote computer connected to the data logger. The hardware settings that are controllable remotely include blanking, threshold voltage, pulse width, and amplifier gain. Blanking refers to a delay between the time that the counter begins and the time that the voltage comparator begins to monitor the return signal. This blanking enables the comparator to only look for a return from the distance where the target should be, thus reducing the number of false returns from sediment in suspension. The threshold used by the comparator can also be controlled remotely. The pulse width, which is generally set at 12 micro-seconds, and the amplifier gain on the return signal are other settings that can be changed to optimize functionality under various environmental conditions.
Many software settings can also be changed remotely. These include the number of scans to record, the time between the beginning of scans, a value for the speed of sound, and any desired averaging technique. This system allows the user to take a few profiles, send them back to a computer where they can be analyzed, and then change the settings in order to reduce the amount of false returns and missed returns. Once the system settings are acceptable, the NITA can then start to collect data. With extended RAM on the datalogger, the NITA can run for over 30 uninterrupted hours while taking one 37 point scan every 6 seconds.
Each scan consists of a time stamp and the distances to the target in millimeters. The files are saved in a format that allows the data to be displayed within seconds of its retrieval on a remote computer.
Testing of the NITA
Laboratory tests were conducted at the University of Florida Coastal and Oceanographic Engineering Laboratory. For one of the tests a corrugated fiberglass sheet was coated with sand to represent a rippled bed. The cross-sectional profile of the




corrugated piece could roughly be represented by a sinusoidal curve with a height of 11 mm and a length of 67 mm. Five scans were recorded using the MTA. The measurements from these scans are plotted in figure 2. The measured distance between the transducer and the target has been subtracted from the average distance between the transducer and the target in order to invert the profile to its correct orientation (crests at higher elevation than troughs). Also plotted is a representative profile curve for the target that was found using the MTA data with analysis techniques which will be described in detail later. The methods used to determine ripple height and length from MTA data found a ripple height of 9 mm and a ripple length of 67 mm. for the sand covered corrugated fiberglass target.
20Measured Ripple' Lt.= 67 rhmn 'MTA Ripll Lt.= 6tmm
E Measured Ripple Ht.= 11 mm MTA Ripple Ht.= 9,mm
-20 I
0 50 100 150 200 250 300 350 400 450
Horizontal Distance in mm
Figure 2. Laboratory test of MTA with sand covered corrugated fiberglass target.
The MTA was able to determine the wavelength of the corrugated target to within one millimeter. However, in determining the simulated ripple height of the target, the MTA data showed an average ripple height 2 mm less than the measured height of 11 mmn between the crests and troughs of the target. This problem is due to the narrowness of the troughs. Since all of the elevations of the ripple crests are constant, it appears that the elevation of the crests has been captured correctly. However the measured elevations of the ripple troughs vary over the profile. This is due to the size of the acoustic footprint (-2 cm) compared to the length of the ripple trough (-I1 to 2 cm).




Since this instrument uses a comparator to find the return echo, the first large return after the blanking period is used to determine the range to the target. If a significant portion of the footprint were to cover an area higher in elevation than the area that the majority of the footprint covered, it is possible that the echo from the area higher in elevation would be recorded as the echo from the target. Most likely the first strong acoustic return from the target, for the transducers above the troughs, came from the sloped area of the target on either side of the troughs. This would tend to vary the measured elevation of the troughs, which was observed.
The average distance to the target for this test was 48.5 cm, which would give a footprint on the order of 2 cm in diameter. Even if a transducer was perfectly lined up over a trough the elevation change of the target over the 2 cm footprint area would be 4 nim. Returns from the sides of the trough would decrease the measured range to the trough, which is what was observed.
Laboratory tests were also conducted in a large wave tank to determine the instrument's capability to measure actual bedforms composed of fine sand in an environment where other methods of bedform measurements were possible. The wave tank had a sloped bed composed of 0.09 mm sand. The measurements were taken after an experiment was run involving spectral waves. Measurements were taken at different stations along the profile in order to cover a variety of size and types of ripples.
The sides of the wave tank were glass so that the ripple profiles at the side of the tank could be measured visually using an incremented measuring device. Figure 3 contains five scans of raw data taken at each station as well as representative profiles for the four separate stations. Again each profile was analyzed using methods that will be described later to find a representative ripple height and length for each profile. These measurements along with visual measurements are also included as text in the plots.
In figure 3, plot (a), the observed ripples at this station could be classified as two dimensional ripples, indicating that the ripple heights and lengths were similar over the




area, and that the ripple crest lengths were long. For plots (b) and (c) the observed ripples were less regular than in the first plot, and in plot (d) the ripples were very irregular, or brick-patterned ripples.
For each of the profiles, the ripple heights and lengths were similar to what was measured at the side of the tank. The differences might have been due to actual changes in ripple dimensions from the side of the tank to the middle of the tank. For the profiles in plots (a) and (b) the ripple heights measured with the MTA were actually greater than the ripple heights measured at the side of the tank. The measured ripple lengths varied slightly between the two methods. For the profile found in plot (c) measured ripple dimensions were virtually identical with the visual measurements. The profile in plot (d) showed an irregular profile, which was observed, and the measurements were similar to the visual observations taken from the side of the tank. The elevation of the MTA above the bed was 18.5 cm, 22 cm, 23.5 cm, and 39.0 cm for profiles a, b, c, and d respectively.
Under ideal conditions, the system is capable of 1 mm vertical resolution and accuracy. However, as described above, the footprint of the acoustic beam at the design elevation of 40 to 50 cm is approximately 2 cm. Thus the system's accuracy will decrease as the change in bottom elevation over the footprint area increases. The horizontal resolution and accuracy for each transducer are taken as the footprint area of 2 cm at the design elevation. As shown, if several bedforms lengths are recorded with the MTA, much better accuratcy in determining bedform lengths can be achieved.




(a)
20 50ie 1= 50 200A () 2e m
visual Ie -t.'6 mm 'MIA Biii5Ie Ht.=8mm
_2
-200 50 100 150 200 (b) 250 300 350 400 450
2 0 *4
E Vsa mm i eL t=r .m
-20
E 20 50 100 150 200 (d) 250 300 350 400 450
20 Visual ipple Ht.'-6 to 8 n 'M IA Bipble Ht.=6!' mm
visual"I pie Lt.= 51 mm :IA Ipple t.=5 mm
> -20
0 50 100 150 200 ()250 300 350 400 450
20 5 0
Figure 3. MTA profiles measured in a laboratory wave tank
A field test
A joint experiment was conducted with the University of Florida and the Army Corps of Engineer's Field Research Facility in Duck, NC during August 23-25, 1995. The sensor insertion system (SIS) located on the research pier was used to deploy the MTA along with many other sensors. The SIS allowed the instruments to be deployed at various locations along the near-shore profile.
During the three day experiment the MTA was deployed under a variety of wave and sediment conditions. Ripple profiles were recorded across the near-shore profile from offshore of the bar, 225 m from shore, to the beach face, 30 m from shore. During the experiment the maximum Hmo wave height was 0.95 m with a 6 second peak period. The median sediment size varied from 0.18 mm to 1.66 mm over the area of measurements. The elevation of the MTA above the seabed was approximately 42 cm for all measurements.




The MTA proved capable of measuring bedform profiles under these wave and sediment conditions. As can be expected with all acoustic instruments, there were some false bottom returns from suspended sediment. However for every run the bedform profile was recorded sufficiently well to determine the geometry of the bedforms. In total there were 30, 13 minute, runs covering 9 locations along the near-shore profile. Analysis of measurements
The prototype MTA system scans at a rate of one scan every six seconds. Several scans were combined in order to find a single representative bedform profile. For most of the analysis ten scans were compared to determine a profile representing the seabed for one minute of sampling.
Because of occasional false returns, an algorithm was developed to estimate the true bedform. shape. For each of the 37 transducers a histogram was constructed of the measured bottom locations from the 10 scans. The bin size for this histogram was 1 mm. Only identical measurements were included in each bin. Then the histogram was analyzed to determine which bin had the highest number of occurrences in it. If two of the bins had the same number of occurrences, then the bin representing the farthest distance from the transducer was used. If the number of occurrences in this bin was three or greater, then this distance was recorded as the distance to the bed for the location being analyzed.
If the number of occurrences in this bin was less than three, another histogram was constructed. This second histogram had a bin size of 2 mm. If the number of occurrences in two of the bins were the same, then the bin farthest from the transducer was used. If the number of occurrences in this bin was 2 or greater, this bin was recorded as the distance to the bed. If the number of occurrences was less than 2, a third histogram was constructed with a bin size of 3 mm. Again the histogram was analyzed and the maximum bin farthest from the transducer was recorded as the distance to the bed. For the cases when the bin size was greater than one millimeter, the location of the bed within the appropriate bin was




set to the lower millimeter for the 2 mmn bin size and to the second lowest millimeter within the bin for the 3 mm. bin size.
This histogram method was carried out for each of the 37 locations along the profile in order to determine the bed location for each of the measured points. A cubic spline fit was then performed on the resultant 37 point profile in order to smooth the profile for further analysis. The smoothed profile had 109 points. This profile was then detrended to remove any linear trend, thus making the profile horizontal. Then the entire profile was subtracted from the mean bed location in order to invert the profile. This inverted profile is the variation of the bed about its mean value; where the positive elevation peaks represent bedform. crests. Note that if the orientation of the MTA is known, then before detrending the data the local bottom slope can also be measured.
The smoothed representative profiles were then plotted with the original raw scan data to make sure that the correct bed-level locations were used. If there was ambiguity, scans taken previous to and after the profile in question were used to help determine the correct bed location. For profiles taken under waves sometimes a large suspension event can cause the MTA system to lose the bed. In these cases, profiles taken before and after the suspension event were used to determine the bed-level.
The smoothed and detrended ripple profiles were analyzed to determine representative ripple lengths and heights from each profile. All maximums and minimums in a profile were first located. A threshold of 3 mmn was then set on the ripple height criterion. This was to eliminate the possibility of mistaking small fluctuations introduced during the smoothing process as ripples. The vertical distance between each adjacent maximum and minimum was then compared to this threshold value. If it was greater, this distance was recorded as a ripple height. If it was less than the threshold distance, and if the last point to exceed the threshold was a maximum, the routine continued to search for a minimum that met the threshold requirement. If during this search a larger maximum was found than the last one recorded, this larger maximum replaced the previously recorded




maximum. This process was continued until a minimum was found that met the threshold requirement. If the last recorded value was a minimum, then a similar process was executed to look for a maximum that met the threshold requirement.
Ripple heights were found by taking the average vertical distance between each adjacent maximum and minimum over the entire profile. Ripple lengths were found by taking the average horizontal distance between adjacent maximums and adjacent minimums. Ripple steepnesses were then found by dividing the average measured ripple height by the average measured ripple length for each profile.
The profiles were then analyzed again with a different threshold value for ripple height. Average ripple height, length, and steepness were then determined by this new set of threshold maximum and minimum values. The threshold value of one-half the average ripple height, which was the value found using a threshold of 3 mm, worked well to find a best fit ripple height and length for most of the ripple profiles in this data set.
An example of MITA data from the field experiment is shown in figure 4. The asterisks represent one minute of raw data (10 scans) with means subtracted. The solid curve is a representative bottom profile found using the data analysis method described above. Plot (a) is from offshore of a longshore bar, 225 m, from shore. Plot (b) is from inside the trough of a longshore bar, and plot (c) is from the beach face, 30 mn from shore. The median sediment sizes for these runs are 0.28, 0.40, and 1.46 mm respectively. The water depths for these three runs were 6.2, 4.4, and 2.5 m and the Hm. wave heights were 0.7, 0.3, and 0.2 mn respectively. The average measured ripple heights for the runs plotted in figure 4 are 13, 40, and 27 mm and the average measured ripple lengths are 103, 275, 197 mm respectively.




(a)
EX
0 50 100 150 200 ()250 300 350 400 450
EIE
>E -20 .... .... ......................... .. .............. ........... ...........
*~ 0 50 10 5 0 5 0 5 0 5
(c
E 2 0 . . K *. .a .6. ... .. . .
(cK
0 50 100 150 200 250 300 350 400 450
Horizontal Distance in mm
Figure 4. MITA profiles measured during a field experiment in Duck, NC
Several instances of false returns from suspended sediment can be seen in figure 4; these points are the asterisks well above the plotted curve. Since one minute of data is plotted in each plot, some of the scatter near the solid curve is most likely due to movement of the sensor frame and the seabed over this one minute period. Also, as mentioned earlier, near the bed there are a larger number of acoustic scatterers that can cause false bottom returns from the area just above the sea bed. .Conclusions
The Multiple Transducer Array (MTA) ripple measurement system can be used to measure small-scale, two-dimensional, bedform profiles. The MITA has several advantages




25
over its predecessors: (1) the system contains no moving parts, which in previous systems have caused vibration and have become fouled: (2) the system has the capability of taking complete profiles in less than 5 seconds, which allows the comparison of multiple scans to better eliminate false returns; (3) the system can be operated remotely even under severe sea-states, and the settings of the system can be changed remotely to better match the environment; (4) the size of the system is small compared to previous profilers which allows easier deployment and creates a lesser impact on flow patterns; (5) the system is capable of order millimeter vertical resolution and two centimeter horizontal resolution.




CHAPTER 4
FIELD MEASUREMENTS OF BEDFORMS AND A COMPARISON OF PREDICTION TECHNIQUES
Introduction
Field observations of the seabed indicate that bedforms are generally present on the inner shelf and near-shore regions. These bedforms can range from vortex ripples, 5 to 40 cm in length, to megaripples with lengths reaching over 1 meter. Small bedforms such as vortex ripples directly impact the structure of the wave boundary layer and thus influence sediment transport processes. Previous studies have observed that under similar flow conditions suspended sediment was limited to the lowest 5 to 10 cm when small scale ripples were not present, and suspensions of 30 to 40 cm were observed when ripples were present (Vincent et al., 1991).
Knowledge of bedforms is necessary to accurately describe the complete sediment transport process. Thus the prediction of bedform geometry is necessary for accurate sediment transport prediction. The major problem with ripple prediction has been the lack of quality field ripple measurements on which to base the models. Due to recent improvements in instrumentation, a large set of field ripple data has been recorded and is presented herein.
A new ultrasonic instrument, the SeaTek multiple transducer array (MTA), was used to measure bedform profiles in the field with millimeter vertical resolution. This instrument proved capable of measuring bedform profiles on temporal and spatial scales that previously were unobtainable. This chapter will present a comparison of existing




ripple predictive models to field data recorded with the MTA and to improve ripple prediction capabilities with a simple new model.
Instrumentation and experimental setup
The first data set to be presented herein was recorded from August 23 to 25, 1995 and will be referred to as the SIS95 data set. The second data set, S1S96, was recorded from October 29 to November 1, 1996. All of these measurements were recorded in Duck, North Carolina at the Army Corps of Engineers Field Research Facility. The Sensor Insertion System (SIS) was used to deploy the instruments off of the research pier in both cases.
The SIS95 bedform data set was recorded using a newly developed 37 element Multiple Transducer Array (MTA), as described in the previous chapter. Bedform profiles were recorded every six seconds during the 13 minute data runs. Other instruments deployed during the experiment include a pressure sensor, a bi-directional electromagnetic current meter, an underwater video camera, and an acoustic concentration profiler. Sediment samples from the bed surface were also obtained with a clam-shell type sediment sampler. The data from all of the instruments was relayed back to a single data logger, which was cabled to a shore based computer system.
Due to the mobility of the Sensor Insertion System (SIS), measurements at many different cross-shore locations were possible. Measurements were made on the beach face, in the inner trough, near the inner bar, and offshore of the bar, as shown in figure 5a. Median sediment sizes ranged from 0.18 to 1.66 mm, water depths ranged from 2 to 7 meters, and Hm. wave heights ranged from 0.2 to 0.9 meters over the course of the experiment. The range in mobility number during this experiment was 2 to 71, with mobility number defined as in equation 2.3.




E 5. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

0 0
10O0 200 300 400 500 600
(b)
(D 5 . . . . . .. . .. .. .. ... . . . . . .. . . .
100 200 300 400 500 600
distance (in)
Figure 5. Cross-shore beach profiles taken during the SIS95 (a) and SIS96 (b) experiments. Vertical bars indicate locations of ripple measurements.
The S1S96 bedform measurements were made with a commercially available 64 element SeaTek MTA. This 2.5 meter long array was composed of three smaller arrays. The center MTA was 50 cm long and consisted of 32, 5 MHz transducers with a spacing of 1.5 cm. This center section was very similar to the single MTA used in the SIS95 experiment. The center MTA was flanked on both sides by 100 cm long arrays consisting of 16, 2 MHz, transducers with 6 cm spacing, as shown in figure 6. The SIS96 SeaTek MTA recorded one bedform profile every 2 seconds. This system allowed for the measurement of small scale bedforms with the center MTA and large scale bedforms with lengths reaching over 1 meter with the whole array.




16 32 1o1

6.0 cm
center to center

1.5 cm
center to center

I< 238.5 cm
total length of measured profile
Figure 6. SeaTek 64 element MTA used in SIS96 experiment.

6.0 cm
center to center

acoustic doppler velocimeter pressure sensor

scanning sonar

Figure 7. SIS96 Instrument setup as viewed from the offshore direction.
The MTA range to the seabed for most runs was -40 cm.

1 16

16 3211




The SIS96 instrument suite consisted of the MTAs, two acoustic concentration profilers (5 and 2.25 MHz) Mesotech model 810 echo sounders, a rotating scanning sonar (2.25 MHz) manufactured by Mesotech, an under water video camera, a pressure sensor, and an Sontek acoustic doppler velocimeter (ADV). The length of data runs for the SIS96 data set ranged from 13 to 32 minutes. The instrument setup is shown in figure 7.
The SIS96 measurements were also made at several different cross-shore locations, as shown in figure 5b. Median sediment sizes ranged from 0.12 to 0.21 mm, depths ranged from 1.4 to 7 meters, and Hm. wave heights ranged from 0.32 to 1.2 meters. These conditions resulted in a range in mobility numbers from 26 to 256. The SIS96 experiment encountered much more energetic flows than the SIS95 experiment. See appendix for measured hydrodynamic and ripple data.
Data Analysis
The presence of suspended sediment can cause false returns and missed returns in the recorded MITA scans. Because of this, representative profiles were determined from several scans of data. Generally, 10 raw profiles were used to determine a single representative profile. For each transducer, the modal distance to the bed was found from the 10 scans. This distance was then recorded as the range to the bed for that transducer. Such analysis was carried out for every transducer in order to determine the whole representative profile for every 20 seconds of data.




(a)
300 ...
E
2 4 6 8 10 12 14 16
time (min.)
(b)
C 0
15
01m0
100 05 10115 120 125 13013 14
horzonal istnc (c), di. horwoalr m
The rih vetia axsidcteiieinintsfrth.rfls




(a)
80
7 0 . . . . . ... . . . . .. . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . ..
- 6 0 .. .. ... ... ....... .... ..... ..... ... .. ........... ...... .. ........ .. ...... .. ... .....
E 5 0 ....... ..... .. ...... .. ....... ...... .. ..... .. ...... ... .. ... .. ..... .. ...... ..... .. .... .....
4 0 . . . .. . . . . . . .. ... . .... . . . . . . . . . ..
3 0 .. ..... .... .. .. .... ...... .... ... .. ...
0
E
2 0 .... .... ..... ... ...
I fill
0
2 4 6 8 10 12 14 16
time (min.)

0.5
0
0.5
1!

horizontal (cm)

10 min. 0 min.

110 115 120 125 130
horizontal distance (cm), + dir. shoreward

Figure 9. Corresponding time series of (a) mobility number and (b) and (c) ripple profiles from SIS96 run 8.




100
$ 80
E
c 60
*40
E
20
. 0.5.
0,
-0.5,
C)
_.)

2 4 6 8 10 12
time (min.)
(b)

.4. 16 8.. 14 16 18

horizontal (cm)

100 105 110 115 120 125 130 135 140
horizontal distance (cm), + dir. shoreward
Figure 10. Corresponding Time series of (a) mobility number and (b) and (c) ripple profiles from the SIS96 experiment for run 26.

t1J '1 l I1 IL

10 min.
0 min.




These representative profiles were then analyzed to find ripple height and length, using a threshold method. First, any linear trend present in the profiles was removed. The removal of the linear trend was to remove the beach slope from the profile data and any settling of the SIS during the run. Settling of the SIS was observed in several of the runs, especially during the first few minutes of the runs. All points of inflection in the profile were then found. The average distance between each minimum and maximum with a vertical separation of at least 3 mm was recorded. A threshold set to one half of this average value was then used to determine the ripple height. Only minimum and maximum range points separated by at least this threshold value were used. Ripple height was found by taking the average vertical separation of these minimum and maximum ranges over the entire profile. Ripple length was found by taking the average of the horizontal separations corresponding to the same maximum and minimum ranges.
For the model comparisons, each run was divided into four sections to show the range of ripple conditions observed during each run. For several of the SIS96 runs the ripple geometry changed over the course of the run. Ripple migration, flattening, and reformation were observed during several of the runs.
An example of ripple flattening and reformation is shown in figure 8. The mobility number time series is plotted in figure 8a. Ripple destruction is seen in figure 8b during periods of high flow intensity, minutes 8 through 9 in the time series. Ripple reformation is observed during periods of lower flow intensity, minutes 10 through 14 in the time series. Figures 9 and 10 are of data recorded under less intense flow conditions. For these runs the bedform profiles remained relatively constant.
It should be noted that many recorded bedformn profiles contained both small scale and large scale bedforms. For the SIS96 experiment where large scale bedforms were measured with the 2.5 meter long array, 14 of the 28 runs had both small and large scale bedforms, 11 of the runs had only small scale ripples, and 3 runs had large scale bedforms exclusively because conditions were too energetic for the existence of small scale




bedforms. The ripple data presented herein is of small scale bedforms with lengths of less than 40 cm. This corresponds to ripples having lengths less than approximately 2000d50. For all of the SIS95 and SIS96 runs, ripples with lengths of less than 40 cm were observed except for a few very energetic flows when their existence was not expected. The large scale ripples measured during the SIS96 experiment with lengths of over 40 cm had rounded crests like megaripples (Clifton, 1976). Measurements and characteristics of these large scale ripples are addressed in chapter 5.
Ripple Model Descriptions
Nielsen (1981) field model
Two semi-empirical ripple prediction models are compared to the SIS95 and SIS96 ripple data sets. The Nielsen (1981) ripple prediction models for field conditions are shown. These models predict ripple height (r) and length (X) using the near bottom semiexcursion (A) and the mobility number (V/). Nondimensional ripple height is expressed as
7 = 2 _1.85 for (ip>10) equation 4.1
A
- = 0.275 0.022t0.5 for (yf<10) equation 4.2
A
and nondimensional ripple length is expressed as
X = exp 693-037In8 equation 4.3
A 10000+ 0.75-In7 )ea
where the mobility number (yf) is as defined in equation 2.3. The near-bottom orbital velocity can be directly substituted for the (Ao) term in the calculation of mobility number.




The Nielsen (1981) expression for ripple steepness
7 = 0.342 0.34V-25 equation 4.4
is in terms of the grain roughness Shields parameter (02.5). Which is defined in equation
2.1.
Wiberg and Harris (1994)
Wiberg and Harris (1994) classified bedforms according to the ratio of the near bed orbital diameter (do=2A) and anorbital ripple height (do /ilano), which is an approximation of the ratio of wave boundary layer thickness to the ripple height (&h1l). From this ratio the ripples were classified as orbital, anorbital, or suborbital by the following criteria:
Table 1: Wiberg and Harris (1994) ripple classification
flow conditions ripple classification
do/lao< 20 orbital ripples
20 do/rao> 100 anorbital ripples
Clifton (1976) classified orbital ripples as ripples with lengths proportional to orbital diameter and anorbital ripples as ripples with lengths that are not proportional to orbital diameter. Wiberg and Harris (1994) found that orbital ripple length and height can be reasonably represented as a fraction of the near bed orbital diameter,
2orb = 0.62do equation 4.5
and that orbital ripple steepness remains roughly constant at
(rl/k)orb = 0. 17 equation 4.6
From these two equations orbital ripple height can be found directly as the product of orbital ripple length and steepness.




Wiberg and Harris (1994) found anorbital ripple length to be a function of only grain size (d50), giving;
X.. = 535 d50 equation 4.7
Previous studies such as Nielsen (1981) and Grant and Madsen (1982), had found ripple steepness to be a function of non-dimensional bed shear stress; however, Wiberg and Harris (1994) found that anorbital ripple steepness can be defined in terms of (do/). This allows the calculation of anorbital ripple height without the calculation of bed shear stress, which eliminates some of the complications and uncertainties that arise from the computation of bed shear stress. Wiberg and Harris (1994) found by comparing ripple steepness (1/X) to non-dimensional orbital diameter (do IT) the relationship:
7 = exp -0.095 In-) + 0.4421n--- 2.28 equation 4.8
for do /1>10; if do/nq<10, qX=O.17.
Wiberg and Harris (1994) found that orbital ripple lengths are a function of orbital diameter and that anorbital ripple lengths are a function of grain diameter. By definition, suborbital ripples have ripple lengths that fall between these two limits. Thus a weighted geometric average of the bounding values of X1ano and %orb was used to determine suborbital ripple length (Xub).
In (d "- In 100]
exb /=exp L (in rA a,,- l r A,,r) + 1n A,, equation 4.9
, ep In 20- In 100 oIbIor -I




Model Comparisons
Nielsen (1981)
The Nielsen (1981) model curves for ripple height, length, and steepness are shown in figure 11 along with measured ripple dimensions from the SIS95 and SIS96 experiments. It is seen that the Nielsen (1981) ripple height model over-predicted heights at low mobility numbers (y<30) and performed better at higher mobility numbers. The model tended to under-predict ripple heights for V>100. It should be noted that only the data from the 5 MHz MTA is included. Thus only the small scale ripples are represented here.
The Nielsen (1981) ripple length model over-predicted ripple lengths at low mobility numbers (W<15), and tended to under-predict at high mobility numbers (W'>100). This is shown in Figure 1 b. Ripple steepness was over-estimated by the Nielsen (1981) irregular wave ripple steepness model for almost every run. The trend present in the steepness data was correctly captured with this model, however flattening of ripples was observed at lower than predicted Shields parameters, as shown in figure 1 Ic.
(a) (b) (c)
.1
0.1 1
00 a
4F C:X
0.001 0.01
10 100 10 100 0.1 1
mobility number mobility number Shields parameter
Figure 11. Measured (a) Nondimensional ripple height (ri/A) and (b) nondimensional ripple length (X/A) versus Mobility Number, and (c) ripple steepness versus Shields Parameter for SIS95 (*) and SIS96 (o) data with Nielsen (1981) model curves.




(a) (b) (c)
EZ E U .)._ "
100 t 00
00
.1 1 10 10 100 0.1
measured height (cm) measured length (cm) measured steepness
Figure 12. Nielsen (1981) predicted versus measured S1S95(*) and S1S96(o) ripple (a) height, (b) length, and (c) steepness.
Wiberg and Harris (1994)
The Wiberg and Harris (1994) ripple model curves are plotted with the measured SIS95 and SIS96 ripple data in figure 13. In figure 13a, ripple length data is compared to both the orbital and anorbital model curves. One problem encountered with the Wiberg and Harris (1994) model is the misclassification of ripples. Using the classification scheme outlined in Table 1, with calculated anorbital ripple height (Tano), the 30 runs of SIS95 data were classified as 21 orbital, 1 anorbital, and 8 suborbital. When measured ripple heights were used the runs were classified as 6 orbital, 1 anorbital, and 23 suborbital. Thus for 15 of the 30 runs the ripples were incorrectly classified. For the SIS96 data set a more energetic environment was encountered. The SIS96 data set consists of 28 runs, 1 was classified as orbital, 25 as anorbital, and 2 as suborbital when calculated values for ripple height were used. When the measured values for ripple height were used, 0 were classified as orbital, 25 were classified as anorbital, and 3 were classified as suborbital. Thus only 1 of the 28 runs was misclassified for the SIS96 data set.




(a) (b) (C)
10000 I0
0.1 0.1
LO 1000. go
o,
0.01 96 0.01I
1000 10000 10 100 1000 10 1 00 1000
do/d50 do/height (measured) do/height (calculated)
Figure 13. Measured SIS95 (*) and SIS96 (o) ripple data with Wiberg and Harris (1994) model curves, orbital (dashed) and anorbital (solid). (a) ripple length versus Orbital diameter, (b) and (c) steepness versus Orbital diameter / ripple height, (b) contains the measured ripple height where as (c) contains the predicted ripple height.
Figure 13a shows underprediction of ripple length for much of the anorbital ripple length data. It is also shown in figure 14b that the Wiberg and Harris (1994) model overpredicted ripple length for most of the SIS95 data. This was partly due to the misclassification of 15 of the 30 runs as orbital ripples, thus predicting large ripple lengths that are proportional to orbital diameter.
Measured ripple height is compared to predicted ripple height in figure 14a. It is shown that for most of the runs ripple height is overpredicted. The Wiberg and Harris (1994) ripple height model was more accurate in predicting ripple height for the SIS96 data set than for the SIS95 data set. This was mainly due to the ability of the model to correctly classify most of the more energetic SIS96 environment as conditions for anorbital ripples.
The Wiberg and Harris (1994) ripple models overpredicted ripple steepness for the majority of the SIS95 and SIS96 measurements. This was due to the overprediction of ripple height for the SIS95 data set and the underprediction of ripple length for the SIS96 data set.




(a)
10
measured height (cm)

E
o100
C:
.2 10
*0 ") 0,.

(b)
C,,
47
0.1
10 100
measured length (cm)

(c)
0
0 (
0.1
measured steepness

Figure 14. Wiberg and Harris (1994) predicted versus measured SIS95(*) and SIS96(o) ripple (a) height, (b) length, and (c) steepness.
The relative error (A) between measured and predicted values can be defined as;

A=exp n (ln(u)- ln(u))2

equation 4.10

where "u" is the measured value and "i" is the predicted value. This quantity is a multiplicative factor that indicates the variation about the predicted value. For example if A equals 1.34, the average error is equal to 34 percent.
Table 2. Relative error between measured and predicted ripple geometries.
Nielsen (1981) field Wiberg and Harris (1994)
ripple height 2.19 2.34
ripple length 2.02 1.93
ripple steepness 1.75 1.89
The Wiberg and Harris (1994) ripple model and the Nielsen (1981) irregular wave, or field, ripple model performed similarly for the SIS95 and SIS96 data sets. As shown in Table 2, the relative error for ripple height is 2.34 and 2.19 for the Wiberg and Harris




(1994) and Nielsen (1981) ripple models, respectively. Both models performed similarly at predicting ripple length and had relative errors of 1.93 and 2.02 for the Wiberg and Harris (1994) and the Nielsen (1981) models, respectively. The models did better at predicting ripple steepness than ripple height or ripple length independently. The Wiberg and Harris (1994) ripple model had a relative error of 1.89 whereas the Nielsen (1981) model had a relative error of 1.75 in predicting ripple steepness.
The Grant and Madsen (1982) ripple model was also compared to the SIS95 and SIS96 data set. This model was developed primarily from the Carstens et al. (1969) laboratory ripple measurements. The Grant and Madsen (1982) model produced large errors when compared to the new field data sets. The relative errors were 4.23, 3.23, and 1.63 for ripple height, length, and steepness respectively. Because of the large errors, this model was not included in the rest of this analysis.
Using near bottom hydrodynamics in predicting ripple geometries
For the model comparisons described earlier, H,,,,, and the peak wave period were determined from a surface elevation spectrum. A power spectrum was constructed from either the pressure time series or the cross-shore velocity time series. This spectrum was then transposed, using linear wave theory, into a surface elevation spectrum. H,,,, was then calculated from the area under this spectrum, and the peak period was calculated from the peak frequency of this spectrum.
One problem with this method is that the transfer function from pressure at a given depth to surface elevation amplifies the higher frequency component more than the lower frequency component. If the surface elevation spectrum is then used to determine the peak frequency, which later is to be used to determine near-bottom orbital diameter, there is a potential to use a peak frequency that does not contain the most energy in the nearbottom flows. This can be shown by correcting a measured pressure time series for a




surface elevation spectrum and also for a near bottom orbital excursion spectrum. The lower frequencies are attenuated less than the higher frequencies in the transformation to near-bottom orbital excursion. Figure 15a is a surface elevation spectrum for a SIS95 run and figure 15b is a near-bottom orbital excursion spectrum from the same pressure time series. Clearly the peak frequencies are different for these two spectra. The transfer function used to determine the bottom orbital excursion spectrum from the surface elevation spectrum shown in figure 15 is the linear wave transformation of I/sinh 2 (kh).
For much of the SIS95 data there was a low frequency swell from a hurricane located off-shore and a locally wind generated wave component. When the peak frequency was taken as the peak in the surface elevation spectrum for this data, the local wind generated wave frequency was recorded as the peak frequency for most of the runs. However, when the near-bottom orbital excursion spectrum was used, for most of the cases the peak frequency was from the lower frequency swell component. Because of this problem, another method was developed to determine the spatial and temporal scales of the nearbottom flow used in the prediction of bedform geometries.
(a) (b)
0.35.
0.25
0.3
0.25- 0.2
U)
0.2- 0.15
E 0.15 E 0.1
0.1
0.05
0.05
0 0.1 0.2 0.3 0.4 0 0.1 0 2 0.3 0.4
freq., Hz freq., Hz
Figure 15. Power spectrums for (a) corrected surface elevation and (b) near-bottom orbital excursion for run 6 of the SIS95 data set.




First, a Fourier transform was performed on either the pressure or the cross-shore velocity time series. It is noted that for the SIS95 data the pressure time series was used, since problems with drift were evident in many of the current meter time series. For the SIS96 data analysis the current meter time series were used for all but three of the runs. During these three runs the current meter was exposed in the wave troughs. The resultant frequency spectrums were transposed to bottom velocity spectrums using linear wave theory. All energy at frequencies lower than 2iT/25, corresponding to a 25 second period, was removed. This was to remove all infragravity components, since the present models are based on gravity wave frequencies. An inverse Fourier transform was then performed on the spectrum, resulting in a time series of near bottom orbital velocity.
All zero upcrossings were then found for this near-bottom orbital velocity time series. The maximum absolute value of velocity was then recorded between each upcrossing. An average of the highest one-third of these maximum velocities was then recorded as the significant near-bottom orbital velocity.
In order to find the significant near-bottom orbital diameter, the near-bottom velocity time series was integrated over time. With the low frequency component removed from the near-bottom velocity time series, the near-bed orbital excursion is about a zero mean. Without filtering the low frequency component, the integrated near-bed velocity is dominated by low frequency oscillations with periods on the order of several minutes. A similar procedure was performed on this near-bottom excursion time series to determine significant values. All zero upcrossings were found, and the maximum (positive) and minimum (negative) excursions from the mean were recorded between each upcrossing. The differences between the maximum and minimum excursions for each upcrossing were recorded as near-bottom orbital diameters. The average of the largest one-third nearbottom orbital diameters was recorded as the significant near-bottom orbital diameter for the time series.




For many of the runs the hydrodynamics and the bedform geometry varied significantly over the course of a run, presumably due to the presence of wave groups. Because of this, each run was divided into four sections. Four values of significant nearbottom orbital diameter and significant near-bottom velocity were computed for each run. These correspond to the same time intervals that the four values for ripple height, length, and steepness were calculated. Figure 16a shows a comparison of orbital diameters calculated from Hmo and from the near bottom significant method for all of the SIS95 and SIS96 runs. Figure 16b is a similar plot showing a comparison of mobility numbers calculated from the two different methods.
(a)
" 88o ~
5 10 15 20 25 ( 35 40 45 50 55
. . I . .!. . .!. . . . . . . . . . . 6 . . . . I.. . .
i250 }
.0:
"1 ..50 ................... . . .....................
W 0
E 50 0
= 1 00.- ~ ~ ~ ~ 0 . . .! . . ... .! . . . . i . . . o &" ... . . : . ... ' . . 3
5 10 15 20 25 30 35 40 45 50 55
Run number
Figure 16. Orbital diameter versus run number (a) and mobility number versus run number
(b) for near-bottom significant values (o) and Hmo derived values (*) for the SIS95 (run 1 to 30) and SIS96 (run 31 to 58) data sets.
The Wiberg and Harris (1994) and the Nielsen (1981) ripple models were then used with significant values for near-bottom velocity and orbital diameter to predict ripple geometry. In figure 17, the Nielsen (1981) model curves are compared to measured parameters with the hydrodynamics described as above. It is seen that for low values of




Mobility number, (y< 20) both nondimensional ripple height (i/A) and nondimensional ripple length (X/A) are overpredicted by the Nielsen (1981) model. Whereas for large values of mobility number (Vr>100) both properties are underpredicted by this model.
(a) (b) (c)
11 0.1
0.01
C0
o 1 0
0.01 ci
0.001 0.01
10 100 10 100 0.1 1
mobility number mobility number Shields parameter
Figure 17. Measured (a) Nondimensional ripple height (rI/A), (b) nondimensional ripple length (X/A), and (c) ripple steepness for SIS95 (*) and SIS96 (o) data with Nielsen (1981) model curves with Mobility Number, Shields parameter, and semi-excursion calculated from significant near-bottom orbital velocity and orbital diameter.
Figure 18 shows the Wiberg and Harris (1994) model curves compared to the significant near-bottom conditions. The model results from the significant near-bottom conditions are very similar to those in the previous section. For most of the SIS95 data set, ripple heights and lengths are overpredicted, and for most of the SIS96 data ripple lengths are underpredicted. The primary reason for the consistent discrepancy in the SIS95 data set is still the misclassification of ripple type. The Wiberg and Harris (1994) classification scheme misclassified 40% of the ripples in the SIS95 data set and 3% in the SIS96 data set. This gives a total misclassification of 22% for both data sets. Most of the problem was in classifying suborbital ripples as orbital ripples. Ripple length was then




based on a proportion of orbital diameter, and ripple steepness was taken as a constant. This then led to the overprediction of both ripple length and height.
(a) (b) (c)
00.1
1000
_ecb Cb .
0.01, 0.01,
1000 10000 10 100 1000 10 100 1000
do/d50 do/height (measured) do/height (calculated)
Figure 18. Measured SIS95 (*) and SIS96 (o) ripple data with Wiberg and Harris (1994) model curves, orbital (dashed) and anorbital (solid) for significant near-bottom orbital diameter data. (a) ripple length versus orbital diameter, (b) and (c) steepness versus orbital diameter / ripple height, (b) contains the measured ripple height where as (c) contains the predicted ripple height.
The Nielsen (1981) ripple height and length models produced larger errors when significant bottom conditions were used than when Hmo derived conditions were used. This was mainly due to the higher mobility numbers associated with the significant bottom velocities than those derived from Hmo and peak period. These larger mobility numbers produced even larger errors in predicting ripple height and length at mobility numbers greater than 100. The tendency of underprediction at high mobility numbers can be seen in figure 17.
The Wiberg and Harris (1994) ripple models produced greater errors in predicting ripple geometry when the near-bottom significant orbital diameter and velocity were used than when Hmo and peak period were used to determine orbital diameter and velocity. This was because for most of the SIS95 data set the significant orbital diameter was larger




than the calculated orbital diameter from Hmo. Since many of the ripples were misclassified, ripple lengths and heights were based on proportions of orbital diameter. For most of the SIS95 data this led to even higher overpredictions. For the SIS96 data, both the significant method and the Hmo method performed similarly. The relative errors associated with the predicted ripple geometries when significant near-bottom conditions are used are shown in Table 3.
Table 3. Relative error between measured and predicted ripple geometries when significant near-bottom conditions are used as input.
Nielsen (1981) field Wiberg and Harris (1994)
ripple height 3.48 2.98
ripple length 2.14 2.12
ripple steepness 1.84 1.94
New Model Development
As shown in the previous figures, the Wiberg and Harris (1994) and Nielsen (1981) ripple models did not correctly capture all of the trends in the SIS95 and SIS96 data. For this reason, we are motivated to develop new ripple prediction formulae. The new formulae are based solely on mobility number and orbital semi-excursion, which are found by using the near-bottom significant values described previously.
Non-dimensional ripple height (ri/A) is described as a function of mobility number (Y) as follows:
77= 0.09 for (V<10) equation 4.11
A




77=09ffor (ijc10) equation 4.12
A
A plot of (ri/A) versus mobility number for the SIS95 and SIS96 data with this curve is shown in figure 19a. The relative error in predicting ripple height with this model is 1.53.
Ripple steepness was also found to be a function of mobility number. The SIS95 and SIS96 data were best described by
77-=-0006V0. +0.15 for it'<190 equation 4.13
77- 01 for yf l9O equation 4.14
2,
This formula was developed by minimizing the relative error with the SIS experimental data and is shown in figure 19b along with the SIS95 and SIS96 data sets. The relative error between predicted and measured values of ripple steepness for the SIS data sets was 1.52 for this formula. It is noted that for mobility numbers greater than 190 a constant steepness of 0.01 is predicted. Since ripple length can be computed by dividing ripple height by ripple steepness, the lower limit of ripple steepness was taken as 0.01 so as to provide reasonable estimates for ripple lengths at high mobility numbers.
The three data points that fall well below the model curves at very low mobility numbers are for sediment with a median diameter of 1.66 mmi. The sediment from the inner trough region where these measurements were made had a bimodal distribution, with one mode located around 1.6 mm and the other around 0.2 mm. Diver observations from this region indicate that the larger sediment was located at the ripple crests and the smaller sediment at the ripple troughs. Since the median value for sediment size is used in this analysis, the smaller sediment present in this particular distribution is not represented in the calculation of mobility number.




(a) (b)
10
1010
1010 12 110
Mobility~ ~~ NubrMbliyNme
S19() ndSSM o )iit daaNihuhmnwboelcrvs Mobility Number ansei
excursion are calculated from the significant near-bottom orbital velocity and significant orbital diameter.
Ripple length can be determined by dividing ripple height by ripple steepness. The relative error in predicting ripple length for the S1S95 and SIS96 data sets was 1.55. Nielsen (1981) observed different trends in ripple steepness versus mobility number for sediments of different densities. It is noted that all of the field data used to construct this model is from areas where the sediment was composed primarily of quartz sand. Comparison of new model with previous field ripple measurements
The relationship between the SIS95 and SIS96 data sets and previous field ripple measurements is shown in figure 20. Only small scale field ripple data with ripple lengths of less than 40 cm were included. This was to reduce the chance of representing megaripple data in this small scale ripple analysis. There was a total of 70 measurements that met these criteria. In the determination of relative errors, flatbed conditions were not considered because artificial errors would be induced when height, length, and steepness values were assigned to these conditions. The flatbed conditions are included in figure 20b, and are plotted as a steepness of 0.01.




It is noted that only the Dingler (1974) data was recorded with ultrasonic instrumentation, and that the other field data sets were recorded through diver observations. It is the authors' experience that the measurement of ripple height while diving is subject to large uncertainties. Some of the variation in observed trends between diver collected data and the data that was recorded with ultrasonic instrumentation could be due to diver measurement errors. The new method agrees well with the Dingler (1974) data, however both ripple height and ripple steepness are underpredicted for the Inman (1957) data at low Mobility Numbers and for some of the Nielsen (1984) data. Some of the Nielsen (1984) and Inman (1957) runs contain data with lengths between 30 and 40 cm. Thus some of the disagreement between the curves and the outlying data points may be due to the inclusion of larger scale bedform data even though a cutoff of 40 cm on ripple length was applied to the data.
(a) (b)
(1)
++
10 +
0 +
101 102 10 1 10,
Mobility Number Mobility Number
Figure 20. Nondimensional ripple height! orbital sem-i-excursion (a) and ripple steepness
(b) versus Mobility Number with new model curves and all available field ripple data. Included are previous field measurements from Inman (1957), Dingler (1974), and Nielsen (1984) (+), SIS95 (*) and SIS96 (o).




Table 4. Relative error between measured and predicted ripple geometries for all previous small scale field data.
New Wiberg and Harris Nielsen field
Model (1994) (1981)
ripple height 1.82 1.65 2.12
ripple steepness 1.62 1.46 1.68
ripple length 2.04 1.30 1.50
In table four the relative error between measured and predicted ripple geometries are shown for the new model, the Wiberg and Harris (1994) model, and the Nielsen (1981) field model. All available small scale field ripple data measured previous to the SIS95 and SIS96 experiments were used in the calculation of these relative errors. These include the field measurements of Inman (1957), Dingler (1974) and Nielsen (1984). It is shown that the new model performed similarly to the other models in predicting ripple height, and that the Wiberg and Harris (1994) model gave the best predictions. All of the models performed similarly in predicting ripple steepness however the Wiberg and Harris (1994) model did slightly better than the other two. Ripple length was better predicted by the Wiberg and Harris (1994) and Nielsen (1981) than it was with the new model for these data sets. It should be noted that the Wiberg and Harris (1994) model was constructed with the data shown, and that the Nielsen (1981) was constructed with the data shown that was collected prior to 1981. Also, only a single wave-height and wave period were given for each of the measured data points in the previous data sets. Thus significant near-bottom conditions could not be determined for these data sets.




Table 5. Relative error between measured and predicted ripple geometries for all available small scale field data; including the SIS95 and SIS96 data sets.
New Wiberg and Harris Nielsen field
Model (1994) (1981)
ripple height 1.60 2.42 2.33
ripple steepness 1.54 1.84 1.70
ripple length 1.68 1.96 2.01
Relative errors between measured and predicted ripple dimensions were calculated for the complete field data set. This data set includes the S1S95, S1S96, and the previously collected data sets discussed above. Table five contains these relative error values. The new model did much better at predicting ripple height than did the other two models. The average error was 60% for the new model whereas errors of 142% and 133% were calculated for the Wiberg and Harris (1994) and the Nielsen (1981) models, respectively. The new model also had better agreement than either of the other two models at predicting ripple steepness and length.
Conclusions
A multiple transducer array was used to remotely measure bedforms during the SIS95 and SIS96 experiments in Duck, North Carolina. Existing methods for predicting smallscale ripple geometry did not capture all of the trends found in this new data set. A new method for predicting ripple geometry is described. This method does not require the calculation of bed shear stress or the classification of ripple regimes as previous models did. This new method relates ripple height and steepness to near bottom mobility number by a few simple relations.




54
All available field data were used to test the new model and the Nielsen (1981) and Wiberg and Harris (1994) models. The new model did substantially better at predicting ripple height, with average errors of 60% compared to 142% and 133% for the Wiberg and Harris (1994) and the Nielsen (1981) models respectively. The new model also gave better estimates of ripple steepness and length. When only the new S1S95 and SIS96 data sets were used, average errors of approximately 50% were observed in predicting ripple height, steepness, and length with the new ripple prediction method.




CHAPTER 5
OBSERVATIONS OF BEDFORM DYNAMICS
Introduction
Small scale vortex ripples and larger scale megaripples populate much of the nearshore coastal environment. The geometry of these bedforms is important to sediment transport processes. Bedforms directly affect the bottom roughness and thus influence the bottom shear stress, the profile of turbulence, and the rate of sediment transport for given wave and current conditions (Vongvissessomjai, 1984).
The dynamics of bedforms are also important in predicting sediment transport. Ripple migration can account for a significant part of the transported sediment under certain wave conditions, (Osbourne and Vincent, 1993, Hay and Bowen, 1993). Ripple flattening and reformation is also important in the prediction of sediment transport. If ripple destruction under wave groups is not accounted for, the predicted sediment transport rate could differ substantially from the actual transport rate. Observations have shown that under many flow conditions the ripples constantly change to match the flow. These changes can occur on time scales as short as a single wave period (Dingler and Inman, 1976).
The purpose of this chapter is to present observations of small scale ripples and megaripples from a field experiment. Ripple migration, flattening, and reformation is demonstrated for a quartz sand seabed in the near-shore environment. Comparisons of ripple behavior to the hydrodynamic forcing are also presented along with a brief description of the instrumentation.




Experimental Setup
A field experiment investigating sediment transport processes was conducted at the U. S. Army Corps of Engineers Field Research Facility in Duck, North Carolina from October 30 to November 2, 1996. The Sensor Insertion System (SIS) on the research pier was used for instrument deployment. The SIS is a movable system that allows measurements to be made at a multitude of cross-shore locations. The measurements presented herein span a variety of water depths, wave heights, and sediment conditions, as summarized in Appendix A. A cross-shore profile measured during this experiment is shown in Figure 5.
Many instruments were deployed during this experiment. To measure bedforms, a SeaTek multiple transducer array (MTA), a rotating scanning fan beam sonar (RSS), and an underwater video camera were deployed. Two acoustic backscatter sensors and an optical backscatter sensor were deployed to measure suspended sediment and turbidity. An acoustic doppler velocimeter and a strain type pressure sensor were used to record hydrodynamic data. All of these instruments, except for the RSS, were controlled via a data logger located in an underwater package. Once the data was recorded by the data logger, the data was downloaded to a shore-based system and archived.
The SeaTek multiple transducer array (MTA) used in this experiment consists of three sections. The center section is 0.5 meter in length and composed of a 32 element linear array of 5 MHz transducers. The center to center spacing of these transducers is 1.5 cm. The acoustic footprint at 0.5 meter range is approximately 2.5 cm. The system is capable of 1 millimeter vertical resolution under ideal conditions. The center array is flanked on both sides by 1.0 meter arrays. These longer arrays operate at 2 MHz and contain 16 transducers with 6 cm spacing. The acoustic footprint at a range of 0.5 meter is approximately 4.5 cm for these arrays. All three sections were attached together to form a continuous 64 element array with a length of 2.5 meters. This 2.5 meter array is capable of




measuring large scale megaripples, with all 64 transducers, and small scale vortex ripples with the center, 32 element, section. Figure 3 is a schematic of the SeaTek MITA used in this experiment.
Each MTA has its own pressure housing that contains a data logger and the electronics necessary to run the transducers. Each of the MTAs log an entire run of data and then offload this data at the end of the run to the main data logger in the under water package. The data from all of the instruments is then stored in a single file and offloaded to shore.
Several MTA parameters can be changed remotely from a shore-based computer. These include the receiver threshold, blanking period, scan rate, and number of scans to record during each run. The MTAs only record a single range measurement for each transducer per scan. The return signal is run through a comparitor. When the return signal voltage meets a certain prescribed threshold, the time delay from transmit is recorded. This time for the reflected ultrasonic signal to return back to the transducer is then used to determine the distance to the bed. Blanking is used to reduce the number of false returns. Blanking is the time delay, after the transmit pulse is sent, before the comparitor starts to monitor the return signal for a bottom return. The maximum scan rate of the MTA was one profile every two seconds. The MTAs are equipped with time varying gain (TVG). Time varying gain is used to compensate for spherical spreading and water attenuation losses in the acoustic signal.
During periods of high sediment suspension some false returns were recorded with the MITA. For this reason, some of the data presented herein is from profiles representing 20 seconds of data. There are also examples of raw data presented so that time scales on the order of wave periods can be studied.
A rotating scanning sonar equipped with a fan beam transducer was also deployed. The sonar operated at 2.25 MHz and was mounted on a stepper motor with 0.225 degree increments. The width of the fan beam in the far field is nominally 0.9 degrees in the




horizontal and 30 degrees in the vertical. Planform images of the seabed within a radius of 5 meters from the sonar were recorded on video tape. The instrument was able to detect large and medium scale bedforms in regions where the transmitted signal propagated nearly perpendicular to the ripple crests. Large scale ripple orientation and rough estimates of migration rate are possible with such a system (Hay and Wilson, 1994).
A monochrome low light underwater video camera was also mounted next to the MTA. The primary purpose of this camera was to document small scale ripple orientation. During many of the runs, visibility permitted video recording of small scale ripples and visual observation of sediment processes near the bed.
An acoustic doppler velocimeter (ADV) and a pressure transducer were used to measure the current and wave conditions. For most of the data presented herein the nearbottom orbital velocity and orbital excursion time-series were derived from the ADV time series. A fast Fourier transform (FFT) was first performed on the cross-shore velocity time series. This velocity spectrum was then transposed to a bottom velocity spectrum using the linear theory correction factor, k.
cosh k(h + z) equation 6.1
sinh kh
An inverse fast Fourier transform (IFFT) was then performed, resulting in the depth corrected bottom time series for orbital velocity. This time series was then integrated to determine a near-bottom orbital excursion time series.
A time series of mobility number, defined as
u(t)u(t)l equation 6.2
((s -)gd5
can then be found from the time series of cross-shore near-bottom orbital velocity, u(t). Mobility number has been shown by several authors to be an important parameter in bedform dynamics (Dingler and Inman, 1976, Nielsen 1981).




Ripple Flattening and Reformation
For several of the SIS96 runs, small scale ripple flattening and reformation were observed. Megaripples exhibit length scales on the order of one meter and do not have regular vortex formation every half-wave cycle (Nielsen, 1992). Megaripple crests are generally more rounded than vortex ripple crests, and can sometimes be superimposed with vortex ripples (Nielsen, 1992). Megaripples are referred to herein as large scale ripples.
Observations of bedform dynamics from the SIS96 experiment indicate that ripple flattening and reformation is related to mobility number. During periods of high peak mobility numbers, ripples tend to be flattened. The reformation of these ripples occurs within a range of mobility numbers that are large enough to move sediment but not so large as to cause flattening.
Figure 21 a is a plot of the mobility number time-series and figure 21lb is a time-series plot of ripple height for run 2. In this case the ripple heights were determined from the spatial standard deviation time-series for the detrended ripple profiles. The ripple heights were calculated by multiplying the standard deviation by 2.83, which is the multiplier used to obtain height from standard deviation for sinusoidal curves. Figure 22 shows the bedform profile time-series from the center 5 MHz MTA for run 2. Mobility numbers reached values greater than 200 near the 3 minute point of the run. During this period the small scale ripples were flattened, which is shown by the decrease in ripple height during this period in figure 21b. This is also shown in figure 22 and was observed on the underwater video. Once the wave energy decreased, the ripples reformed during minute 4. The ripple profile remained relatively constant during minutes 5 through 8, with no significant migration or growth. During the event between minutes 8 and 9 when peak mobility numbers briefly reached values greater than 150, some decrease in ripple height was observed. The large mobility numbers during minutes 11I and 12 flattened most of the




ripples shown in figure 22. This flattening is also shown in figure 21b by the decrease in ripple height from 0.6 to 0.35 cm. Peak mobility numbers reached values of over 200 during this one minute event. The ripples reformed during the minute following the event. Even though mobility numbers reached values of nearly 200 during the remainder of this run, the ripples were not flattened. This is most likely due to the duration of the event. Only one or two waves created mobility numbers near 200 on this occasion. It appears several waves with mobility numbers larger than 150 are required for significant flattening of small scale ripples.
(a)
2 0 0 ........... .......... ..... ..... ........
E
0
. . .. .. . . . . . . . . . . . . . ... . . . . . . . . . . ... . . . . . .. . . . . .
-200 .... .. I I I I
E 0 2 4 6 (b) 8 10 12 14 16
E
P0.5
CL I
0
0 2 4 6 8 10 12 14 16
time (min.)
Figure 21. Time series of mobility number (a) and ripple height (b) for run 2.
Figure 22b is a plot of the raw MTA data for minutes 8 through 15 of run 2. The time between profiles is 2 seconds. Light shades correspond to higher bed elevations than do darker shades. False returns from suspended sediment can be seen as light areas during periods of significant sediment suspension. Such events occurred at minutes 9, 10 and 11.5. The ripples were flattened during the event near minute 12. When the ripples were flattened, the seabed was located very close to the mean elevation of the previously rippled bed. It should be noted that during some periods of high suspended sediment concentrations, the MTA recorded false returns from the suspended sediment. Most likely




the true bottom returns, shown after the suspension events, were measured after a significant portion of the suspended sediment settled out of suspension.
For this run, complete ripple reformation occurred during the minute after the ripples were flattened. The peak mobility numbers during minute 13 were near 100. Such mobility numbers seem to be conducive for fast recovery of flattened bedforms. Such flows are strong enough to readily move sediment, however they are not too strong so as to impede ripple formation or cause flattening.

(a)

time (min) 0 100
(b)

120 110
horizontal (cm)

100 105 110 115 120 125 130 135 140
horizontal (cm)
Figure 22. Small scale bedform profiles for run 2. Mesh plot of bedform profiles (a) with time on the y-axis and elevation in centimeters as indicated by the colorbar. Raw MTA bedform profiles for minutes 8 through 15 (b). The time between scans is 2 seconds. Range to the seabed is in centimeters as indicated by the colorbar. Time is plotted on the y-axis and horizontal distance is plotted on the x-axis.
The time-series for mobility number and ripple height are shown in figures 23a and
23b for run number 14. Bedform profiles are shown in figure 24. It is shown in figure

0.2
0
-0.2
-0.4
-0.6
-31
-32

t
12
E "10




23b that for the first 1.5 minutes the ripples increased in height from 0.5 to 0.75 cm. During the second minute peak mobility numbers briefly reached values over 200. This resulted in a reduction of ripple height to 0.4 cm that quickly recovered to 0.62 cm over the next minute. The ripple profile remained relatively constant until minute 8. During minute 8, the peak mobility numbers reached values of over 200 on several occasions, and remained above 150 for a large portion of the minute. The bedforms were almost completely flattened by this episode. The ripple height, as determined from the spatial standard deviation, for this period was 0.2 cm. The heights of the small scale bedforms remained very small during the period of low mobility numbers through minutes 10 and 11. The peak mobility numbers of near 100 around minutes 12 through 14 resulted in growth of the ripples. Another wave group with a peak mobility number of 200 during minute 15 resulted in a decrease in ripple height. It is noteworthy that the data from the NITA and the video data indicate a nearly flat bed during minutes 9 through 11, and that the ripples begin to reform at their previous locations during minute 12.
Figure 24b is a plot of the raw MTA data for minutes 10 through 15 of run 14. Rapid reformation of the previously flattened ripples is shown in this figure at minute 13. This corresponds to the end of the wave group seen in the mobility number plot between minutes I I and 13. Reformation of the ripples occurred within 20 seconds after the passing of the wave group. Continual growth in ripple height was observed from minutes 12 through 14 for the ripples with heights of 0.5 cm and lengths of I I cm. It is noted that the bedslope has not been removed from the data plotted in figure 24b. The elevation of the bed is increasing in the onshore, positive Y, direction.




- 0 2 4 6 8 10 12 14 16
time (min.)
Figure 23. Time series of mobility number (a) and ripple height (b) for run 14

0120
110
0 100 horizontal (cm)

100 105 110 115 120 125 130 135 140
horizontal (cm)

Figure 24. Small scale bedform profiles for run 14. Mesh plot of bedform profiles (a) with time on the y-axis and elevation as indicated by the colorbar in centimeters. Raw MTA bedform profiles for minutes 10 through 15 (b). The time between scans is 2 seconds. Range to the seabed is in centimeters as indicated by the colorbar. Time is plotted on the y-axis and horizontal distance is plotted on the x-axis.

E 05
0.5
15

time (min)

E- 13
E

0.2
0
-0.2
-0.4
-53
-54
-55
-56




Ripple flattening and reformation were observed during run 20 both with the MTA and under-water video. For most of the other runs, when conditions were energetic enough for ripple flattening, visibility was not sufficient for under-water video viewing of the bedforms. The mobility number time-series for run 20 is plotted in figure 25a and ripple height as determined from the standard deviation is plotted in figure 25b. For several periods during this run the mobility number exceeded values of 200. The small scale bedforms recorded with the 5 MHz MTA are shown in figure 26. It is noted that the small scale ripples shown are superimposed on a migrating megaripple that will be addressed later. Because of this, the ripple heights determined by the standard deviation contain some of the variance due to this larger megaripple and do not accurately depict the heights of only the small scale ripples.
Ripple flattening was observed during the wave groups at minutes 4 and 8. At around the 4.5 minute mark the peak mobility number exceeded 200 during several waves, and ripple flattening was recorded with both the video and MTA. At the 5 minute mark in figure 26b, it can be seen that the measured profile is approximately flat. Again, the white areas in this figure are false returns from suspended sediment. Figure 27 is a compilation of several images from the under-water video recordings during this run. Figure 27a is an image recorded just after most of the suspended sediment settled out of suspension at minute 5:00. This image indicates that the seabed is essentially flat since shadows from ripples are not seen. Figure 27b is an image recorded 35 seconds later. Ripples can be seen in this image. Figure 27c was recorded 1 m-inute and 10 seconds after the first image. Again, small scale ripples can be clearly seen in this image. The MTA profiles were recorded on a transect parallel with the horizontal axis in figure 27.




8
time (min.)

Figure 25. Time series of mobility number (a) and ripple height (b) for run 20.
(a)

time (min) 0 100
horizontal (cm)
(b)

0.2
0
-0.2
-0.4
-39
-39.5
-40

100 105 110 115 120 125 130 135 140
horizontal (cm)

Figure 26. Small scale bedform profiles for run 20. Mesh plot of bedform profiles (a) with time on the y-axis, and elevation in centimeters as indicated by the colorbar. Raw MTA bedform profiles for minutes 4 through 10 (b). The time between scans is 2 seconds. Range to the seabed is in centimeters as indicated by the colorbar. Time is plotted on the y-axis and horizontal distance is plotted on the x-axis.

0 0 . .
15




Figure 27. Under-water video images from Run 20 for time 5:00 (a), 5:35 (b), and 6: 10
(c). Ripple reformation over the course of one minute is shown.
The flattening of small scale vortex ripples appears to be a function of mobility number. Peak mobility numbers greater than approximately 150 can potentially flatten ripples. Dingler and Inman (1976) suggested that a single wave with a mobility number of over 240 can flatten a rippled bed. The SIS96 bedform observations support this finding. For mobility numbers less than approximately 240 and greater than approximately 150, several waves are required to completely flatten the ripples. Higher mobility numbers require fewer waves to flatten ripples than do lower mobility numbers.




Ripple reformation can occur within a minute after flattening, and in certain conditions within a couple of wave periods. Once the peak mobility numbers decrease to values below approximately 150, the ripples tend to reform. If the peak mobility numbers are too low, the ripples will not reform until appropriate mobility numbers are reached. For most of the data presented, it appears that if the peak mobility numbers remain below approximately 50, ripple reformation is slow. Once mobility numbers of greater than 50 and less than 150 are reached and maintained, the ripples appear to reform rapidly, often within one minute.
Most of the ripple destruction and reformation observed show ripple reformation in roughly the same position as the previous ripple. For many runs slight ripple remnants were observed. This would logically lead to reformation in the same location. However, for other runs an apparently smooth bed was observed with MTA profiles and underwater video recordings. Even during some of these cases the ripples reformed in the same locations. This possibly could be due to some remnant ripples outside of the measurement area or ripple remnants of less than 2 or 3 mm leading to ripple repopulation in the same locations. Since the vertical resolution of the MTA is decreased during periods of high suspended concentration, remnants of less than 2 or 3 mm would not be seen under such flows.
Ripple migration
Large Scale Migration
Migration of large scale ripples, or megaripples. was observed with the complete 64 element MTA. These bedforms; had lengths of greater than 40 cm, and heights of 3 to 10 cm. During runs 19 through 21, megaripple migration was observed. Runs 19, 20, and 21 were recorded over an 85 minute period. The profiles from these runs are plotted in figure 29. Run 19, shown by the lowest group of curves, spans minutes 1 to 16. Run 20




is shown in the center group of curves and spans minutes 50 to 66. Run 21 is shown at the top of figure 29 and spans minutes 69 to 85. During these runs the megaripple crest in the center of the profiles migrated approximately 50 cm. This corresponds to an average migration rate of 0.59 cm/min. The height of this bedform ranged from 4 to 6 cm and the length, as measured from a cross-shore transect, ranged from 100 to 130 cm during the course of these runs.
The mobility number time series for runs 19, 20 and 21 are shown in figure 28. During run 20 the wave energy increased, and the mobility number reached values of over 300. The steepness of the megaripple crest decreased during this period. Run 21 had several waves that resulted in mobility numbers of approximately 200. The megaripple in figure 29 remained similar in shape during runs 20 and 21. The forward face of the megaripple migrated approximately 18 cm shoreward during run 20. This compares to 10 cm for run 19 and 8 cm for run 21, when the waves were less energetic.
(a)
2 0 0 . . . . . ... . . .. . . . . . . . . . .. . . . . . . . . . . . . .. -. . . . . . . . . . .
0
-200 . .
2 4 6 8(b) 10 12 14 16
2 0 0 L i i l ... . . . .. .. . . . . .. . . . .. .... . . . . ... . . . . . . . . . . . . . . .. .
20...................... ....... ........... ......
- 2 0 0 . . . .... .. . . 11
2 4 6 8(c) 10 12 14 16
-o 2 0 0 . .... . . . . ... . . . . .. . . . . . ... . . . . ... . . . . . . .. . . . . . . . . . ..
" -200
E
0
o 2 0 0 .. . . .. . I .. . . .. . . .. . . . . . .
2 4 6 8 10 12 14 16
time (min.)
Figure 28. Time series of mobility number for runs 19 (a), 20 (b), and 21 (c).




1 2 ........................................................ . / ............... 8 5 m in .
1 .................. ..................... ......... . ..... ........
.. . .
.0 16 min.
0 . .... . ... . . . . . . . . . . . . . . . .
0 min.
0 50 100 150 200
horizontal (cm), + onshore
Figure 29. Bedform profiles for runs 19, 20, and 21. One minute separation between profiles, each profile is offset by +2 mm. Time for the profiles is recorded on the right vertical axis.
The megarripples measured with the MTA during runs 19, 20, and 21 were uniform and long crested. Planform images of these megaripples were recorded with the rotating scanning sonar (RSS). An image recorded at the beginning of run 19 is shown in figure 30. The returns from the instrument frame and MTA are outlined in white boxes. The returns from the upslope of the megaripples are shown as white lines extending from the upper right to the lower left of the figure.




Figure 30. Rotating scanning sonar (RSS) image from the beginning of run 19. The white boxes enclose returns from the MITA (top) and instrument mounting arms (lower). Light patches running from upper right to lower left correspond to returns from ripple crests of large scale bedforms. Concentric circles indicate range in one meter increments. The onshore direction is towards the right in the figure.
Ripple migration may account for a significant portion of the total sediment transport for certain wave and current conditions. The migration of the megaripple shown in the center of figure 29 represents a cross-shore transport rate per unit length of beach of 144.7 g cm~f1 hr This transport rate was calculated by first finding the area under the ripple crest, Ar, for the starting and ending profile of runs 19 and 21, respectively. The center of mass for both of these areas was then found. The cross-shore velocity of the ripple, c, was determined by the horizontal displacement of the center of mass divided by the time between profiles. The length of the ripple, X, was measured as the cross-shore




distance between troughs. The porosity of the sediment, 0, was taken as 0.35 for medium sand. The transport rate associated with the migration of the megripple per unit length of beach was then calculated as

(1 )Arc
Q, = A

equation 6.3

with Ar = 357 cm2, c=23.1 cm/hr, and 1=100 cm.
To estimate bedload transport, a Meyer-Peter (1948) type formula was used with a slope modification as suggested by Fresdsoe (1978). This type of bedload formula was designed for unidirectional flows. However an adapted instantaneous version can be used as an indicator for bedload transport under oscillatory flows (Nielsen, 1992). The timeseries of mobility number and Swart's (1974) friction factor were used to determine the time series of the grain roughness Shields parameter, 02.5(t). The nondimensional bedload transport, 4(t), is expressed as:

O~h ozz(t)) 625(t dI(t) = 8 2.5 (t) -0.05 0.1 dl 02. 5 02.5
dx 02.5 (t) 102.5(01

equation 6.4

Sh 02.5(t)
025., (t)[ > 0.05 + 0.1
dx 102.5(t)

equation 6.5

Sh 62.5(t for 02.5(t) < 0.05+ 0.1 2(t)
dx 02.5 (t)

equation 6.6

where the bedload transport rate, Qb(t), is expressed as

Qb(t) = D(t)dso50 (s 1)gdso

equation 6.7




A positive 02.5(t) represents a flow in the on-shore (positive x) direction. A positive bottom slope (dh/dx) corresponds to an increase in bed elevation in the on-shore direction.
The average value of the predicted bedload transport rate, Qb(t), was then calculated for runs 19, 20, and 21 to be 47, 68, and 10 g cm-l hr-1 respectively. These compare to a transport rate of 144.7 g cn- hrl calculated from the megaripple migration over the course of these runs. It is seen that the sediment transport due to the migration of this megaripple exceeds estimates of bedload transport. Previous authors have suggested that the cross-shore bedload transport can be accounted for in the migration of meter-scale bedforms (Hay and Bowen, 1993). These calculations are consistent with this statement. The bedload transport calculations presented herein are an estimate of the bedload transport rate. Generally such calculations are only used as an order of magnitude estimate of transport.
Megaripple migration was also observed in run 9. The time series for mobility number is shown in figure 31a, and the bedform profiles are plotted in figure 3lb. The peak mobility numbers were frequently over 150 and on several occasions during the 16 minute run the mobility numbers were near 250. These measurements were made very near the breakpoint. Because of the highly energetic conditions, no small scale bedforms were present. Two megaripple crests were recorded with the MTA for this run. The heights of the megaripples were approximately 4.5 to 5.5 cm and the cross-shore lengths were approximately 1 meter. The megaripples shown in figure 14b migrated approximately 7 cm during the 16 minute run. This corresponds to a migration rate of 0.44 cm/min.




(a)
- 2 0 0 .. ........... .... ............. ............. ..........
1 0 .. . . . . -. . .... . . .. . . .
Ei
o0 10 0 . . ......... .. . ...... ......
E -0
-200 .... ................ ....... ............. .............
1 2 3 4 5 6 7
time (min.)
(b)
oE 8 ..... ......... .. .: .................. .. ... ... ............ ......!
6 ... ... ... .... .. .i ............. ................... 1 5 m in .
.2
2. 4.. ... ...
0
I I
0 50 100 150 200
horizontal distance (cm), + dir. shoreward
Figure 31. Mobility number time-series for run 9, (a), and bedform time-series (b) showing megaripple migration. Time between profiles is 30 seconds, and each profile is offset by +2 mm. Time is on the right vertical axis.
The megaripple migration of the center crest in run 9 represents a sediment transport rate of 137 g cm1 hr1. This was calculated using the method described above with an average area of the ripple crest cross-section of 330 cm2 and a shoreward horizontal translation of the center of mass of this area of 6.7 cm over a 16.7 minute period. The cross-shore length of the ripple was taken as 100 cm. Using equations 6.4, 6.5, and 6.6, the estimated bedload transport rate for this run was 53 g cm1 hr'. Sediment transport due to megaripple migration during run 9 was approximately 2.5 times larger than the bedload transport estimate for this run. Again, the estimation of bedload transport presented should not be taken as the exact value of bedload transport but only as an order of magnitude estimate.




Small Scale Migration
During several of the runs small scale, or vortex, ripple migration was observed with the 5 MHz MTA. The best indicator for ripple migration found for this data set was the estimated bedload transport rate as predicted by equation 6.6. Equation 6.6 is not only a function of the flow intensity near the bed, but also of flow asymmetry. Ripple migration was generally observed during relatively high mobility number flows with asymmetry between the shoreward and seaward flows. For all instances of migration during this experiment the migration was in the shoreward direction, which was the direction of predicted bedload transport as predicted by equation 6.6.
Figure 32 contains the mobility number time-series and the ripple height time series for run 15. Figure 33 is the corresponding bedform profile time-series for run 15. The average ripple height was 0.4 cm and the average ripple length was 12 cm for this run. Ripple migration was observed over the first 8.5 minutes of run 15. The ripples migrated approximately 4.5 cm during this period which results in an average migration rate of 0.53 cm/min. The calculated average rate of bedload transport was 30.4 g cm'l hr-1 for this period. From minutes 8.5 to 13 the average rate of bedload transport was calculated as 3.0 g cm'f hr-1. During this period no significant migration was observed and the ripples gradually increased in height from 0.4 to 0.55 cm. A wave group producing a peak mobility number of over 300 resulted in ripple flattening at minute 14.
Figure 33a is a three-dimensional depiction of the two-dimensional bedform profiles over time. Time is plotted on the y-axis with horizontal distance and elevation on the x and y axis. Figure 33b is a surface plot of the same data. The ripple crests are the light areas and the troughs are the darker areas. Elevations are defined by the color bar to the right of the plot in centimeters. Ripple migration and flattening are observed in this plot.




10 12 14 16

2 4 6 8
time (min.)

11
10 12 14 16

Figure 32. Time series of mobility number (a) and ripple height (b) for run 15.
(a)

time (min)

100 105

120 125 130 135 140

horizontal (cm)

0.5
0
-0.5
-1

100 1U5 11U 115 12U 125 13U 135 14U horizontal (cm)

Figure 33. Small scale bedform profiles for run 15. Mesh plot of bedform profiles (a) with time on the y-axis. Surface plot of bedform profiles (b) with time on the y-axis, horizontal distance on the x-axis, and vertical elevation as indicated by the colorbar in centimeters.

0
E a-)
cO.5"

0_ 0

I I I I I




Figure 34 contains the mobility number and ripple height time-series for run 16. Figure 35 shows the bedform profile time-series for this same run. The ripples shown migrated an average of 6 cm during the first 8 minutes of run 16. This corresponds to an average migration rate of 0.75 cmmn. Most of this migration occurred during the first 4.5 minutes of the run when the estimated average bedload transport rate was 21.3 g cm7 hr-1. During minutes 4.5 to 7, when the average bedload transport was calculated as 9.7 g cm'l hr'1, little migration occurred. The ripples present during this period had average heights of 0.6 cm and lengths of 9 cm. Ripple height continued to build for the first 7.5 minutes of this run from 0.4 to 0.7 cm. The ripples were then flattened by a group of waves at the 7.5 minute mark. The maximum mobility number during this wave group was 216.
Ripple reformation was not observed until minute 12.5. This probably is due to the destructive effects of the intermittent waves between minutes 9 and 12 that resulted in mobility numbers of near 150. The ripples reformed within a minute during the calmer conditions between minutes 12.5 and 13. No significant ripple migration was observed during minutes 13 to 16. During this period the calculated average bedload transport rate was 19 g cm' hr'.
(a)
. ~ 2 0 0 . . . .. . . .. . . .. . . .. . . .. . . .. .. . . . .
E 0
E- 0 b 0 1 41
E
-50.5
0
~- 0 2 4 6 8 10 12 14 16
time (min.)
Figure 34. Time series of mobility number (a) and ripple height (b) for run 16.




(a)
E 0 .5 ...... .... .... ... '.
-0.5,
15 '
10
5 14"0 '
time (min) 0 100 105 110 115 120 125 130 135
horizontal (cm)
(b)
15 !0.5
10_ 0
E E
S5 -0.5
0-1
100 105 110 115 120 125 130 135 140
horizontal (cm)
Figure 35. Small scale bedform profiles for run 16. Mesh plot of bedform profiles (a) with time on the y-axis. Surface plot of bedform profiles (b) with time on the y-axis, horizontal distance on the x-axis, and vertical elevation as indicated by the colorbar in centimeters.
During a few of the runs, no ripple migration was seen even during high mobility number flows. An example of this is run 3, which is shown in figures 36 and 37. Single events with mobility numbers of over 150 caused some flattening throughout the run, however no significant migration was observed. The estimated average bedload transport rate, from equation 6.6, for this run was 5.2 g cm1 hr". Even during the relatively energetic first four minutes of the run the estimated bedload transport rate was 7 g cm-' hr". This low value indicates little asymmetry in the near-bottom orbital flows. This is also the case for run 2, which is plotted in figures 21 and 22. Both of these runs were in depths of over 5 meters, thus the orbital velocities were more symmetric than those measured in more shallow water.




- 0 2 4 6 (b) 8 10 12 14
E
C)
..c1
"=02 4 68 10 12 14

time (min.)

16 16

Figure 36. Time series of mobility number (a) and ripple height (b) for run 3.
(a)

10 ..
time (min) 0 100 105 110 115
horizontal (cm)
(b)
15(
.10
E
E
5 -

100 105 110 115 120 125 130 135 140
horizontal (cm)

Figure 37. Small scale bedform profiles for run 3. Mesh plot of bedform profiles (a) with time on the y-axis. Surface plot of bedform profiles (b) with time on the y-axis, horizontal distance on the x-axis, and vertical elevation as indicated by the colorbar in centimeters.




Figure 38 is a time series of bedform profiles from run 19. It should be noted that the small scale bedforms in figure 38 are superimposed on the crest of a larger bedform that is migrating shoreward. This larger scale migration was discussed in the previous section and is shown in figure 29. The mobility number time-series for run 19 is plotted in figure 28a.

E 1 ..... ........ ..../'*75
, 0 .: ......
1590

. . .. . . ." '
110 115" 120 125 130 135 0
0100 105 10 15
horizontal (cm)

time (min)

. . . . . . . . . . . ..... .. . . . .. .

15 min. 10 min.
5 min.
0 min.

135 140

100 105 110 115 120 125 130
horizontal distance (cm), + dir. shoreward

Figure 38. Small scale bedform profiles for run 19. Mesh plot of bedform profiles (a) with time on the yaxis. Bedform profiles offset to show development (b). The first profile of the run is plotted as the lower curve, subsequent profiles are offset by +0.2 cm. The time separation between profiles is 20 seconds. The right vertical axis indicates time in minutes for the profiles.




Run 19 was chosen because an apparent change in ripple migration rate was observed during this run. The solid line intersecting the profiles in figure 38b roughly corresponds to the crest location of a single ripple as it ingrates shoreward. The elapsed time in minutes from the beginning of the run is shown on the right vertical axis next to the appropriate profiles. This time corresponds to the time shown in figure 28 of the mobility number time series.
Figure 39 is an image of the seabed from the under water video taken at the beginning of run 19. Under water video recorded during this run indicate a brick-pattern ripple formation with many bifurcations in the ripple crests. The 5 MHz NITA was oriented so that profiles were recorded along a line parallel to the horizontal axis in figure 39. When wave groups were present large suspension events would result. The sediment would then settle out of suspension between wave groups. On video, the ripples appear to become flat after these suspension events and then rebuild between wave groups, such as those shown for run 20 in figure 27. The average small scale ripple height and length during this run were 0.7 cm and 12 cm, respectively.
Figure 39. Underwater video image from the beginning of run 19 showing brick-pattern ripples.




It is shown by the line following a single ripple crest in figure 38b that the ripple migration rate varied over the course of the run. The estimated rate of bedload transport, from equation 6.6, was 39 g cm' hr-1 for minutes 1.5 to 6. The ripple crest marked in figure 38 migrated 10 cm during this 4.5 minute period. This corresponds to an average migration rate of 2.2 cm/min.
During minutes 6 to 8.5, the estimated bedload transport rate was 13.2 g cml hr1 and no significant migration was observed. A single wave group resulted in mobility numbers of over 100 for minutes 8.5 to 9. This flattened the ripples and caused an apparent shoreward shift of 2 cm in the ripple crests over this 30 second period. Such a shift would correspond to a 4 cm/min migration, however the bedform profile time series does not indicate a coherent ripple migration during this period, but rather a flattening and reformation.
No significant migration was observed during minutes 9 to 11.5. The estimated bedload transport rate for this period was 16 g cm-1 hr-l. During minutes 12 to 13.5 significant ripple migration was observed, and the estimated rate of bedload transport was 110 g cm' hr-1. The marked ripple in figure 11 migrated approximately 8 cm during this 1.5 minute period. This corresponds to a migration rate of 5.3 cm/min. From minutes 13.5 to 14.5 no significant ripple migration was observed. During this period the estimated average rate of bedload transport was 17 g cm' hr'. Peak mobility numbers were relatively high, between 150 and 200, for the remainder of the run. During this period, minutes 14.5 to 17, the bedforms were flattened. Shoreward migration appears to be the case, however it is difficult to determine a migration rate for this energetic period.
Since these ripples are superimposed on the crest of a megaripple, contraction of flow might have resulted in lower mobility number flows producing a larger than expected impact on the small scale ripples. Ripple flattening was not observed for mobility numbers of near 100 in the any of the other runs. Also, migration rates of over 5 cm/min were not observed during any other runs. Such high migration rates could in part be due to the




migration of the underlying megaripple. The mechanics of ripple migration, especially when the ripple is superimposed on a migrating megaripple, are not well understood.
Previous authors have suggested that mean bottom currents can result in ripple migration (Dingler and Inman, 1976). It is noted that the average distance from the seabed to the velocity meter was over 1.8 m for this experiment. The maximum measured mean cross-shore current for the SIS96 experiment was 0.03 m/s. Since the actual nearbottom velocities were not measured, it is difficult to compare ripple migration with mean bottom currents. No correlation was seen between measured mean currents, measured over one meter off of the seabed, and ripple migration for this experiment.
The small scale ripple migration observed during the SIS96 experiment supports the use of the estimated bedload transport, by equation 6.6, as an indicator for small scale ripple migration. For the SIS96 experiment, ripple migration was observed when the estimated rate of bedload transport exceeded a value of approximately 20 g crnf1 hr. The dimensional bedload transport rate in equation 6.6 had better agreement with observations of ripple migration than did the non-dimensional bedload transport rate shown in equations 6.4 and 6.5. A distinct threshold for the existence of ripple migration could not be found for the SIS96 data when the non-dimensional bedload transport rate was used as an indicator. The dimensional bedload transport rate contains an additional d50 to the 1.5 power term. It appears, at least for this data set, that this term has importance in determining ripple migration.
Model Comparisons
Model comparisons of the small scale ripple measurements from the SIS96 experiment were presented in chapter 5. Megaripple measurements made with the whole MTA are presented in this chapter. For some of the runs, the megaripple crests were not normal to the MTA. For these runs the RSS was used to measure the angle between the




MTA and the megaripple crests. This angle was then used to correct the ripple lengths measured with the MTA.
Observations of megaripples indicate that the spacing between megaripple crests, ripple length, is proportional to orbital diameter. Ripples with lengths proportional to orbital diameter have previously been referred to as orbital ripples (Clifton, 1976). Wiberg and Harris (1994) proposed that orbital ripple length, Xorb, is proportional to orbital diameter, d,, by the relation
?,orb=0.62 d. equation 6.8
and that orbital ripple steepness is equal to a constant value of 0.17. This follows from Miller and Komar (1980) who proposed a multiplier of 0.65 instead of 0.62.
Wiberg and Harris (1994) adopt the approximate criteria proposed by Clifton and Dingler (1984) that orbital ripples occur when the ratio of orbital diameter to grain diameter, do d50, is less than 2000. During the SIS96 experiment ripples with length scales proportional to orbital diameter were observed in conditions too energetic for the existence of vortex ripples, such as in run 9 shown in figure 14. Megaripples with length scales proportional to orbital diameter were measured even when significant mobility numbers of over 200 and djd50 values of over 10000 were observed.
Megaripple measurements are shown in figure 40 as '+'s and small scale ripples are shown as 'o's with the Wiberg and Harris (1994) orbital ripple model curves plotted as the solid lines. Measurements of megaripple lengths agree well with the Wiberg and Harris (1994) orbital ripple length formula. However, measured megaripple steepnesses were generally below their predicted value of 0.17 for orbital ripples. The dashed line in figure 40c is at a constant steepness of 0.065. This was the average value of megaripple steepness for the SIS96 data set. It should be noted that megaripples were considered present if the steepness of the bedforms from the whole 2.5 meter MTA were larger than




0.03. In figure 40a nondimensional ripple lengths are plotted versus nondimensional orbital diameter. The solid curve is the Wiberg and Harris (1994) orbital ripple curve. The dashed line was calculated using equation 6.7 for orbital ripple length and a constant steepness value of 0.065.

1 UUU
+
+ +
100.-- + 0
10 .- o o D o o of o oo
10o a 0 0
- 00 0

2000

oI
10 0
:1000

0.1
)0.01
0.01

2000

4000

4000

10000

10000

do/d50

Figure 40. Measured SIS96 small scale (o) and large scale (+) ripple data versus nondimensional orbital diameter for nondimensional ripple height (a), length (b), and steepness (c). The Wiberg and Harris (1994) orbital ripple curves are shown as solid lines, new suggested curves are shown as dashed lines.

0o 0
0 0




It is shown in figure 40 that the coexistence of both small scale vortex ripples and large scale megaripples is possible at many values of dj dso. It appears that at low values of dj d50so length and height scales of the small and large scale bedforms begin to merge. The region where small scale bedforms begin to scale with the near bottom orbital diameter has generally been referred to as the orbital ripple range. These measurements show that orbital ripples can exist at both large and small values of dj ds50.
(a)

00 (b) do/d50 10000
++ +++ + + + + + V+ + ++++ ++ + + +
S+ + + + ++ +

100
(c) mobility number

t
0.05 0.1 0.2
0.05 0.1 0.2

mean current (m/s)
Figure 41. Megaripple steepness versus nondimensional orbital diameter, d/ d50so, (a), mobility number (b), and mean net current measured 1.5 meters above the bed (c) for the SIS96 data set.

0.1
CD
0
CL U)
I)
0.01
20
S0.1
a)
a)
U)
0.01

+ 4+ ++~ ++ i++++T1++ t+ + ++
.+: +-+++++ +
+T ++ + + A-+ + + ++ +

O 0.1
00 U) U) 00
0.01




As shown in figure 41, megaripples were observed at the full range of orbital diameters, mobility numbers, and mean current values encountered during the SIS96 data set. For many of the runs megaripples were not present. For these runs the steepness of the bedforms measured with the whole MTA were generally less than 0.03. When the bedform steepness was less than 0.03, megaripples were said not to exist. The dashed lines in figure 41 represent a steepness of 0.03. The data below this line is said to not contain megaripples and was not included in the previous analysis of megaripple dimensions.
It should be noted that the steepnesses determined from the whole MTA are always greater than zero, even when no large scale bedforms were present. This is mainly due to the aliasing of small scale ripples in the 2 MHz MTA profiles. Since transducer spacing on the 2 MHz MTA is 6 cm, it is possible to alias small bedforms with lengths less than 12 cm into the 2 MHz MTA profile.
No significant correlations between flow and sediment parameters were found for the existence of megaripples. Mean cross-shore and long-shore currents, median sediment diameter, local bedslope, orbital diameter, mobility number, grain roughness Shields parameter, and bedload transport rates were all compared to the data and did not result in any relation for the existence of megaripples.
Conclusions
A 64 element multiple transducer array (MTA) is capable of high-resolution measurements of small and large scale bedform profiles. Temporal scales on the order of wave periods can be measured as well as the migration, flattening, and rebuilding of smallscale bedforms with heights of less than 1 cm.
Observations from the SIS96 experiment indicate that the flattening of small-scale vortex ripples is a function of mobility number. Groups of waves with peak mobility




numbers of over approximately 150 can potentially flatten ripples. Single waves with peak mobility numbers of larger than approximately 240, which also was suggested by Dingler and Inman (1976), can flatten a rippled bed. Ripple reformation can occur within a minute after flattening, and in certain conditions within a couple of wave periods. Rapid ripple reformation appears to occur when peak mobility numbers are between 50 and 150. Ripple reformation is slow when peak mobility numbers remain less than approximately 50. When mobility numbers reach values of over approximately 150 during a rebuilding period, ripple flattening can occur. Thus mobility numbers larger than 150 can impede the reformation process.
It was observed that a significant part of the total cross-shore sediment transport may be accounted for in the migration of megaripples. Small scale ripple migration was observed when the estimated bedload transport rate was greater than approximately 20 g cm-1 hr-' as predicted by equation 6.6. Ripples appear to migrate in the same direction as indicated by the direction of the predicted bedload transport. Since a net transport predicted by equation 6.6 represents skewness in near bottom velocities, it appears that small scale ripple migration is a function of the asymmetry in near-bed orbital flows.
Megaripple lengths measured during the SIS96 experiment were roughly proportional to near-bottom orbital diameter by the relation proposed by Wiberg and Harris (1994) for orbital ripples. It has previously been proposed that these orbital ripples exist when dod50 <2000 (Clifton and Dingier, 1984). However, megaripples with lengths proportional to orbital diameter were observed at the full range of do/d0, even at values of over 10000. Measured megaripple steepness was found to be approximately 0.065.
Ripples were observed in every run during the SIS96 field experiment. Seabed measurements were made at several cross-shore locations and under many different wave conditions. Many of the measurements contained both small and large scale ripples, while others contained only small or only large scale ripples. Even in some conditions when plane-bed conditions were anticipated, small scale ripples were observed to form in the




Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E6D88H4ND_MV4L4P INGEST_TIME 2017-07-14T23:02:39Z PACKAGE UF00075478_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES