Wave and Sediment Dynamics on the Chenier Plain of Louisiana. George Paul Kemp Louisiana State University and Agricultural & Mechanical College

Louisiana State University LSU Digital Commons

LSU Historical Dissertations and Theses Graduate School

1986 Mud Deposition at the Shoreface: Wave and Sediment Dynamics on the Chenier Plain of Louisiana. George Paul Kemp Louisiana State University and Agricultural & Mechanical College

Follow this and additional works at: https://digitalcommons.lsu.edu/gradschool_disstheses

Recommended Citation Kemp, George Paul, "Mud Deposition at the Shoreface: Wave and Sediment Dynamics on the Chenier Plain of Louisiana." (1986). LSU Historical Dissertations and Theses. 4305. https://digitalcommons.lsu.edu/gradschool_disstheses/4305

This Dissertation is brought to you for free and open access by the Graduate School at LSU Digital Commons. It has been accepted for inclusion in LSU Historical Dissertations and Theses by an authorized administrator of LSU Digital Commons. For more information, please contact [email protected]. INFORMATION TO USERS

While the most advanced technology has been used to photograph and reproduce this manuscript, the quality of the reproduction is heavily dependent upon the quality of the material submitted. For example:

• Manuscript pages may have indistinct print. In such cases, the best available copy has been filmed.

• Manuscripts may not always be complete. In such cases, a note will indicate that it is not possible to obtain missing pages.

• Copyrighted material may have been removed from the manuscript. In such cases, a note will indicate the deletion.

Oversize materials (e.g., maps, drawings, and charts) are photographed by sectioning the original, beginning at the upper left-hand comer and continuing from left to right in equal sections with small overlaps. Each oversize page is also filmed as one exposure and is available, for an additional charge, as a standard 35mm slide or as a 17”x 23” black and white photographic print.

Most photographs reproduce acceptably on positive microfilm or microfiche but lack the clarity on xerographic copies made from the microfilm. For an additional charge, 35mm slides of 6”x 9” black and white photographic prints are available for any photographs or illustrations that cannot be reproduced satisfactorily by xerography.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 8710568

Kemp, George Paul

MUD DEPOSITION AT THE SHOREFACE: WAVE AND SEDIMENT DYNAMICS ON THE CHENIER PLAIN OF LOUISIANA

The Louisiana State University and Agricultural and Mechanical Col. Ph.D. 1986

University Microfilms

International300 N. Zeeb Road, Ann Arbor, Ml 48106

In all cases this material has been filmed in the best possible way from the available copy. Problems encountered with this document have been identified here with a check mark V .

1. Glossy photographs or pages.

2. Colored illustrations, paper or print ______

3. Photographs with dark background i / ^

4. Illustrations are poor cop y ______

5. Pages with black marks, not original co p y ______

6. Print shows through as there is text on both sides of page ______

7. Indistinct, broken or small print on several pages

8. Print exceeds margin requirem ents ______

9. Tightly bound cop y with print lost in sp in e ______

10. Computer printout pages with indistinct print ______

11. Page(s) ______lacking when material received, and not available from school or author.

12. Page(s) ______seem to be missing in numbering only as text follows.

13. Two pages numbered . Text follows.

14. Curling and wrinkled p a g es ______

15. Dissertation contains pages with print at a slant, filmed as received t /

16. Other ______

University Microfilms International

A Dissertation

Submitted to the Graduate Faculty of the Louisiana State University and Agricultural and Mechanical College in partial fulfillment of the requirements for the degree of Doctor of Philosophy

The Department of Marine Sciences

by G. Paul Kemp B.S., Cornell University, 1975 M.S., Louisiana State University, 1978 December 1986

GEORGE PAUL KEMP

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ACKNOWLEDGEMENTS Support for this project was provided by the Louisiana Sea Grant University Program, a part of the National Sea Grant University Program maintained by the National Oceanic and Atmospheric Administration of the U.S. Department of Commerce. Additional field support was generously furnished by the Louisiana Universities Marine Consortium (LUMCON), and the U.S. Army Corps of Engineers staff at Freshwater Bayou Canal Lock. . John Wells introduced me to the mudflat problem and supervised this research. Dag Nummedal participated actively in the field program, and was unfailing in his encouragement. William Wiseman, Jr., patiently schook i me in the art of time-series analysis. Charles Adams, Jr., gave invaluable assistance with boundary layer theory. James Coleman and James Gosselink contributed sound advice throughout an unorthodox student career. W. David Constant assisted ably as a member of the examining committee. Floyd Demers, Annie McK. Prior, and Gerald McHugh at the Coastal Studies Institute, and Mike DeTraz of LUMCON, were instrumental in bringing this project to a successful completion. Celia Harrod and Kerry Lyle provided much needed assistance with the graphics. Scott Dinnel, Ivor LI. van Heerden, Lauro Calliari, and Allison Drew deserve recognition among the many students who contributed. A special thanks is extended to my understanding colleagues at Groundwater Technology, Inc., to the Joseph Lipsey, Sr., Scholarship award committee, and to Geraldine Newman at the Department of Marine Sciences. The greatest measures of appreciation go to my parents, and to Linda Fowler, my partner in all endeavors.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE OF CONTENTS

Page ACKNOWLEDGEMENTS...... ii

LIST OF TABLES...... v

LIST OF FIGURES...... vi

ABSTRACT...... x

INTRODUCTION...... 1

PREVIOUS STUDIES...... 7

FIELD AREA ...... 14

G eology...... 16 Inner-Shelf Circulation ...... 18 Storms and Modem Mudflat Sedimentation ...... 20

DATA ACQUISITION...... 23

Sediment Survey ...... 23 Dynamics Experiment ...... 25

RESULTS...... 29

Sediment Survey ...... 29 Geomorphology ...... 29 Mudflat Sedimentation ...... 36 Mineralogy and Rheology...... 41 Dynamics Experiment...... 50 Meteorologic Influences on the Tide Record ...... 52 Waves and Oscillatory Flow ...... 52 SuspendedSediments ...... 61 Infragravity Water Level Changes and Non-oscillatory Flow...... 64

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. DISCUSSION 80

Formation and Transport of Fluid Mud ...... 85 Kinematics of Fluid Mud Deposition at the Shoreface ...... 91 Shelf Sediment Dispersal...... 100 Future Chenier Plain Progradation ...... 102 Implications for Other Coasts and the Geologic Record 103

CONCLUSIONS...... 106

BIBLIOGRAPHY...... I l l

APPENDICES...... 133

A. Beach Profiles...... 133 B. Program for Grant-Madsen Boundary Layer Closure 142

VITA ...... 148

APPROVAL SHEETS...... 149

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF TABLES

Table . Page

1. Chenier Plain Shoreline Characteristics...... 31

2. Clay Mineral Abundances ...... 46

3. Sea and Swell Wave Statistics...... 53

4. Total Suspended Sediment...... 62

5. Water Level Variance Below 0.005 cps ...... 67

6. Non-Oscillatory Onshore-Offshore Flow ...... 70

7. Steady Alongshore Flow...... 73

8. Characteristic Alongshore Velocity Profile Parameters 78

9. Grant-Madsen Boundary Layer Model Results ...... 88

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF FIGURES

Figure Page

1. Location of chenier plain study area ...... 6

2. Map of study area, showing major shoreline features and beach survey stations ...... 15

3. Cross-section of shoreface at station 6, showing deployment of sensors during cold front experiment...... 26

4. Nearshore surficial sediments and bathymetry in the study area ...... 30

5a-b. Muddy shoreface at station 7 (5a); reactivated chenier beach (station 2) at Cheniere au Tigre (5b) ...... 32

6a-b. Time-lapse views of the vegetated mudflat mapped by Morgan et al. (1953) at station 3; eroding marsh scarp, May, 1982 (6a), replaced by clastic washover beach, May, 1985 (6b) ...... 34

7. Bar migration at Tigre Point (station 4)...... 35

8. Oblique aerial view of western limb of large mud bar formed in Spring, 1982, between stations 7 and 8...... 37

9. Profile time-series of mudflat accretion and erosion at station 8, February - July, 1981 ...... 37

lOa-b. Shoreface mud at station 7; Freshly-deposited fluid mud, Spring 1982 (10a); eroding surface showing development of shore-normal flutes 5 months later (10b) ...... 38

11. Log of vibracore from marsh now covering mudflat described by Morgan et al. (1953) west of Cheniere au Tigre at station 3 ...... 40

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 12a-c. Bedding structure in radiograph positives of mudflat sequences ...... 42

13. Texture and bulk density changes with depth in upper meter of newly deposited mudflat...... 43

14. X-ray diffractograms of smears from the <1 jx fraction of mudflat sediments treated as indicated ...... 45

15. Shear stress-strain rate plot from capillary viscometer measurements on mudflat sediment 48

16. Fluid mud yield strength plotted as a function of sediment concentration (from McCave 1984)...... 49

17. Winds and tides during cold front experiment, 10-13 December, 1982 ...... 51

18. Representative energy density spectra from 20 minute pressure and flowmeter time-series for pre-frontal, frontal, and post-frontal runs ...... 55

19a-b. Concurrent spectra, and cross-spectral estimates of coherence and phase for records from run 3; between seaward and landward sensors located 10 m apart (19a), and between near-surface and near-bottom flowmeters (19b) ...... 57

20. The areas of application of the several wave theories as a function of H/d and d/L (from Komar 1976) ...... 59

21a-c. Relationships between wave parameters determined from the spectra and cross-spectra, and those predicted by shallow water LWT; wavelength (21a), and orbital velocity (21b) ...... 60

22. Mean suspended sediment concentrations from representative pre-frontal, frontal, and post- frontal runs ...... 63

vii

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 23. Relationship between vertical sediment concentration gradient and the relative heights of sea and sw ell...... 65

24. Low frequency water level and onshore-offshore flowmeter data from frontal run 3 ...... 68

25. Sequence of vertical velocity profiles during onshore- offshore flow event in run 3 ...... 71

26. Low frequency water level and alongshore flowmeter data from frontal run 5 ...... 74

27. Characteristic logarithmic alongshore velocity profiles for frontal runs 4 and 5, and post-frontal run 8 ...... 79

28. B-L-S model for sorting of cohesive and non-cohesive sediments in the boundary layer of a turbidity current (from Stow and Bowen 1980)...... 82

29. Generalized shoreface geometry, showing slopes used in determining standing wave node position, and changes in the cross-section affecting cross shore velocities...... 94

30a-b. Cross-shore trends in low-frequency flow under standing waves (30a), and bed shear stress due to waves attenuated at a constant H/d ratio of 0.2 (30b) ...... 95

A-l. Station 1 profile sequence ...... 134

A-2. Station 2 profile.sequence ...... 135

A-3. Station 3 profile.sequence ...... 136

A-4. Station 4 profile sequence ...... 137

A-5. Station 5 profile .sequence ...... 138

A-6. Station 6 profile sequence ...... 139 viii

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. A-7. Station 7 profile sequence...... 140

A-8. Station 8 profile sequence...... 141

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ABSTRACT Sedimentological data, and high-resolution current velocity time- series were acquired nearshore along the muddy coast of Louisiana's chenier plain. Results suggest that the 2 to 10 cm thick, laminated and graded silt/shell + mud beds which characterize shoreface mudflat sedimentation form through intermittent dumping of fluid mud suspensions. Cohesive and non-cohesive sediments are first separated and concentrated by shear within a wave boundary layer (WBL). Rapid sedimentation, or "freezing", of the mud fraction occurs when the yield strength of the fluid mud suspension equals that of the WBL shear stress. Incident waves were well described by shallow water linear wave theory, but exhibited the progressive nearshore attenuation and absence of breaking characteristic of muddy shorelines. An onshore decrease in wave shear stress results, which is the reverse of that on surf-dominated coasts. Fluid mud layers, characterized by significant yield strength, are proposed to form when high concentrations of suspended sediments are introduced nearshore, but are prevented from depositing by WBL shear stresses close to the bed. In cross-section, WBL sorting causes a near-bottom discontinuity in sediment concentration, and the associated vertical yield strength gradient. Expanded spatially, this point of weakness forms a plane along which the fluid mud layer can move in response to shear exerted at its upper margin by an accelerating flow. Adjustment of the steady alongshore current velocity profile to the fluid mud layer restricts coastwise sediment transport. Significant cross shore transport of fluid mud may occur, however, under the influence of unsteady flows. Such low-frequency flows were found to be associated with x

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. shore-amplified water level oscillations which have characteristics of a standing wave with an antinode at the shoreline. The onshore decreasing wave shear stress gradient leads to preferential deposition at the shoreface. Widespread renewal of chenier plain shoreline advance through mudflat accretion is predicted when the Atchafalaya delta, now forming 100 km to the east in Atchafalaya Bay, emerges on the inner shelf, and the nearshore supply of fine-grained sediment is dramatically increased.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. INTRODUCTION

The past 15 years have witnessed a growth in knowledge about silt and clay sedimentation which was recently likened to an "information explosion" (Stow and Piper 1984). An early casualty of this "explosion" was the belief that marine muds accumulate only in quiescent environments, isolated from currents exceeding a few cm-s_l. Nowhere does the "quiescent environment" paradigm contrast more with what is observed than on a significant fraction of the world's coastlines where fine-grained sediment is deposited at the beach face itself. There, mud sedimentation occurs in non- estuarine settings exposed to high-velocity wave-generated flows, in addition to significant steady currents (Dolan et al. 1972, Rine and Ginsburg 1985). In most locations it alternates spatially and temporally with the more familiar coarse-grained clastic deposition (Delaney 1963, Wells and Coleman 1981a, b). Elsewhere, it appears to occur only during the high energy conditions accompanying storms (Morgan et al. 1958, Nair 1976, Martins et al. 1979). Modem mud coasts can provide guidance in the interpretation of ancient sequences in which marine shales are observed to merge laterally or vertically into non-marine deposits without the transitional beach and tidal facies normally considered indicative of shoreline processes (Rine and Ginsburg 1985, Walker and Harms 1971,1976; Keulegan and Krumbein 1949, Moore 1929). An understanding of the hydrodynamic and oceanographic factors which give rise to any muddy shoreface would greatly improve the utility of a facies model of widespread applicability. Unlike the dynamics of littoral onshore-offshore clastic sediment transport, which have been studied for nearly a century (Cornish 1898), those pertaining to

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2

nearshore mud deposition have been largely overlooked. The investigation reported here was initiated to, at least partially, close this gap. The overall goal of this study was to identify and describe the processes by which a coast incorporates influxes of mud introduced from offshore. More specifically, I sought to develop a physical model for the onshore translation and deposition of mud under waves which could explain the distinctive bedding observed in coastal mudflat cores. Shoreline mud accumulations are commonly linked by along-isobath "advective mud streams" (McCave 1972) to river deltas. Such turbid currents restrict terrigenous sediment transfer from the continental shelf to deep water, and can deliver silt and clay to nearshore locations hundreds to thousands of kilometers distant from the original point of introduction (Delft Hydraulics Laboratory 1963, Drake 1976, Gibbs 1976). Some of the best known inner-shelf mud streams play a significant role in coastal progradation by providing a source of unconsolidated sediments that are captured for onshore transport and deposition. The mud streams of the Amazon, Yangtze, and Mississippi Rivers are responsible for extensive mudflat deposition along 1400 km of the Guiana (Allersma 1971), 800 km of the East China Sea (Meie et al. 1983), and 60 km of the Louisiana chenier plain coasts (Morgan et al. 1953, Coleman 1966, Wells and Kemp 1981), respectively. Littoral mud deposits have also been described flanking the deltas of rivers that discharge into epicontinental seas, including the Gulf of Bohai, China (Zenkovich 1967, Meie et al. 1983), the Gulf of Carpentaria, northern Australia (Rhodes 1982), the Firth of Thames, New Zealand (Woodroffe et al. 1983), and the Gulf of California (Thompson 1968). Less is known about

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 3

apparently non-deltaic coastal mudflat deposition reported adjacent to lagoon systems on the south-west coast of India (Nair 1976) and on the Rio Grande do Sul coast of southern Brazil (Delaney 1963, Martins et al. 1979). These sources document littoral mud deposition in all tide ranges and wave energy regimes. The coastal features formed exhibit a spectrum of morphologies. Ephemeral mud "arcs", spits and bars with a few square kilometers of subaerial expression have been described from Louisiana (Morgan et al. 1958, Wells and Kemp 1981). In contrast, migratory "mud banks" a hundred times larger have been tracked for decades off the coast of Surinam and the Guianas (NEDECO 1968, Rine and Ginsburg 1985). Elsewhere, coastal mud deposits have been designated only with the generic terms "flat" or "tidal flat", suggesting that features with discrete alongshore boundaries are not observed. While our knowledge of the processes important to fine-grained sedimentation under waves is growing rapidly, lessons learned in the laboratory about the complexity of this subject have led some to believe that little might be gained from field research. The singular investigation by Wells and Coleman (1981a, b) of the relationship between shallow-water waves and mud suspension on the coast of Surinam suggests, however, that such work can provide insights which might not be recognized at laboratory scales. The eastern margin of Louisiana's chenier plain coast today offers an ideal location for study of the dynamics of fine-grained sediment deposition and erosion in a wave-dominated environment (Fig. 1). This area is presently receiving an influx of mud from the Atchafalaya River, a Mississippi distributary now building a delta lobe in a bay 100 km to the east

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 4

(van Heerden et al. 1981). In some locations, fine-grained sediments accumulate near shore under conditions similar to those described by Wells and Coleman (1981a) in a process which appears to involve deposition from highly concentrated mud-water suspensions or fluid mud. Nearby, however, muddy shorelines exposed to the same wind and wave conditions are erosional, while on others, normal surf zone processes and sandy beach sedimentation prevail (Wells and Kemp 1981, Kaczorowski and Gemant 1980). A two-staged approach was adopted to meet the objectives. First, a survey of offshore bathymetry and sediment composition was combined with a 2.5 year beach profiling program to supply regional coverage of the temporal and spatial patterns of sand and mud deposition along the 30 km of coast indicated in Figure 1. Cores from marsh, mudflat, and nearshore environments were analyzed to provide information on depositional processes, as well as the mineralogy and rheology of the mud itself. Second, a dynamics experiment was conducted during the passage of a winter cold front, when a range of energy conditions, typical to this coast, could be examined and compared over a short time interval. Rapid-response flowmeters, pressure sensors, and a multi-level suspended sediment sampler were deployed in the shallow nearshore zone where mud deposition occurs. This equipment was used to record wave characteristics, and to monitor suspended sediment concentrations and flow velocities at various elevations above the bed. Evidence of major depositional events was acquired during the long term monitoring program. Although significant mud sedimentation did not take place during the four day period of the dynamics experiment, analysis of

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. the time-series records yielded new insight into the nature of nearshore fine grained sediment transport under waves.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 6

o o e CM 0 )

9. CO 0 CO > . •O 13 w o eo "co I 0 CL CM 0 ) k - 0 ‘c 0 -C O *♦—o oc CO o • o o —I o , e T~ CO 0 CD k _ o>3 'l l

c m eo

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. PREVIOUS STUDIES

Even a brief survey of the literature reveals that the mechanics which govern deposition of fine-grained sediments close to shore in the presence of waves are largely unknown (Wells and Kemp 1985). The mere presence of mud at the shoreface is troubling, as it appears to violate the consensus of nearly a century of work establishing the preferential onshore transport of the coarsest sediment fraction in clastic settings (Cornish 1898). The current lack of understanding stems, first, from the absence of a general theory relating mud deposition to fluid and sediment parameters, and, second, from a paucity of field data. I. N. McCave (1984) addressed both of these factors when he stated in a recent review that:

"there are no good ways of predicting the state of aggregation of polydisperse, multimineralic suspensions ..., thus in natural situations empiricism reigns, but based on very few measurements".

Muddy coasts pose unique difficulties for researchers not only in the theoretical sense, but also from the standpoint of logistics and instrumentation. The development of new sensors, continued progress in the laboratory, and an increased awareness of the disparity between what is known about littoral sand and mud transport, have provided new impetus for field efforts. Gibbs (1970,1976) inferred from work in the vicinity of the Amazon River mouth that a nearshore concentration of fine-grained sediments occurred down coast through an Ekman-type coastal upwelling process. Suspended sediments introduced from the Amazon are initially carried

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 8

offshore and the sand fraction is deposited relatively close to the river mouth. The great majority of the suspended load is, however, deflected to the west by the Guiana Current. There, it becomes concentrated in bottom waters with a significant shoreward velocity component which returns only the clay fraction to the coast. Van Straaten and Kuenen (1958) and Postma (1961) provided the first physical model to account for the selective nearshore deposition of fine grained sediments in a coastal marine setting, albeit one dominated by tidal rather than wave generated flows. On the North Sea tidal flats studied, deposits of mud were restricted to the higher elevation surfaces which experienced only the final stages of the flood tidal current. These workers developed the well known "scour-lag" hypothesis to explain this size segregation. In it, they proposed that fine-grained sediments which settled from the waning flood current and during slack tide formed deposits sufficiently competent to resist removal by the subsequent ebb tidal scour. These models provide an indication of the span of spatial and temporal scales at which processes important to nearshore mud deposition can be expected to operate, but do not specifically address the basic dynamics problem at the shoreface, that is, how mud is deposited under waves. The limited data available suggests that the presence of gel-like fluid mud is characteristic of most open littoral settings where significant shoreface deposition occurs (Morgan et al. 1953, Delft Hydraulics Laboratory 1962, NEDECO 1965,1968; Zenkovitch 1967, Nair 1976, Martins et al. 1979, Wells and Kemp 1985). This term has been loosely applied to unconsolidated clay and silt suspensions with sediment concentrations ranging from 10 to 300 g-1"1 (Einstein and Krone 1962).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 9

Chronic channel siltation associated with the occurrence of fluid mud in San Francisco Bay led Einstein and Krone (1962) to conduct a classic series of laboratory experiments on the modes of cohesive sediment transport

in salt water. Suspensions with salinities of l ° / o o and greater were found to

be of sufficient ionic strength to ensure that collisions between clay particles would result in aggregates with lasting bonds. They proposed that fluid mud forms from coagulation of dispersed fine-grained sediments in a flocculation process involving collisions caused by both Brownian motion and fluid shear. In the dispersed suspension, away from the bed, collisions between particles take place primarily as a consequence of Brownian motion. Closer to the bed, however, where both the particle population and the range of particle sizes are greater, Einstein and Krone (1962) showed that local shear could cause collisions at a rate far greater than Brownian motion alone. Under these conditions large floes have a high probability of collision with small ones and thereby continue to grow in size as they fall toward the bed. The maximum size attainable by cohesive particles in the suspension is thought to be determined by a complex and largely unknown interplay between settling velocity, shear at the bed, and the strength of aggregate bonds (McCave 1984). Einstein and Krone (1962) proposed that destruction of floes by shear stress near the bed might return disaggregated particles to the suspension and thereby restrict clay sedimentation. Stow and Bowen (1980) suggest from their examination of silt laminated turbidite muds that shear within the bottom boundary layer may indeed act to sort silts and clays by concentrating clays in suspension while allowing silts to deposit. In order to explain the graded and laminated silt- mud couplets they observed, they proposed that at some critical

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 10

concentration, shear is "overcome" and clays deposit rapidly as a blanket over the silt laminae. Rine and Ginsburg (1985) found silt and clay couplets in radiographs of Surinam mudflat deposits and have proposed that a similar sorting process may operate in this coastal setting. No mechanism was identified in either study to explain the required rapid clay deposition. More than 40 years ago, however, Einstein (1941) explained sudden deposition, or "freezing", of mud-charged subaqueous underflows as a consequence of the tendency for such suspensions to acquire internal, or yield, strength in excess of the shear stress applied. It has long been known from viscometer measurements, that clay suspensions in the fluid mud range behave as plastic solids rather than true fluids (Bingham 1922, p. 215). As such, they are characterized by a capacity to resist deformation up to some critical level of imposed shear stess (Einstein 1941, Einstein and Krone 1962). The magnitude of the applied stress required to initiate flow, or conversely, the level below which flow ceases, defines the yield strength of the suspension. Krone (1962) found that the yield strength of fluid mud from San Francisco Bay increased in proportion to the sediment concentration raised to the power of 2.5. The results of subsequent work on harbor muds from other locations suggest that this empirical relationship may have broad applicability (Owen 1975, Hydraulic Research Station 1979). Because of the cohesive nature of the clay particles, deposits formed from fluid mud suspensions produce beds which also exhibit non-Newtonian rheological characteristics (Faas 1981). Maa and Mehta (in press) have recently prepared a useful review of the extensive engineering literature on this topic, which is briefly summarized here.. Beds of cohesive sediments

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 11

respond to wave-induced shear in a complex manner which ranges from elastic deformation to mass erosion (Alishahi and Krone 1964, Migniot 1968). Orbital motion and pressure oscillations associated with surface waves have been observed to cause horizontal and vertical displacements at and below the mudline both in the laboratory (Lhermitte 1960, Migniot 1968, Doyle 1973), and in the field (Tubman 1977). Furthermore, response to one set of wave conditions may change with time. Mud beds "soften" under cyclic loading (Thiers and Seed 1968, Schuckman and Yamamoto 1982) or suddenly liquify when pore pressures build beyond some critical point (Turcotte et al. 1984). The work performed by waves at and below the mudline effectively dissipates wave energy, thereby attenuating wave heights at a far greater rate than does normal friction over a consolidated sand bed. To explain wave damping observed over the "mud hole", a region of fluid mud located offshore of the central Louisiana coast, Gade (1958) provided an analytical solution for the dissipation of wave energy by a deformable mud bed. In this case, the bed was considered a viscous fluid. Since then, other analyses have assumed bed response to be elastic (Gade 1959, Mallard and Dalrymple 1977, Dawson 1980) or viscoelastic (Tubman 1977, Macpherson 1980, Hsiao and Shemdin 1980, Maa 1986). The study of Wells and Coleman (1981a) was the first to address the effect of shoaling waves on mud suspension and transport nearshore. Working along the coast of Surinam, they observed not only attenuation as waves propagated over a gently sloping (0.0005) bottom blanketed with fluid mud, but also a deformation of the characteristic wave profile from sinusoidal in deep water to the steep, symmetrically crested, flat-troughed

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 12

shape best described by solitary wave theory. Plunging breakers were observed at mud-free interbank locations, but at stations fronted by fluid mud, the long-crested solitary-like waves disappeared completely before reaching the shoreline. Wave height decreased with water depth at a constant ratio (H/d) of 0.23, far below 0.78, the critical value at which breaking can theoretically occur (McGowan 1894). By monitoring pressure variations in the upper 0.5 m of fluid mud, Wells and Coleman were able to obtain time- series records of sediment suspension and deposition. These data indicated that sediment was suspended at frequencies ranging from tidal (T=12.4 hr) to that of the incident waves (T=10 s). An intermediate, infragravity range was also identified (T=0.5 to 5 min) in this low-slope, dissipative environment. Suspended sediment concentrations were highest in areas where solitary-like waves were observed. Arguing that these waves might, in fact, be waves of translation, Wells et al. (1979) proposed that high rates of onshore sediment transport might be possible without invoking breaking waves or nearshore circulation cells. Solitary wave theory predicted free-surface elevations well, but as Wells et al. (1979) point out, several of the assumptions upon which this theory is based (Boussinesq 1872), particularly the requirements for an inviscid fluid, irrotational motion, and a rigid bottom, are certainly violated. High quality velocity time-series from the muddy nearshore are clearly needed, but, heretofore, have been unavailable. In the past 10 years, the advent of rapid-response flowmeters has permitted direct measurements of nearshore flows associated with shoaling and breaking waves (Huntley and Bowen 1973, Meadows 1976). Spectral analysis of such time-series has been used effectively to isolate the low

/ Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 13

frequency components of the nearshore velocity field which govern the net transport of sediments suspended by waves (Meadows 1976). When flowmeter data is acquired simultaneously at several levels above the bed, boundary layer models, developed by analogy from laminar flow theory, can be used to parameterize the turbulent shear flow. Results generated by these simple models, when applied to sediment-laden flows (Smith and McLean 1977a, b; Adams and Weatherly 1981a, b), and to flows affected by waves (Grant and Madsen 1979), have guided recent efforts to understand factors critical to sediment erosion, transport, and deposition. In the present study, I extend the work begun by Wells and Coleman (1981a, b) to include nearshore velocity data. The equatorial coast of Surinam enjoys relatively constant wind and wave conditions year-round and is rarely affected by storms. In contrast, the muddy coast of Louisiana, located 30° to the north, outside the trade wind belt, experiences a far greater range of weather types. Data was collected on time-scales ranging from seasonal to wave frequency in order to place a quantitative study of the nearshore sediment dynamics in the context of this evolving shoreline.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. F IE L D AREA

The 30 km section of coast selected for study lies west of the Southwest Pass of Vermilion Bay on the eastern margin of the Louisiana chenier plain (Fig. 2). This area has historically been the primary locus for chenier plain mudflat deposition (Morgan et al. 1953, Morgan and Larimore 1957, Coleman 1966, Morgan and Morgan 1983). It also includes segments of three other shoreline types which are more representative of the rest of the modem chenier plain coast, namely, unprotected marsh scarps, perched washover beaches, and reactivated strandline or chenier beaches (Kaczorowski and Gemant 1980, Wells and Kemp 1981). In plan view, this shoreline projects into the Gulf as a gradual headland with an apex at Tigre Point. The curvature is more pronounced on the eastern arm of this bulge, where the broad, shallow Trinity Shoal platform extends 30 km offshore. The presence of this shoal results in a lessening of the offshore slope from 0.001 (0.06°) at Dewitt Canal on the western margin of the study area to 0.0005 (0.03°) at the eastern edge, 10 km west of Southwest Pass. From offshore, this low-lying shoreline appears almost featureless, broken only by the outlet of Freshwater Bayou Canal. This artificial ship channel is periodically dredged to maintain a 4 m depth in its offshore approach, but discharge of freshwater and sediment is regulated by locks 2 km inland. Otherwise, the only landmarks are spoil banks associated with pipeline canals, and the 2 to 3 m relief of Cheniere au Tigre, a truncated beach complex which obliquely intersects the modem coast 5 km east of Tigre Point.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. cn i S f e > f O ^^CHEN|EREAUT!iatti iTIGRE: .POINT; DjU.S.A.C.d.E.'TsTIDE GAGE ; : /FRESHWATER 'BAYOU CANAL. PIPELINE CANALS 92-25' DEWITT CANAL □ Marsh □ Beach Ridge IB Tidal Mudflat Q Profile Station Figure 2. Map of study area, showing major shoreline features and beach survey stations.

Reproduced with permission of the copyright owner. Further reproduction prohibited withouto permissh 16

GEOLOGY Early workers recognized that the chenier plain evolved as a result of the coastwise transport and deposition of sediments derived from Mississippi deltaic systems to the east. Russell and Howe (1935) were the first to hypothesize mudflat formation as the primary mechanism for Recent coastal progradation. They interpreted the linear, generally coast-parallel sand/shell ridges, locally called "cheniers", which now form "islands" surrounded by marsh, as beaches built by waves during intervals of slackened fine-grained sediment supply. Fisk (1948) suggested on the basis of limited borings that it might be possible to correlate major progradational episodes on the chenier plain with the proximal location of specific Mississippi subdeltas. Price (1955) generalized the facies relationships described by Fisk (1948) into what he called the "chenier plain" type coast and was the first to note the analogy between this shoreline and that of the Guianas in South America. However, it was not until the completion of a major drilling program in 1959 that the geochronology and stratigraphic complexity of the chenier plain really began to be understood. Gould and McFarlan (1959) and Byrne et al. (1959) were then able to reconstruct its genesis on the basis of radiocarbon dating and paleofaunal analysis, respectively. Briefly, their interpretation indicates that as post-glacial sea level rose from -5 m to near its present level between 5600 and 3000 years B.P., the Gulf of Mexico inundated a dissected Pleistocene prairie surface forming a complex coastline of shallow lakes and bays. Sea-level rise ushered in a new era of delta development to the east as lobes spread out over large areas of the shallow shelf. Fine-grained sediments accumulated in the estuaries and filled

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 17

depressions in the Pleistocene surface. The oldest, most inland chenier ridges formed as spits or detached barriers coincident with the gradual leveling of the rate of eustatic rise at approximately 3000 years BP. Subsequently, two Mississippi delta complexes, the Teche, and the Lafourche, built out from the coast in the area between the present river mouth and the chenier plain (Kolb and Van Lopik 1958). The supply of both fine and coarse-grained sediment to the chenier plain was increased during these episodes, and progradation accelerated. Sand and shell chenier ridges continued to form, but, from this point on, mudflats, which previously were confined to back-barrier bays, began to appear at the coast itself, onlapping beaches on the Gulf side. These mudflats make up the bulk of the late Recent regressive sequence. In geological terms, the stratigraphy of the chenier plain documents a major progradational episode. Along a 150 km long front, the position of the shoreline was extended 20 to 30 km south into the Gulf of Mexico, advancing at a mean rate of 5 to 10 m -yrl. The presence of the abandoned coarse grained chenier ridges indicates, however, that this has not been a continuous process. Today, in fact, truncated and eroding chenier deposits all along the modem shoreline signify that the long-term progradational trend is, at least temporarily, in abeyance (Kaczorowski and Gemant 1980). Indeed, historical records suggest, paradoxically, that coastal retreat on the chenier plain has averaged nearly 10 m -yrl in the 200 years since the first reliable surveys were made (Morgan and Larimore 1957, Morgan and Morgan 1983). Wave reworking of chenier beaches provides most of the clastic sediments currently in transport along the coast (Beall 1968). It is only

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 18

within the coastal bulge encompassed by the study area, and to a lesser extent, near Rollover Bayou, 20 km farther west (Wells and Kemp 1981), that significant fine-grained deposition and localized progradation continue today. Despite the regional erosional trend, these areas experienced a dramatic influx of mud in the early 1950's. It was at this time that Morgan et al. (1953) mapped subaerial mudflats fronting the coast throughout the study area from Cheniere au Tigre west. These workers attributed this deposition, which meant economic disaster to the small beach resort developed at Cheniere au Tigre (Bailey 1934), to the growing importance of the Atchafalaya River as a Mississippi distributary, and a source of fine-grained sediments to the shelf.

INNER-SHELF CIRCULATION Atchafalaya River discharge, comprising roughly a third of the combined Mississippi and Red River flows at Simmesport, Louisiana, averages 5126 m^-s-l (USACOE 1974). It is dispersed onto the shelf through a dredged navigation channel and, to a lesser extent, through other breaks in an oyster reef at the seaward margin of Atchafalaya Bay (Roberts et al. 1980, Wells et al. 1983). The most significant influx of low salinity, sediment-laden waters from the Bay occurs during the spring flood which typically builds from late December through April, and declines rapidly thereafter (Angelovic 1976, Dinnel and Wiseman 1986, Cochrane and Kelly 1986). Flood discharge commonly exceeds the average rate by 300 percent (USACOE 1974). Tidal and wind-driven currents control the mixing of Atchafalaya and Gulf waters, and the dispersal of river-borne sediments on the inner

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 19

continental shelf. Mixed diumal and semi-diurnal tides in the study area have a mean amplitude of 60 cm (Manner 1954; NOAA, NOS 1981). Currents associated with these tides rotate in a clockwise manner with average speeds inside the 10 m isobath of less than 10 cm-s-* (Murray 1976). Because of the well-behaved rotation and low velocity of these flows, wind-driven currents with speeds ranging up to 50 cm-s“l (Adams et. al. 1982) play a more significant role in shelf sediment transport except in the immediate vicinity of the tidal passes of Atchafalaya and Vermilion Bays (Todd 1968). Cochrane and Kelly (1986) have recently integrated a variety of shelf current data to provide a coherent regional perspective. Their work indicates that flow on the Texas-Louisiana shelf is dominated for 10 months out of the year by cyclonic circulation around a mid-shelf geopotential low which appears off Texas in September and elongates to the east through the winter and spring. Westerly flow in the coastal limb is balanced by an easterly countercurrent over the outer shelf. Cochrane and Kelly (1986) show convincingly that the nearshore portion of this gyre is a wind-driven coastal boundary current (Csanady 1977), which is set up by the mean alongshore component of the wind stress, directed to the west. This interpretation is supported by the rapid change in circulation which occurs when the regional wind stress shifts to easterly in July and August. The cyclone which controls the inner-shelf current regime is abruptly replaced for these two months by anti-cyclonic flow. The geopotential high around which this circulation develops is located just off the coast of the study area, but drives significant northerly and easterly nearshore currents only along the Texas coast. Cyclonic circulation is re-established in September with the onset of winds from the east.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 20

High winds promote vertical mixing of the water column and build waves which put bottom sediments into suspension. Barring hurricanes and tropical storms, sustained high winds of regional extent are common only between September and April when extratropical cyclones or cold fronts cross the coast every 5 to 10 days. Femandez-Partagas and Mooers (1975) have described the detailed structure of surface wind systems associated with cold fronts traversing the northern Gulf of Mexico. They found that hourly observations of wind speed and direction made at a single station exhibit a characteristic sequence. A pre-frontal period of steady southerlies of increasing magnitude ends with a rapid clockwise shift in direction coincident with front passage. After a 180° rotation, a second steady wind regime, this time with a northerly orientation, becomes established, and wind speeds gradually diminish. The dominant cross-shelf component of the wind stress during these fronts results in sequential water level set-up and set-down along the east- west oriented coast of the study area (Chuang and Wiseman 1983). Wave energy also builds during the steady southeasterlies which preceed front passage. Wind-forced flow is to the west and onshore (Daddio et al. 1978). The onset of northwesterly offshore winds gives rise to short-lived, but relatively high-velocity, eastward setting currents (Crout et al. 1985, Adams et al. 1982), and strong inertial oscillations (Daddio et al. 1978).

STORMS AND MODERN MUDFLAT SEDIMENTATION Some of the same investigators who originally mapped the mudflats on the eastern margin of the chenier plain (Morgan et al. 1953) revisited this coast five years later, following the passage of Hurricane Audrey in June, 1957 (Morgan et al. 1958). This devastating storm went ashore near the

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 21

Texas-Louisiana border and generated a 3 to 4 m storm surge in the study area. Morgan et al. (1958) noted that although the hurricane caused an average of nearly 100 m of coastal retreat along most of the chenier plain, the mudflat-fronted shoreline experienced very little erosion. In two locations, in fact, shoreline progradation was documented. There, mud was deposited as discrete shore-welded "mud arcs" or bars 2 m thick with alongshore dimensions of nearly 4 km. The mud in these features had abrupt, easily defined lateral boundaries. The sharpness of the contact suggested to the researchers "that the mass of fluid mud was transported by the storm tide and deposited as a unit". Seeing the effects of the hurricane brought to mind an observation made during the initial surveys in the spring of 1952. In 1958, Morgan et al. wrote:

"There is morphological evidence of the effectiveness of suspended mud during the height of the storm. With normal storms this material is moved toward the shore where some of it may be permanently or temporarily incorporated into the beach. During field work in 1952 a beach profile was surveyed across an essentially non- mudflat area. Three days later following a minor storm the area was revisited. The storm had blanketed the foreshore with a layer of gelatinous clay which had a maximum thickness of about 18 inches."

In the mid-1960s, Coleman (1966) found little change with respect to the distribution of mud in the study area, but added much new information on the physical and biological properties of the mudflat sediments. He X-rayed mudflat cores which, to the unaided eye, appeared almost featureless. Radiographs of "massive" sections disclosed multiple sets of parallel laminations relatively undisturbed by biological reworking. The laminae

/ Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. were composed of concentrated silts slightly coarser and denser than the predominantly clay matrix. The observations made by Morgan and his colleagues in 1952 and 1957, together with the sedimentological data provided by Coleman (1966), suggest, somewhat surprisingly, that chenier plain mudflats are deposited rapidly under the highest energy conditions this coast experiences.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. DATA ACQUISITION

Data were collected over a 2.5 year hurricane-free period from December, 1980, to May, 1983, and during a follow-up reconnaissance made in May, 1985. The bathymetry and surficial sediment distribution was surveyed along the 30 km length of the study area to 4 km offshore, or roughly the 3 m isobath. Geomorphological, sedimentological, and rheological data were acquired in beach and nearshore environments at shoreline stations spaced 3 to 5 km apart (Fig. 2). This work, summarized in the first subsection below, provided the information necessary to design a dynamics experiment and interpret its results in a regional context. The second subsection covers the experiment conducted in December, 1982.

SEDIMENT SURVEY Depth and positioning data for the offshore survey were obtained with a Raytheon 731 depth recorder and a Decca-DelNorte trisponder system, respectively. One hundred-eighty bottom samples were collected at a 0.5 km spacing along arced traverses which intersected the coast at 1 km intervals. A long-handled scoop-type sampler was used to minimize loss of fine-grained sediments in this shallow water setting. Samples were analyzed for percent sand, shell, silt and clay using standard techniques (Folk 1980). Sizing of the silt and clay fractions was accomplished by Coulter Counter (Sheldon and Parsons 1967). Benchmarks were established at eight field stations and beach profiles were surveyed at these locations every 3 to 4 months. The stations were located using the trisponder system and benchmark elevation was determined by referencing local water levels at each station to the datum of a recording 23

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 24

tide gauge maintained by the U. S. Army Corps of Engineers at Freshwater Bayou Canal. Survey lines were initiated on a stable marsh surface landward of washover effects, carried over the berm, if present, and continued offshore as far as visibility and water depths permitted (approximately 200 m). Where soft muds were present, the bottom was defined as the upper surface of the mud layer. Profiles were digitized relative to mean sea level (MSL = +0.24 m Gulf Coast Low Water Datum). The shoreline was defined as the point where the MSL datum intersects the profile. Shoreline progradation or retreat was determined from the translation of this point relative to the benchmark. Push-cores (Van Straaten 1954) up to one meter long were obtained to characterize deposition in nearshore sub-environments. These cores were split and described from X-ray radiographs of 1.5 cm thick slabs using the methods of Roberts et al. (1976). They were compared with sequences identified in a 5.5 m long vibracore (Laneski et al. 1979) recovered at profile station 3 from a vegetated mudflat formed in the early 1950s (Morgan et al. 1953). In addition to being X-rayed, this core was sampled at 5 cm intervals and analyzed for textural variability in the same way as the offshore sediments. A push-core obtained in the sub-aqueous section of a newly deposited mudflat was analyzed to determine the mass physical properties of this sediment. The core was transported to the lab in a vertical position, and was subsampled with a glass tube composed of pre-weighed, detachable 5 cm sections of known volume. The tube was vibrated slightly and allowed to sink into the sediment as described by Sikora et al. (1981). Upon extraction, the tube was segmented and each section weighed to determine bulk density.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 25

In this way a qualitative record of the vertical variation in density and degree of consolidation was obtained. These data were then compared with grain size measures determined by pipetting (Folk 1980). Surficial sediments from the same location were subjected to X-ray diffraction to determine clay mineral composition (Schultz 1964, Griffin 1971) and analyzed to determine viscosity and Bingham yield strength in a capillary viscometer linked to a manometer to allow application of variable known head pressures. This equipment and the method were first described by Einstein (1941), and more completely later by Das (1970).

DYNAMICS EXPERIMENT Nearshore wave characteristics, flow velocities, and suspended sediment concentrations were monitored at profile station 6 between 9 and 13 December, 1982. A cold front traversed the study area during this time, and data was collected in three sessions corresponding to the pre-frontal, frontal, and post-frontal periods. At the time of the experiment, nearshore soft mud accumulations averaged 0.5 m in thickness. A tide record was obtained for the duration of the experiment using a pressure transducer water-level gauge fixed to a platform at the mouth of Freshwater Bayou Canal, 2 km east of station 6. Hourly observations of wind speed and direction were recorded at the Freshwater Bayou Locks. The smooth mud bottom at station 6 had a slope of 0.004 (Fig. 3). Time-series data were acquired at two shallow water locations (d ~ 1.0 m) 10 m apart along the shore-normal transit. The landward station (B) was approximately 90 m from shore. At each location, a 2 cm-diameter steel rod was driven vertically into the bottom. The transducer of a pressure-type

/ Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. rv> CD

ANALOG ANALOG RECORDERS / / / / // / DUCTE DUCTE OWMETEHS MPELLO MPELLO BI-DIRECTIONAL BI-DIRECTIONAL V Y L RODS SUPPORT INSTRUMENT INSTRUMENT SENSORS / V/V/ **A / 7 RELEASE ELECTRONIC ELECTRONIC SEDIMENT SAMPLERS SUSPENDED

/ / / / / / / / / / / . /’..BOT TOM / / // //.' //.' // / ' / / / / / / // / / / / ••' / / / / /••' - ‘ / - ‘

/ / / // 7 7 j - L . / .. / / / / ■' ■ ■' Figure 3. Cross-section of shoreface at station 6, showing deployment of sensors during cold front experiment.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 27

wave gage (Fredericks and Wells 1980) was clamped to the landward rod such that the diaphragm of the sensor was fixed at the sediment-water interface. An identical transducer was attached to the seaward rod (A) at the same level as the first to ensure that the frequency response factor was the same for both sensors. Three miniature (8 cm diameter), ducted, bidirectional impellor-type flowmeters fabricated at Coastal Studies Institute were fixed with a common orientation to a 2.5 cm diameter stainless steel sleeve. The sleeve was dropped over the landward steel rod and lowered until it rested against the housing of the pressure transducer. In this position, by rotating and clamping the sleeve, the flowmeters could be oriented to measure either onshore-offshore or alongshore flows at 13, 30, and 60 cm above the bed. All data were recorded in analog form aboard a small boat moored 30 m from the instrument array. Signals from the wave gages were logged by a 2- channel Gould Brush strip chart recorder. The sense of rotation and intervals between discrete on-off signals generated by the rotating flowmeter contacts were converted electronically to positive and negative instantaneous velocities which were logged on heat-sensitive paper by two Astromed recorders. Nine 20 minute data runs, in addition to a trial run (No. 1, not used in this analysis), were completed during the course of the experiment, including 4 paired runs in which the flowmeters were sequentially oriented onshore and then alongshore. During the last two runs, a recorder malfunction limited the number of time-series obtained. During these runs, pressure data was acquired only at the shoreward station (B), and flowmeter data was limited to the 60 and 13 cm elevations.

/ Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 28

During runs 2 through 10, 39 instantaneous suspended sediment concentration profiles were obtained within 2 m of the instrument array using a multi-level sampler modified from that of Kana (1979). The sampler consisted of four 1-liter Van Dorn bottles stacked on a frame and fitted with a common release. When triggered with the bottom of the frame at the bed, samples were obtained simultaneously at mean elevations of 10, 35, 65, and 95 cm above the bottom. In the laboratory, measured aliquots of the water samples were passed through Millipore 0.45 micron filters using a pressure filtration system. The filters were weighed to determine sediment concentration following standard procedures (Meade et al. 1975). Pressure and flowmeter records were digitized at a 0.5 s time step which was determined to preclude aliasing (Nyquist frequency = 2 cps). The power spectra were computed (Bendat and Piersol 1966) using a Fast Fourier Transform (FFT) algorithm (Cooley et al. 1969). Coherence-squared statistics and phase relations were calculated between similar time-series (Jenkins and Watts 1968). Low-pass smoothing of the records was accomplished in the frequency domain by suppressing Fourier coefficients with a frequency greater than 0.005 cps (T = 200 s).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. RESULTS

SEDIMENT SURVEY Geomorphologv The nearshore bathymetry and distribution of sediments in the study area are mapped in Figure 4. Isobaths show the effect of Trinity Shoal on the offshore gradient and outline a 3 km^ mudflat between profile stations 7 and 8. Poorly sorted muds with a median size of 2.5 microns and a silt content ranging from 20 to 40 percent dominate the nearshore surficial sediment suite. Fine sand (3 phi median) fronts the coast east of Freshwater Bayou Canal. There, it expands west to east from a ribbon a few 10's of meters wide near station 5 to an apron nearly a kilometer wide off station 1. A second band of sandy sediments in this area also trends to the east from Tigre Point somewhat seaward of the first. The only sand-sized sediments on the other side of the canal are found trailing west from the dredged spoil disposal area on the margin of the navigation channel. A summary of grain-size, morphological, and shoreline change statistics at the eight survey stations is given in Table 1. These results confirm indications in the offshore data that Freshwater Bayou Canal divides the shoreline of the study area into two distinct sedimentary provinces. Unvegetated mudflats 10's to 100's of meters wide characterize the coast west of the canal (Fig. 5a). There, clastic sedimentation is limited to pockets of small bivalve shells (Mulinea lateralis. Nuculana concentricaV and organic "coffee grounds". In contrast, sand and shell beaches are continuous east of the canal, except for a gap of eroding marsh between station 3 and Cheniere au Tigre. This space of less than 2 km separates two beach systems which differ significantly in composition. Oyster shell fragments fCrassostrea

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CO o Station saxet MuddySand ♦ Bottom Sampling E3 E3 CHENIERE AUTIGR CHENIERE Sandy Mud E3 E3 OINTJ CANAL FRESHWATER- FRESHWATER- o PIPELINE CANALSPIPELINE mSEHR DEWITT CANAL BELOW MEAN LOW WATER OAtUM CONTOUR INTERVAL 0.5 METERS □ Marsh □ Bsach Rldga □ Tidal Mudflat Q Profit* Station □ Mud □ Silt Figure 4. Nearshore surficial sediments and bathymetry in the study area.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CO

Jul FM 107.5 1.6 -3 3 .0 .01 May 8

+20.0 1.1 .02 Mar 90.0 19.0 M M OWB/M <4.0 <4.0 <4.0 <4.0 144.0 Jul Sep 1.1 .02 .005 .004 .004 Mar M 20.5 27.5 Oct -4 .0 <4.0 Mar OWB 5 6 7 .009 -9.0 +15.0 B/FM FM FM 1.4 1.4 May .05 .04 OWB Jul 3.0 3.0 3.75 <4.0 5.5 1.0 5.8 4.5 FM +1.5 Table 1 B Jul 3 4 .003 .009 .08 .08 B 0.02.4 3.0 1.1 .002 2 6.5 2.0 -1.0 -7.8 Jul December, 1980 - May, 1983 Chenier Plain Shoreline Characteristics Features!2) B Max retreat (-m) 10.0 13.5 6.2 Median size (phi) 3.0Slope 3.0 3.0 .002 Month observed Jul Slope .05 Median size Height (phi) (m) 1.5 1.9 Max advance (+m) 8.0 Month observed Dec Oct May May, 1981- May, 1983 (m) -15.5 Inshore (MSL to100 m offshore) Net 2 year change: Foreshore (MSL to berm crest) <2)B = <2)B sand bars; FM = fluid mud 0)OWB = overwash beach; CB = chenier beach; MS = marsh scarp; M = mudflat. Morphology!') OWB CB MS Shoreline change between surveys (3 to 4 months) Station (E to W) 1

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

A Figures 5a-b. Muddy shoreface at station 7 (5a); Cheniere au Tigre (5b). reactivated chenier beach (station 2) at

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 33

virginical eroding out of the reactivated beach ridge form an important component of the relatively steep beaches extending from Cheniere au Tigre to the east (Fig. 5b). In contrast, low-profile overwash beaches with little shell characterize the shoreline extending east and west of Tigre Point. An abrupt scarp marks the seaward edge of the marsh in the area between the two beach systems. This marsh is the remains of a large expanse of unvegetated mudflats which Morgan et al. (1953) observed during the 1950's. Significant clastic beach sedimentation did not occur at survey station 3, which was located in this section, throughout the duration of the field program (Fig. 6a), although bars of fine sand resting on consolidated marsh sediments were often present just offshore. When this location was revisited in 1985, however, a low-profile sandy beach covered the survey station (Fig. 6b). This deposition was clearly a consequence of westerly longshore transport, as the boundary of the sandless shoreface could still be found 0.5 km to the east. Fine details of shoreline change are not resolved in beach profiles surveyed at 3 to 4 month intervals. The results of the 2.5 year program do, however, provide important information about seasonal and spatial patterns of deposition and erosion. A complete set of the profiles obtained at the eight survey stations is compiled in Appendix A. The sandy beaches east of the canal undergo an annual cycle of winter erosion followed by onshore translation of a bar, or multiple bars, during the summer and fall (Fig. 7). The most extensive bar development occurs at the easternmost stations, 1 and 2. Although the position of the shoreline was stable at Tigre Point (Sta. 4) and Cheniere au Tigre (Sta. 2), the remainder of the stations east of the canal experienced between 8 and 16 m of coastal

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 34

■ h

Figure 6a-b. Time-lapse views of the vegetated mudflat mapped by Morgan et al. (1953) at station 3; eroding marsh scarp, May, 1982 (5a), replaced by clastic washover beach, May, 1985 (5b).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Non-winter Accretion at Clastic Station 4

14 DEC'81 15 MAR'82

1.0m

Sand Bar Formation 250m50m MSL 100m 150m 200m

13 JUL'82

15 MAR'82

20 OCT'82 13 JUL‘82

Onshore Bar Translation

Figure 7. Bar migration at Tigre Point (station 4).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 36

retreat over the two-year period between May, 1981, and May, 1983 (Table 1).

Mudflat Sedimentation The beach survey program demonstrated that, non-hurricane, littoral deposition of low-density muds takes place west of the canal only between the months of January and May. The most extensive mudflat observed during the study period formed between stations 7 and 8 after a survey in mid- December, 1981, but before the next visit in late February, 1982. It was initially arcuate in form, and closely resembled those mapped by Morgan et al. (1958) after Hurricane Audiy. The shore-normal and alongshore dimensions were estimated at 300 m, and 1500 m, respectively, and the maximum thickness at 2 m. Although the eastern arm of the arc was not visible after 6 months, the western arm survived as a subaerial spit obliquely intersecting the coast for at least 1.5 years (Fig. 8). An elevated bar or berm on the seaward margin of the mudflat is typical of new deposits (Fig. 9) of highly fluid sediment (Fig. 10a). Subsequently, the bar and other regularly exposed areas dry and the surface cracks into ever smaller polygons (Fig. 5a). When these areas are subject to wave action, pieces of the surface crust are ripped up and the mudflat acquires an irregular surface. The bar is usually gone by the beginning of the winter following deposition (Fig. 9). Mud clasts, shells, and "coffee grounds" accumulate in offshore-oriented erosional grooves or flutes (Fig. 10b). If new mud is deposited the next spring, that part of the mudflat which survives the winter will be protected from further erosion. Oyster grass (Spartina altemifloral then propagates onto this newly available land. Freshly deposited mud is easily removed, and mudflats undergo

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 8. Oblique aerial view of western limb of large mud bar formed in Spring, 1982, between stations 7 and 8.

Prollla Station 8

1.5m- ACCRETION

EROSION 1.0m

13 FEB’81 0.5m 2 9 MAY’8 1

MSL

1.0m -

0 .5 m 2 9 MAY’S 1

MSL 50 m 1 0 0 m 1 5 0 m 2 5 0 m 3 0 0 m 2 9 J U L '8 1

-0 .5 m

Figure 9. Profile time-series of mudflat accretion and erosion at station 8, February - July, 1981.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 38

Figure 10a-b. Shoreface mud at station 7; Freshly-deposited fluid mud, Spring 1982 (10a); eroding surface showing development of shore-normal flutes 5 months later (10b).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 39

erosion in all seasons. The most rapid retreat, however, occurs during the summer immediately following a major depositional event (Table 1). Then, the bar loses definition as sediments are redistributed both landward and seaward along the profile section (Fig. 9). Progradation may more than offset shoreline retreat in any given year, but the required sedimentation does not occur at a uniform rate, or at the same locations each year. As a result, zones experiencing major deposition in any single spring may be separated by long stretches of shoreline which receive little or no mud. The data summarized in Table 1 shows that between May, 1981, and May, 1982, shoreline translation, whether through accretion or erosion, was far more active at mudflat stations 6,7, and 8, than at any of the stations east of the canal. The shoreline shifted 15 and 20 m seaward in this period at stations 6 and 7, respectively, but moved 33 m landward at station 8. It is illustrative of the dynamic nature of the mud-fronted coast that between February and May, 1981, the months just prior to the two year period considered above, the shoreline at station 8 moved seaward 108 m (Fig. 9). A log of the 5.5 m vibracore from station 3 is provided in Figure 11. Coring took place 20 m from the shoreline in the marsh which now covers a mudflat described by Morgan et al. (1953). The core penetrated a bioturbated sandy mud at 3.2 m below the marsh surface which matches Coleman's (1966) description of nearshore marine sedimentation. The upper 2 m of the core record the mudflat and marsh sequence. The basal 0.5 m of this sequence fines up and is composed of poorly-defined layers of silty-sand and clayey-silt with no discernible erosion surfaces. Radiographs of this section show wavy bedding distorted by water escape structures (Fig. 12a) which may be artifacts of the coring process. In

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHENIER AU TIGRE-CORE T 3

«SAND tSHELL «ORG C LfTHOLOGVUNIT DESCRIPTION DEPOSmONAL 0 SO 0 10 0 2 ENVIRONMENT wim Imbricated eJwie BEACH/MARSH

2to lOcmttttoKbattocomooeedof lemmatad dbownoyrnaeaaamudiacwand M common at baaa of ootoata; btrrowa. roots, mud cracfca, and concisions tHupaauaon: •hall haen and aand present atbaaa MUDFLAT ofutt ;iarmnationa change from oonvaua and tontiojnr naar oaaa to wavy and flat hmg farther up; Aama and toed svucaras convnon naar Oaaa

dean aand and anal; tomnaoona ndacrrt btf no burrows teoma wavy iamrae andamel- BAR seals i-baddtoq; tnoncatedshalel baaa

Intaroaddad aand end mud;wedge-shaped TRANSITIONAL poorty-aonedlaminattons; shelt-noah at baaa; no burrowa (BAR TROUGH)

heevfy bisrowed. faw tontteufar and wavy aand/attt laminae dUcernaols ; aftail NEARSHORE haan secumuettone{burrowingrHenwty MARINE Ineraaaaa witti depth

Figure 11. Log of vibracore from marsh now covering mudflat described by Morgan et al. (1953) west of Cheniere au Tigre at station 3; see Figure 6a-b.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. contrast, bedding is well-preserved in the overlying sand-free mudflat sediments and shows repeated 2 to 10 cm thick beds consisting of flat and lenticular silt laminae overlain conformably by a layer of featureless silty- clay. The clay layer makes up an estimated 60 to 90 percent of the thickness of each couplet. Erosional surfaces separating the couplets are marked by scour-and-fill structures (Fig. 12b), and, in some cases, by thin layers of imbricated Mulinea shells (Fig. 12c). Five of these cycles were recognized below the bioturbated, organic carbon-rich marsh horizon. A 0.2 m thick cross-bedded sand and shell washover deposit caps the rooted zone.

Mineralogy and Rheologv The cyclic bedding observed in the upper mudflat section of the vibracore produces variations in bulk density with depth. Overall, the trend in a push-core obtained through newly deposited mud is from 1.2 g-cm-3 at the surface to 1.4 g-cm-3 a meter below (Fig. 13). The density increase occurs as a series of steps, however, increasing downward within each silt- mud or shell-mud couplet to the point where grain-size data indicates the base of the cycle. Bulk density drops off abruptly below the erosional interfaces so that the upper part of each cycle is underconsolidated. The percent of sand-sized particles (mainly shell) in the mudflat samples from this core ranged from near zero to 14 percent. Silt content varied between 20 and 50 percent, while clay made up the remainder, between 50 and 80 percent of the total. The within-bed pattern of consolidation, apart from the trend which characterizes the deposit as a whole, suggests that each of these couplets represents sedimentation occuring during a single depositional event. X-ray diffraction traces from the <1 micron fraction of mudflat sediments subjected to the standard suite of laboratory treatments are shown

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

positives of mudflat sequences laminae distorted by loading and laminae filling scour depressions silt and shell); flat and lenticular water escape (12a), flat-lying silt (light tones indicate higher density lamina of disarticulated, imbricated in in underlying erosional surface (12b), Mulinea valves at erosional surface (12c). Figures 12a-c. Bedding structure in radiograph

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 43

COARSE FRACTION DENSITY (g -c m " 3) (% > 63 M) DEPTH

Figure 13. Texture and bulk density changes with depth in upper meter of newly deposited mudflat.

/ Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 44

in Figure 14. The major clay mineral groups identified were illite, smectite, and kaolinite. As is shown in Table 2, the two analytical methods used to obtain semi-quantitative estimates of the abundances of these minerals give somewhat different results for the dominant expanding-lattice illite and smectite groups. The method of Griffin (1971) is similar to techniques which have been used in previous studies of sediments in Atchafalaya Bay and the inner shelf (Powers 1957, Johns and Grim 1958, Whitehurst et al. 1960, Ho and Coleman 1969, Brooks 1970, Mobbs 1981, Van Heerden 1983). Mobbs (1981) worked both in the Bay and just offshore, and found insignificant differences in composition between the two locations. Comparing his results with earlier work, he did, however, suggest a long-term trend toward an increased illite fraction in the most recent deltaic sediments. Van Heerden (1983) noted a similar trend in cores containing both bay bottom sediments and those associated with the newly emergent subaerial delta. He further observed some indication of an increased illite content in flood deposition relative to low-flow sedimentation. Mean percentages for the < 2 micron fraction from Mobbs (1981) are presented in Table 2, and show a somewhat higher percentage of illite relative to smectite than do the comparable mudflat results. A mixed-layer illite-smectite group was identified in the mudflat sediment using the more detailed method of Schultz (1964). This mixed- layer clay makes up almost half of the smectite identified following Griffin (1971). This mineral group can form when degraded illite enters seawater and regains much of its original potassium content (Johns and Grim 1958). The dominance of the transitional mixed-layer illite-smectite clay in

/ Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 45

AIR-DRIED

GLYCOLATED

330

550

10 15 20 25 30 DIFFRACTION ANGLE (29)

Figure 14. X-ray diffractograms of smears from the< 1j i fraction of mudflat sediments treated as indicated: S=Smectite, l=lllite, K=Kaolinite, and Q=Quartz.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 46

Table 2 Clay Mineral Abundances Chenier Plain Mudflat and Atchafalaya Bay

Clay Mineral Atchafalaya Bay Chenier Plain <2 jim <1jim Methods------Mobbs (1981) Griffin(1971) Schultz(1964) % % — % ...... Kao Unite 16 d ) 17 19

Illite 53 43 31

S m ectite 31 39 20

Mixed-Layer Illite-Sm ectite 30

(^Includes Chlorite

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 47

sediments from the study area suggests a close connection between mudflat deposition and the illite-rich Atchafalaya suspended load. Results of viscometer runs on newly deposited mudflat sediment with a bulk density of 1.26 g-cm-3 (sediment concentration = 416 g-l_l) and a salinity of 22 °/OQ are summarized in the shear stress - strain rate plot in

Figure 15. The data describe an approximately linear relationship (r2 = 0.93) between the applied shear stress, T, and the rate of deformation, du/dy. The equation for this line describes the behavior of a Bingham plastic

T = Tp+|i (du/dy) (1)

where Tp is the shear stress corresponding to the yield strength, and |i is the

sediment viscosity (Bingham 1922). Deformation ceases when the applied shear stress falls below the yield strength. The yield strength, 171 dynes-cm"2, and viscosity of the mudflat sediment, 52 dyne-s-cm_2, are determined by the intercept and slope, respectively, in Figure 15. Krone's (1962) relationship between yield strength and sediment concentration has the form

Tp = FC2.5 (2)

where C is the sediment concentration in g-1‘1, if Tp is measured in dynes-

cm-2. The experimental results indicate a value of 4.9 * 10"5 for the proportionality factor, F. If Krone’s empirical relation holds, determination of this factor allows estimation of yield strength for a range of sediment concentrations. The line described by this equation is plotted on Figure 16 (adapted from McCave 1984) along with results for other marine muds (Krone 1962,1963, Owen 1975, Hydraulic Research Station 1979).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 48

5 0 0 i

400

/* = 51.7 dyne-s-cm-2 S t 3 0 0 OC o < > Z 2 0) > 9 200

100 Tp =170.9 dynes-cm-2

0 1 2 3 4 s-1 DEFORMATION RATE(du/dy)

Figure 15. Shear stress-strain rate plot from capillary viscometer measurements made at 22° C on mudflat sediment with a density of 1.26 g-cnr3 (sediment conc. = 416 g-M), and a pore water salinity of 22 °/00. Bingham plastic behavior is indicated by the presence of a Bingham yield strength, Tp, of 170.9 dynes-crrr2

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 49

1 0 0 0 .0

C9 1 0 0 .0

1.0

"S.

0.4 10 20 40 60 100 200 400 SEDIMENT CONCENTRATION (g-l-1 )

Figure 16. Fluid mud yield strength plotted as a function of sediment concentration (from McCave 1984). Data from Krone (1963), Owen (1970), and Hydraulics Research Station(1979) for a variety of marine muds plot with approximately the same trend (Tp=FC2-5). This trend is extrapolated from the mudflat data along the dashed line shown.

/ Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 50

DYNAMICS EXPERIMENT The wind record for 10 through 13, December, 1982, in Figure 17, shows the 180° shift in direction which marks the passage of a cold front in the early morning hours of the 1 l^h . Wave characteristics, flow velocities, and suspended sediment concentrations were monitored in the three sessions indicated. The pre-frontal regime began at midday on the lO^1 with the onset of easterlies ranging between 10 and 15 knots. The steady pre-frontal regime lasted 15 hours and includes the period of the first two instrument runs. Starting at midnight, wind direction rotated in a clockwise manner from easterly to northwesterly during the next 8 hours. Speeds averaged less than 10 knots during this shift, which we identify as the frontal period. A rapid increase in velocity after runs 3 through 6, on the afternoon on the 1l^1, signaled the onset of the post-frontal regime. Northwesterly winds strengthened over the next 8 hours to sustained speeds between 25 and 35 knots. Eight hours later, velocity dropped to between 15 and 25 knots, but continued steady from the north throughout the 12^. Wind azimuth then shifted clockwise a second time in the early hours of the 13^ and speeds dropped off to 10 to 15 knots. By midday, during the period of runs 7 through 10, winds were from the east once again. The effects of front passage were manifested in departures of the observed tide record from that predicted, in changes in the gravity wave regime, and in the distribution of suspended sediments above the bed. These effects are described in the first three sections below. The final results section is focused on the relationship between infragravity water level oscillations and low pass flow velocities.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. RUNS RUNS 1-2 7-10 100 PREDICTED E 3 50 z O < MGL > UJ mi UJ OBSERVED HI “50 O

\-/

H Z s N <

(0 g 30 Z * o 20 1U aUJ (0 a 10 z 5

10 13

Figure 17. Winds and tides during cold front experiment, 10-13 December, 1982. Tide level recorded at the experiment site are compared with those predicted at the mouth of the Mermentau River (NOAA, NOS 1981). Wind data are hourly observations at Freshwater Bayou Canal Lock. The periods of the pre-frontal, frontal, and post- frontal instrument runs are indicated by stippling.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 52

Meteorological Influences on the Tide Record Water level at Freshwater Bayou Canal is plotted along with that predicted at the mouth of the Mermantau River in Figure 17 (NOAA, NOS 1981). This gauging station, located 70 km west of the Canal, is the closest for which predictions are made. Although the mean tide range there exceeds that at Freshwater Bayou by 0.1 m, a general comparison can be made. Water level is an average of 0.2 m above predicted for the pre-frontal period on the 10th Predicted and observed elevations are in agreement on the 11 & coincident with front passage. Following the onset of the strong northwesterlies later that day and continuing through the 12th, Freshwater Bayou levels average 0.3 m below predicted. On the 13^, the water level depression was reversed as easterlies again set in.

Waves and Oscillatory Flow Four kilometers offshore of station 6, significant wave height during the first 6 data runs (Pre-frontal and frontal periods) was estimated at 1 m. The wave field at this time included a long-crested swell which approached the shoreline from the south-southeast at a 5 to 10 degree angle of incidence. Following front passage, wave height offshore was less than 0.5 m, but a short-crested, confused sea made the dominant direction of approach impossible to assess. Wave data acquired at the experiment site are summarized in Table 3. Water depth ranged from 85 to 105 cm, and wave heights averaged less than 20 cm. These low-amplitude waves formed spilling breakers at a point 30 m landward of the instrument array during the first two runs (pre-frontal period). In all subsequent runs, waves were completely attenuated before reaching the shoreline, and breaking waves were not observed.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. cn co

— 8 8 9 11 6 4 Swell/Sea — —— —— —— 17 18 20 9 15 11 22 8 Orbital Vel. Urms Urms (cm-s-1)

2097 761 L L (cm) 2397 976 1806 1107 2097 965 Wavelength 4.8 2290 1025 3.6 2000 976 5.7 4.3 4.4 2174 829 10.2

8.4 5.4 2130 953 3.1 1.9 173 123 8.9 5.1 13.0 11.3 Height ^rms (cm)

2.5 5.0 4.9 2.8 3.5 2.8 5.6 6.3 2067 829 3.4 T (s) Period 0.7 0.4 6.8 3.1 Table 3 6.8 Swell/Sea Swell/Sea Swell/Sea 6.9 6.5 December, 1982 8 Cold Front Experiment 96 10-13 90 8.0 3.2 5.3 90 6.185 3.2 d 85 6.8 3.0 10.4 100 7.3 105 100 100 7.3 3.5 11.0 4.6 2242 1107 100 5.8 W ater Depth (cm) Sea and Swell Wave Statistics Along Along On/Off Along Along On/Off On/Off On/Off On/Off Flowmeter Orientation

13:01 13:38 12:34 11:30 13:07 10:51 12/11 5 14:23 3 4 6 15:09 2 14:20 7 8 9 10 Frontal Run Run No. Time Pre-Frontal 12/10 Post-Frontal 12/13 Mean Std. Dev.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 54

The effects of front passage on wave height and oscillatory flow can be seen in the sequence of pressure and flowmeter spectra shown in Figure 18. Spectra computed from representative 20 minute pressure and flowmeter records acquired before, during, and after front passage all show energy density peaks at two frequencies in the gravity wave range. These peaks define distinct wave trains; swell with an average period of 6.8 s (0.1471 cps) and locally generated seas of 3.1 s period (0.3226 cps). The energy contributed by linear trends and infragravity oscillations of period greater than 200 s (0.005 cps), combine to form a third peak which appears close to the origin in all the spectra. Root-mean-square wave heights (Hrmg) and oscillatory flow velocities (Um s) calculated for the principal gravity wave

peaks (Thompson 1981) are also included in Table 3. Two changes in the wave field associated with the transition from pre- to post-frontal conditions are apparent. First, the total gravity wave energy, consisting of the combined contributions of both sea and swell, decreases through the sequence. Second, the distribution of energy between swell and sea frequencies shifts significantly. Hnns and UnxiS in both bands are of

comparable amplitude in the pre-frontal runs. Energy associated with the swell is retained during the second sampling period coincident with the shift to northwesterly winds, while that of the seas is greatly diminished. Finally, a reduced, residual swell and building seas characterize the final, post-frontal records. The differences in inferred wave height between seaward (A) and landward (B) pressure sensors 10 m apart were less than 5 percent in all runs where simultaneous data were obtained. Although it was noted qualitatively that, except during the pre-frontal period, waves were completely attenuated

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. PERIOD (•) 10 5 300 -I PRESSURE FLOWMETER 60 CM RUN 2 RUN 2 SWELL ■ SWELL 200-I 400- SIa SEA

100-1 200-1

3001 600 RUN 3 RUN 3

200 ■! 400-

200-

300-; 600 RUN 7 RUN 7

200 - 400-j

1 0 0 - 200

0.25 0.50 0.75 1.00 0.25 0.50 0.75 1.00 FREQUENCY (cps)

Figure 18. Representative energy density spectra from 20 minute pressure and flowmeter time-series for pre-, frontal, and post- frontal runs. Energy concentrations at the characteristic sea and swell frequencies are indicated.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 56

before reaching the shoreline, comparison of water level spectra at the two stations indicates that in four runs, wave height actually increased slightly in the landward direction. Such an increase can be seen in spectra from run 3 plotted in Figure 19a. Coherence and phase estimates computed from the cross-spectra of simultaneous water level records from run 3 are also shown in Figure 19a. The relationships between coherence, phase, and frequency shown in this figure are representative of all the runs and provide information about the incident waves which cannot be deduced from individual spectra. As is expected, coherence between the records is highly significant through all frequencies with appreciable energy. The strength of the correlation lends credibility to the phase trend shown. Below 0.005 cps, in the infragravity peak, the two records are in phase. The gradual increase in lag with increasing frequency suggests that the spectral peaks observed in the gravity wave range are associated with progressive waves rather than standing waves or spectral harmonics (Suhayda 1974). Given that the spacing between the sensors is known exactly, the magnitude of the phase lag at the principal gravity wave frequencies provides a direct measure of swell and sea wavelengths. Mean wavelengths of the swell and sea calculated in this way were 21.3 m and 9.5 m, respectively (Table 3). Spectra of flowmeter records obtained at 60 and 13 cm above the bed during the same run are shown in Figure 19b. Energy densities at swell frequency during this run were somewhat higher at 13 cm than at 60 cm, but in most onshore-offshore runs, gravity wave energy was almost identical at all levels. Coherence and phase estimates from the cross-spectra indicate that

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. PERIOD (a) 10 s 1 10 5 3 0 0 -l PRESSURE: RUN 3 600 FLOWMETERS: RUN 3 ••award sanaor (A) * ..... 60 cm •horaward sanaor (B) — 13 cm O 200- 400 UMl 100 A 200i

• V i■fe <»J c^J i Of V o-f v

1-0-1 1.0-1

uttl iuz 0 .5 - 0.5- Uls z o

i 0 . 0 -

(B) lags (A)

0.25 0.50 0.250.75 0.50 0.75 1.00 FREQUENCY (cpa)

Figures 19a-b. Concurrent spectra, and cross-spectral estimates of coherence and phase for records from run 3; between seaward and landward pressure sensors located 10 m apart (19a), and between near-surface and near-bottom flowmeters (19b).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 58

oscillatory flow at the two elevations is highly coherent and in phase. These observations suggest that oscillatory wave motion is reasonably approximated by potential flow, at least down to 13 cm above the bed (Komar 1976, p. 39). A knowledge of wave height, water depth, and wavelength is sufficient to classify surface gravity waves within the bounds of specific wave theories. As is shown in Figure 20, adapted from Komar (1976, p. 61), both sea and swell fall within the range of Airy or Linear Wave Theory (LWT), well below the line beyond which wave steepness is limited by breaking. Because of their limited wavelength, the seas are considered "intermediate water" waves even in 1 m water depth and lie outside the region where LWT shallow water approximations are most valid. Wavelengths determined from the cross-spectra average 6 percent longer than those calculated using shallow water LWT, as is indicated by a slope slightly less than unity in Figure 21a, but otherwise are well predicted by this theory (r2 = 0.92). The oscillatory velocity data plotted in Figure 21b also shows a good fit (r2=0.99) between Urms determined from spectra, and the theoretical maxima predicted. Umax exceeds Unns, as is expected. The

62 percent difference is, however, an indication that orbital velocities calculated from the spectra provide a somewhat inadequate representation of oscillatory flow for the shoaling waves considered here (Cornish 1898). In all runs, at all depths, an assymetry was observed between maximum onshore and offshore velocities, both before and after correction for the low- frequency mean flow component. Corrected peak onshore orbital velocities exceeded offshore maxima by an average of 28 percent.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Lina of limiting wiv« staapnasa

£ '/S W E L L > v EA

H/d 0.1

AIRY WAVE THEORY

Airy Airy Airy ’■N shallow 'o intermediate deep water water water 0.01 0.01 0.1 1 10

Figure 20. The areas of application of the several wave theories as a function of H/d and d/L (from Komar 1976, p. 61). The regions occupied, within the span of Airy or Linear Wave Theory (LWT), by seas and swell monitored during the cold front experiment are shown in stippling.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 60

(A) 2 2

* y = 0.91x ♦ 0.57 14 0.92

6 10 14 18 22 26 WAVELENGTH FROM CROSS-SPECTRA (m) Q

(B)

« 30

y = 1 .6 1x ♦ 0.17 22 0.99

14 22 30 38 46 Urms FROM SPECTRA (cm-*’ 1) x < 2

Figures 21a-b. Relationships between wave parameters determined from the spectra and cross-spectra, and those predicted by shallow water LWT; wavelength (21a), and orbital velocity (21b).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 61

Suspended Sediments Total suspended sediment concentration results from the multi-level sampler are summarized for the pre-frontal, frontal, and post-frontal periods in Table 4. Concentrations ranged from 160 to 9400 mg-l“l. Both concentration and between-sample variability decreased with elevation above the bed. The highest concentrations, which approached those of fluid mud, were monitored at 10 cm above the bed during the frontal period (runs 3-6). Concentrations at this elevation, during these runs, averaged more than 10 times greater than during the pre-frontal period, and 3 times higher than the post-frontal mean. Ratios calculated between concentrations at the 65 and 10 cm levels are shown in Table 4. Near-bottom concentrations (10 cm elevation) during the frontal period were 550 percent higher than at 65 cm, but only 2 and 54 percent greater during the pre- and post-frontal periods, respectively. These results indicate that a significant vertical gradient in sediment concentration was developed in the frontal runs which was not present during the other two periods. Representative concentration profiles from runs 2, 3, 5, and 7, plotted in Figure 22, show the progression of gradient development through the frontal sequence. It is clear that suspended sediment concentrations are not directly related to wave energy as the highest values near the water surface were found during the post-frontal period when wave heights were lowest. One interpretation of the distinctive character of the frontal profiles, relative to those from the pre- and post-frontal periods, is that gradient development may be linked to the frequency composition of the gravity wave energy, rather than to wave amplitude. Ratios of sea to swell wave

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CD ro 0.65(.23) 0.65(.23) 0.61(.28) 0.66(.21) 0.98(.03) 0.98(.03) 0.16(.11) 0.09(.06) 0.09(.06) 0.54(.33) 0.30(.09) 998( 443) 732( 250) 325( 56) 4650(2792) 1155( 358) 3213(3531) 0.13(.07) ...... 618( 86) 1131(455) 439( 29) 466( 18) 773( 391) 0.69(.27) 634(232) 515(129) 320( 45) 325( 56)

820( 61) ( 901 82) 1355( 398)

— 318( 53) 320( 45) — 318( 53) 364(93) 473(185) 533(200) 1658(2020) 413(12) 426( 16) 431(20) 477( 20) Table 4 December, 1982 1.08 582( 46) 1.12 0.94 0.98 0.42 304(21) 355( 4) 0.33 360(54) 341 ( 25) 362( 11) 5666(3597) 0.40 337( 95) 0.39 205(28) 256( 92) 350( 61) 0.98 0.98 Cold Front Experiment 1.03(.08) 425(19) 568(168) 606(196) 0.74(.34) 10-13 0.39(.04) 290(99) 317( 76) 443(152) 3582(3037) Total Suspended Sediment 6 6 2 3 3 6 6 3 4 39 85 96 100 depth Samples Sea/Swell sampled at 4 elev. 65cm/10cm d (cm) 95 cm 65 cm 35 cm 10 cm - mg-h1 -- 5 6 14:23 15:09 100 90 4 13:38 100 3 13:07 9 12:34 100 8 11:30 90 7 10:51 2 14:20 85 10 13:01 105 Run Run No. Time Water No. Height Ratio Mean Cone. (Std. Dev.) Cone. Ratio Frontal Mean 11 Pre-Frontal MeanFrontal 12/11 4 Post-Frontal 12/13 Pre-Frontal 12/10 Post-Frontal Mean 24 ...... Grand Mean

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 100

tTl fcml

E o RUN 2 RUN 3

a * Ui V “ 10 — ,— imu inmi~lQmi|lmjni|i iu > o m <

-i 50 RUN 5 RUN 7

10 J ,li minmfan'lilfl.nnmri 100 1000 10000 SUSPENDED SEDIMENT (mg-f1)

Figure 22. Mean suspended sediment concentrations from representative pre-frontal (2), frontal (3 and 5), and post-frontal (7) runs. Error bars show two standard deviations about the mean.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 64

heights, given in Table 4, provide one measure of the relative importance of these components of the wave field. When the near-surface/near-bottom concentration ratios (65cm/10cm) from each run are plotted against the respective wave height ratios, the data from the pre- and post-frontal periods form one cluster distinctly separate from a second frontal cluster, as is shown in Figure 23. The concentration ratios exhibit large within-run variability, as is indicated by the error bars in the figure. Within-run variability in the wave height ratios cannot be estimated. Even so, when all the data are considered together, a weak linear relationship emerges (r2 = 0.69). This result suggests that swell waves may contribute to the formation of a suspended sediment gradient, which does not develop if higher frequency seas make up an important part of the wave field.

Infragravitv Water Level Changes and Non-Oscillatorv Flow Energy contained in the low-frequency peak below 0.005 cps constitutes between 30 and 60 percent of the total variance in the water level data, and from 7 to 22 percent of that in onshore-offshore flowmeter records. An average of 54 percent of the low-frequency variance in the pressure records is contributed by uni-directional tidal trends in water level which are long with respect to the 20 minute sample length. These can be approximated by a straight line and are readily removed using a linear detrending procedure. Non-tidal water level fluctuations contribute the remaining low-frequency variance. As was noted earlier, concurrent water level spectra obtained 10 m apart differ by less than 5 percent in the gravity wave frequencies, and, overall, indicate no consistent cross-shore pattern with respect to wave height. In contrast, a systematic amplitude increase in the direction of the shoreline is observed below the gravity wave range. The

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. o 5 1 -° oc z o j= 0.8 < OC z E g S 0 .6 -

Osi o IA O* £<<>0.4<0 UJ 2 E w 0.2 Q UJ a z UJ 0.2 0.4 0.6 0.8 1.0 1.2 CL CO Hsea/Hswei! WAVE TYPE RATIO

Figure 23. Relationship between vertical sediment concentration gradient, indicated by the ratio between near-surface and near bottom values, and the relative significance of sea and swell, indicated by the ratio of their respective heights determined from the spectra. Circled numerals correspond to run number, and error bars indicate two standard deviations about the mean concentration ratio for each run.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 6 6

sum of variances in spectral frequency bands below 0.005 cps from the detrended water level records are listed by station in Table 5. Between-station differences provide a relative measure of the landward amplification of the non-tidal, low-frequency water level changes occurring during each run. This amplification was less than 5 percent in 2 of the 7 runs from which data at both (A) and (B) are available, but ranged from 11 to 33 percent in the remainder, and averaged 12 percent overall. Concurrent water level records are in phase at frequencies below 0.005 cps, as the cross-spectral results in Figure 19a show for run 3. An amplitude increase in the direction of the shoreline, without a progressive phase shift over the same distance, is characteristic of a standing wave with an offshore node and an antinode at the shoreline (Suhayda 1972, Waddell 1973). Attention is drawn to the variance values for run 3 in Table 5 which exceed the means for all runs by nearly 300 percent. Low-pass wave and flowmeter data (T > 200 s) from run 3 are shown in Figure 24 and clearly show the source of this energy. Mean water level drops gradually in the first half of this segment, then abruptly rises 10 cm in 340 s. During this period of rapid water level rise, flowmeters at all depths register significant onshore flow. Most of the low-frequency adjustment in flow velocities occurs away from the bed at the upper flowmeter elevations. The records shown in Figure 24 clearly establish that coherent onshore-offshore flows lasting for minutes are associated with non-tidal fluctuations in nearshore water elevation. Local and regional wind data of comparable resolution are lacking. It appears, however, that the wind-forced sequential water level set up and set-down which characterizes the frontal tide record is reflected in higher-frequency changes observed in 20 minute records.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 5 Water Level Variance Below 0.005 cps in Concurrent Detrended Records from Stations A and B 10 m Apart

Run Variance (cm)2 D ifference % No. ------Station A Station B

Pre-frontal 2 1.59 1.78 +0.19 + 12

Frontal 3 7.74 8.63 +0.86 + 11

4 1.18 1.21 +0.03 + 3

5 1.54 1.71 +0.17 + 11

6 1.20 1.21 +0.01 + 1

Post-frontal 7 0.87 1.16 +0.29 + 33

8 1.02 1.14 +0.12 + 11

Mean 2.16 2.41 +0.24 + 12 Std. Dev. 2.47 2.76 0.29 10

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. O) 00

1050 1 3cm 1 750 900600 60cm sr- 30cm sr-

TIME (s) 450 300 RUN RUN 3 LOW PASS OFFSHORE ONSHORE 150 I E o 111 III a Q. 0) T" Figure 24. Low frequency water level and onshore-offshore flowmeter data from frontal run 3. Low pass true zero point are shown in the lower frame. Logarithmic flow periods stippled. water level is superposed on demeaned pressure record in upper frame. Low pass flowmeter velocities about a

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 69

The existence of a connection between infragravity water level changes and cross-shore flow is supported by analysis of the flowmeter variance statistics presented in Table 6. Flowmeter variance less than 0.005 cps increases with elevation above the bed. Covariances were computed between the low-frequency velocity time-series and the low-pass water level records. Then the same comparison was made between the velocity data and the time-series which results from evaluating the change in water level between each time step. This simple derivative of the low-pass water level time-series exhibits positive values when water level rises, and negative values when it falls. Change in the rate of water level rise and fall, represented by the derivative time-series, is positively correlated with the low-frequency flowmeter records and explains, on average, 60 percent of the variance in velocity below 0.005 cps. The undifferentiated water level record explains only 22 percent of the low-frequency onshore-offshore flow. The poor correlation between untreated water level and velocity is not unexpected given that maxima and minima in water level represent end points which are out of phase with cross-shore flow. Physically, as the surface elevation fluctuates about a mean position, high and low water stands are both periods of diminished shore-normal transport. The covariance data in Table 6 indicates that the correlation between the rate of water level change (derivative) and shore-normal velocity is far stronger at the 60 and 30 cm levels than it is closer to the bed, at the 13 cm level. The reason for the decreased correlation there can be seen in Figure 25. This figure shows a progression of velocity profiles extracted at 50 s intervals during the low-frequency flow event shown in Figure 24 (run 3).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 6 Non-Oscillatory Onshore-Offshore Flow: Covariance Between Water Level and Flowmeter Records Below 0.005 cps

Run Flowmeter Flowmeter % Flowmeter Variance No. Elevation Variance Explained by:

(cm) (cm-s-1)2 Untreated Derivative Water Level Water Level

2 60 4.5 - 16 + 70 30 3.4 - 15 + 70 13 2.4 - 7 + 62

3 60 3.9 - 4 + 90 30 2.6 - 6 + 88 13 1.2 - 60 + 49

6 60 3.0 + 10 + 58 30 2.2 + 15 + 63 13 0.9 - 1 + 49

7 60 3.9 - 29 + 76 30 2.9 - 22 + 66 13 1.2 - 20 + 43

10 60 2.9 - 42 + 45 30 N/A N/A N/A 13 0.7 - 41 + 19

Mean 2.6 - 22 + 60 Std. Dev. 1.2 17 19

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

4 T=626 s T=626 3 2 0 1 1 T=575 8 T=575 ONSHORE VELOCITY CCM-S-1) O F F 8H O R E 4 3 T=675» 2 1 0 1 -3 -3 -2 60 20 10 60 40 30 70 20 30 80 80 80 70 40 o u o > D < Ul o Z o > _] I- < 111 ui Figure 25. Sequence of vertical velocity profiles from 525 to 775 s during onshore-offshore flow event in run 3 (see Fig. 24). !

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 72

The sequence demonstrates that both onshore and offshore transport begin as flow reversals close to the bed. This observation lends credibility to the standing wave concept introduced earlier, as near-bottom flow reversals are characteristic of transport predicted under both standing (Longuet-Higgins 1953) and partially reflected progressive waves (Carter et al. 1973). The 20 minute length of the records obtained during this study does not permit resolution of the periodicity of the low-frequency water level fluctuations observed, if, indeed, any such periodicity exists. Establishment of periodic forcing is not required with respect to the kinematics, however, as any perturbation of the free surface will involve a redistribution of mass requiring shore-normal currents. These currents emulate those developed under a true standing wave characterized by a particular node position and frequency. Spectra from flowmeters oriented alongshore have a significant peak only at frequencies below 0.005 cps as the ducts eliminated an estimated 95 percent of the gravity wave signal. Alongshore flowmeter statistics compiled in Table 7 lead to three important observations. First, alongshore flow was to the west throughout the course of the experiment. Second, mean velocities remained relatively constant, ranging from 7.6 to 8.1 cm-s"l at the 60 cm elevation, through the frontal and post-frontal sequence when wave heights dropped significantly. Finally, mean flow velocities in all runs increase with elevation above the bed. Low-pass wave and flowmeter data from run 5 are shown in Figure 26. The dependence of velocity on flowmeter elevation is more pronounced in this alongshore data than in the onshore-offshore record from the same period (Fig. 24). Water level drops through most of this segment but rises

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 73

Table 7 Steady Alongshore Flow: Covariance Between Water Level and Flowmeter Records Below 0.005 cps

Run Flowmeter Mean % Flowmeter Variance No. Elevation Westward Explained by: Velocity ...... (cm) (cm-s-1) - Untreated Derivative Water Level Water Level

4 60 7.8 + 52 + 11 30 4.5 + 49 + 10 13 1.7 + 56 + 11

5 60 7.9 + 57 + 35 30 5.6 + 66 + 40 13 1.3 + 81 + 24

8 60 8.1 + 54 + 20 30 6.2 + 48 + 10 13 3.0 + 40 + 3

9 60 6.0 + 86 - 16 30 N/A N/A N/A 13 1.7 + 82 + 24

Mean 4.9 + 61 + 19 Std. Dev. 2.6 16 11

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ■vl

< < ■■ ■ i w ------1 30cm ------1 ------1 TIME (s) ------1 ------1 ------1 ------1 ------1 RUN RUN 5 LOW PASS ------ISO 300 450 600 750 900 1050 o n. n. ■- w Figure 26. Low frequency water level and alongshore flowmeter data from frontal run 5. Low pass water level is superposed on demeanedpoint are pressure shown record in in the upper lower frame. frame. Westerly Logarithmic low pass flow flowmeter periods velocities indicated about by a stippling. true zero

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 75

toward the end. Flowmeter velocities generally mirror the water level curve, decreasing when water level falls, and increasing when water level rises. As before, covariances were computed first between the low- frequency flowmeter time-series and the untreated water elevation record, and then between the velocity data and the derivative water level time-series. For the alongshore runs, the untreated water elevation data is positively correlated with velocity (positive to the west) and explains an average of 61 percent of the low-frequency flow variance. The derivative of the water level time-series explains, on average, only 19 percent of the flowmeter variance below 0.005 cps. Physically, the positive correlation between alongshore flow velocity and water level means that maximum and minimum velocities occur at high and low water stands, respectively. A rise in water level, then, results in an increase in westerly alongshore flow velocities at the instrument array. The depth dependence of the mean alongshore velocities in Table 7 suggests that the alongshore current might be well characterized as a boundary shear flow. The physical constraint that velocities must be zero at the bed gives rise in a neutrally stable flow to a boundary modified layer in which mean steady horizontal velocities typically approach free-stream values as a logarithmic function of elevation above the bed (Adams and Weatherly 1981a). The equation which expresses the relationship of horizontal velocity to depth in the bottom boundary layer has the form,

u* z U= — In— (3) K zo

/ Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 76

where U is the velocity measured at a reference height z above the bed and K is von Karman's constant (0.4). This profile is fitted to the data by minimizing the square of the velocity deviations with respect to a characteristic shear velocity, u*, and an apparent bottom roughness length, z0 (Wilkinson 1984).

In the present investigation, low-pass velocities averaged over 5 s intervals at each of the three flowmeter elevations were fit to equation (3) to obtain 240 profiles, with associated estimates of u* and z0, for each 20

minute record. A profile was defined as logarithmic when A, the coefficient of determination, exceeded 0.9755, the t-test criterion for a 90 percent confidence level in a three point fit (Adams et al. 1982). Low-frequency onshore-offshore flows met this test between 5 and 25 percent of the time. Alongshore flow, in contrast, was logarithmic for 65 to 85 percent of the runs where three point velocity data was available (runs 4, 5, and 8), and was close to logarithmic most of the rest of the time. Periods of significant logarithmic flow are shown in the onshore-offshore record in Figure 24 and in the alongshore data in Figure 26. A subsample of velocity profiles meeting the logarithmicity test was selected from periods of steady flow when the mean current was undergoing neither acceleration or deceleration. It is important to note here, for consideration later, that no periods of steady flow were found in the onshore- offshore data. Ten steady flow profiles with r^ values above 0.99 were identified, however, in each of alongshore runs 4, 5, and 8. Mean values of u*, z0, and the velocities at each reference elevation derived from these

subsamples are given in Table 8 and were used to generate characteristic steady flow profiles for each of these runs.

/ Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 77

The characteristic velocity profiles are shown in Figure 27. In these semi-log plots the y-intercept corresponds to z0, the apparent bed roughness, while the slope is K/u*. The profiles for runs 4 and 5, which were obtained

less than 1 hour apart, are similar, while that from run 8 in the post-frontal period differs significantly. At 60 cm, velocities in runs 4 and 5 exceed that for run 8, while at 30 and 13 cm the reverse is true. Values for z0 in all of

these profiles suggest the presence of bottom roughness elements 2 to 5 times the amplitude of any actually found on the smooth mud bed at the experiment site (2 cm maximum). In briefly reviewing the results of the cold front experiment, we note that both sea and swell waves at the study site were well described by shallow water LWT. The major effect of front passage on the wave field appears to be a modulation of the relative importance of the high-frequency sea component. A significant vertical gradient in suspended sediment concentration was developed during the frontal period under essentially monochromatic swell waves (runs 3 - 6). This gradient was not present during the pre- or post-frontal runs, when seas contributed more significantly to the spectral energy distribution. Water level oscillations in the infragravity region were observed to be correlated with unsteady cross shore flows. The low pass alongshore current, in contrast, was found to be a fully developed, and relatively steady, boundary shear flow.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 78

Table 8 Characteristic Alongshore Velocity Profile Parameters

Run Time Water u* z0 Velocity % — at— Log No. Depth 60cm 30cm 13cm. (cm) (cm-s-1) (cm) (cm-s-1)

Frontal Period 12/11 4 13:38 100 1.9 10.1 8.3 5.0 1.2 85

5 14:23 100 1.8 10.1 8.0 4.8 1.1 65

Post-Frontal Period 12/13 8 12:34 85 1.2 5.1 7.4 5.3 2.8 80

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 79

100

— RUN 4 inO m RUN 5

run 3

a aUJ UJ > O a < z o < > UJ m l UJ

0 2 4 0 8 10 14. 1 12

VELOCITY

Figure 27. Characteristic logarithmic alongshore velocity profiles for frontal runs 4 and 5, and post-frontal run 8.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. DISCUSSION

The results of this study show that mudflat deposition has ceased east of Freshwater Bayou Canal, but remains active in the western half of the study area. Mudflats form there only in late winter and early spring, and are characterized in cross-section by distinctive 2 to 10 cm thick beds displaying both grading and flat-lying basal silt or shell laminae. The goal of monitoring nearshore dynamics over a broad range of wave energy conditions was achieved during the December cold front experiment. Fine grained sediments were in motion during this event, but sequential bed profiles conducted in the course of the experiment showed no significant changes indicative of accretion or erosion. A conspicuous lack of sedimentation is not surprising given the intermittency which characterizes mud deposition in this area. It is, however, an important indication that all factors necessary for nearshore mud deposition were not present at the time of the experiment. Significant water level set-up and set-down were associated with the characteristic clockwise rotation in wind direction which accompanies front passage. These effects were clearly manifested in departures from the predicted tide, but also showed up as infragravity oscillations in 20 minute nearshore water level records. These oscillations exhibited characteristics of a standing wave with an antinode at the shoreline. Significant low-frequency, onshore-offshore flows were correlated with these shore-amplified water level changes and support an assumption of standing wave motion. The sea and swell waves monitored during the cold front experiment were found to be well described by shallow water linear wave theory, but

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 81

also experienced the nearshore attenuation documented by Wells and Coleman (1981a, b) on the muddy coast of Surinam. A significant near bottom gradient in suspended sediment concentration was observed under monochromatic swell waves, at the time of front passage, which was absent when higher-frequency seas contributed significantly to the total wave energy. Mean alongshore velocities near the surface responded to low- frequency changes in the depth of flow at the flowmeter array, rather than to changes observed in the incident wave field. This current consistently set to the west in spite of the sequential shifts in wind direction which characterized the cold front progression. The vertical velocity distribution of the alongshore current, unlike that of the the low-frequency shore-normal transport, was indicative of a steady, logarithmic boundary shear flow. Apparent bottom roughness estimates (z0)

were, however, 2 to 5 times greater than the amplitude of any physical roughness elements present at the experiment site. A marked decrease in both apparent bottom roughness, and the characteristic shear velocity (u*),

was noted in the transition between frontal and post-frontal profiles. Now that we have some information about the nearshore dynamics, it is appropriate to re-examine in more detail the boundary layer sorting model (B-L-S model) proposed by Stow and Bowen (1980). Their representation of the near-bottom dynamical environment is reproduced in Figure 28. Stow and Bowen (1980) inferred, on the basis of sedimentological data alone, that the origin of graded and laminated silt + mud beds must depend on a fundamental difference in the way suspended cohesive and non-cohesive sediments respond to boundary layer shear. Differential settlement of these two types of particles in a waning flow could not explain the within-bed

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission SEA SURFACE

— z = h

TURBIDITY CURRENT

SUSPENSION

SHEAR

BOUNDARY 10- LAYER

>VISCOUS SUB- { \ LAYER SETTLING lam inar SUB-LAYER

Figure 28. B-L-S model for sorting of cohesive and non-cohesive sediments in the boundary layer of a turbidity current (from Stow and Bowen 1980). Elevation, z, measured from the bed in a millimeter logarithmic scale.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 83

sorting they found in fine-grained, shelf margin turbidite sedimentation, as clay floes and silt grains with equivalent settling velocities would be expected to deposit together. According to the B-L-S model, initially only silt grains can settle through the high-shear zone to form the characteristic basal laminae, while clay floes approaching the bottom are disaggregated and returned to the flow interior. Stow and Bowen (1980) suggest that clay concentrations build up above the boundary shear layer to the point that aggregates form which are sufficiently large to "overcome" shear. At some critical concentration, they penetrate the high shear zone, "break-up", and deposit rapidly to form the characteristic mud "blanket" overlying the silt laminae. Following deposition of a silt + mud couplet, the process is repeated at the new surface, such that a sequence of stacked laminated and graded beds is developed. Observations made during the course of the present study indicate that the first part of this process, the selective passing of non-cohesive particles through the shear zone, could well explain the flat-lying basal laminations found in mudflat sedimentation. It is not clear, however, how large clay floes, which initially could not survive passage to the bed, could later resist disaggregation by the same shear stress. An alternative mechanism for deposition of the mud part of the silt + mud couplet is proposed here which does not require any diminution of boundary layer shear. This approach is founded on the assumption that the mud suspension supported above the high shear zone acquires yield strength as its sediment concentration rises. Sudden deposition might then occur, as suggested by Einstein (1941), without break-up of the floes, through a "freezing" process, similar to that observed in mudflows on land (Johnson 1970). Freezing would be initiated when the

/ Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. yield strength of the suspension'reached equivalence with the boundary layer shear stress. As yield strength is a property which characterizes fine-grained sediment suspensions only at fluid mud concentrations, deposition by the mechanism proposed requires the presence of a fluid mud layer above the high shear zone. The dynamics data collected in the present study permit a more detailed investigation of parameters relevant to boundary layer sorting than has been possible in the past. In the first section below, we address the problem, also recognized by Einstein (1941), of how turbulence, generated in this case primarily by wave shear at the bed, is prevented from propagating into the flow interior, where it would prevent development of the significant near-bottom sediment concentrations required. This analysis establishes a potential for the intact transport of a fluid mud layer by a uni directional current. In the second section below, we consider the potential for nearshore deposition of fluid mud transported along the unusual cross shore shear stress gradient developed under progressively attenuated waves. A scaling argument is advanced to explain why deposition did not occur during the cold front investigated, and, more importantly, when it could. This leads, in the third section, to a discussion of the temporal and spatial connections between coastal mud deposition, the shelf sediment dispersal system, and the Atchafalaya River. Predictions for the future of the chenier plain are made in a fourth part, based on projections of Atchafalya delta development. Finally, generalizable features of the chenier plain model are extrapolated to other muddy coasts, both modem and ancient.

permission of the copyright owner. Further reproduction prohibited without permission. 85.

FORMATION AND TRANSPORT OF FLUID MUD Implicit in the simple logarithmic boundary layer model defined by Equation (3) is the assumption that a single boundary layer length scale is sufficient to fully characterize interaction between the flow and the bottom (Tennekes and Lumley 1972, p. 146). Elevated estimates of the bottom roughness, such as those derived from the alongshore velocity profiles in this study, are common in shear flows where additional length scales play a role (Adams and Weatherly 1981a). Multiple scales arise when the flow is influenced by turbulence from sources other than that due to the steady current boundary shear, or when turbulence production is restricted by a density gradient. Turbulence generated by unsteady oscillatory wave flows at the bed gives rise to a second, shorter length scale, that of the wave boundary layer (WBL), which must be considered in addition to the scale associated with the steady current boundary layer. Grant and Madsen (1979) have developed a widely applied method for parameterizing the effects on the steady current which result from combined wave and current interactions in the WBL. Their analysis, which will not be repeated here, indicates that this interaction is non-linear and results in estimates of bottom shear stress which are greater than the sum of wave or current shear stresses treated separately. The steady current in the region above the WBL, in which wave motion is characterized by non-turbulent potential flow, experiences a resistance which depends not only on the physical bottom roughness, but also on characteristics of the WBL. The velocity profile above the WBL has a form similar to that given in Equation (3):

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 86

u* 30z U = — In — Z > 8yy (4) K ZcW

where the only new term is ZcW, an "apparent roughness length" containing the combined effects of the physical roughness (z0), and that due to the

presence of a WBL of thickness Sw. Inside the WBL, the appropriate roughness is once again z0, the amplitude of the bottom relief, and the

velocity profile is determined by:

u* u* 30z U = — ( — ) In — z < 8W (5) K u*cw zo

where u*cw is the characteristic turbulence intensity inside the WBL due to

the nonlinear wave and current interaction. Maximum shear stress at the bed due to both waves and the current is given by:

Tew,max = P(u*cw)^ (6)

where p is the density of the fluid. The thickness of the WBL is estimated as:

2K(U*CW) (7) (0

WBL thickness is thus proportional to u*cw and inversely proportional to co,

the radian wave frequency.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 87

Values for u*, u*Cw> zcw> zo> Tcw,max> and §w are determined from

the velocity at a reference elevation in an iterative closure dependent on a small number of wave and current parameters (See FORTRAN listing in Appendix B). Results of this analysis, based on current velocities at 13 cm and swell parameters for the three alongshore, cross-wave runs in Figure 27 are given in Table 9. Calculated WBL thicknesses (8W) for the runs range from 1.5 to 2.0

cm, roughly 2 orders of magnitude less than that of the current boundary layer, which can be considered to cover the full depth of the flow (100 cm). The characteristic turbulence intensity in this limited region, u*cw, is nearly

an order of magnitude greater than that determined in the model for the flow above the WBL, from 1.5 cm-s-! in run 8 to 2.2 cm-s-1 in run 4. The estimated bed shear stress due to the combined wave and current motion ranges from 2.1 to 4.6 dynes-cm-2 for runs 8 and 4, respectively. Calculated WBL shear velocities are competent to suspend fine to medium sand (McCave 1984). More significantly, bed shear stresses equal or exceed Bingham yield strengths experimentally determined by Krone (1962, 1963) and Owen (1975) for fluid mud suspensions in the 20 to 100 g-l“l range (Fig. 16). These shear stresses are, however, far below those required to maintain in suspension a slurry with the concentration of the surficial mudflat sediment, which, the results of this study suggest, has a yield strength greater than 100 dynes-cm-2 (Fig. 15). The large magnitude of the wave-induced flow, relative to that of the steady current, leads to estimates of an apparent roughness length, zcw,

which range up to 36 cm (Table 9). This is the amplitude of physical roughness elements which the model predicts would produce the same drag

/ Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. OO oo

1.48 1.58 2.02 ...... 2.1 4.3 4.6 cw.max 'w -cw (cm) (dynes-cnr2) (cm) —Model Output

......

------0.79 0.40 1.47 0.02 11 1.08 0.23 2.12 0.02 34 Table 9 December, 1982

Cold Front Experiment 10-13 11 20 ...... Grant-Madsen Boundary Layer ModelO) Wave Effects on Steady Alongshore Flow

18 Model Input 1.1 ref. elev. Vel. Amp.(2) 05 ...... (cm) (cm-s-1) (cm-s-1) (cm) (rad-s-1) (cm-s-1) 100 Frontal Period 12/11 Post-Frontal Period 12/13 OIGrant and Madsen (2)From (1979) LWT Run Depth Velocity Maximum Oscillatory Wave u* u*cw ; 8 90 2.8 10 No. at 13 cm Orbital Excursion Freq. 4 100 1.2 18 20 0.87 0.23 2.20 0.02 36

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission 89

on the steady current as the combination of physical roughness, and the non linear interaction of oscillatory and non-oscillatory flows in the WBL. This resistance is transmitted to the flow interior in calculated values for u* (Table

9) which are nearly an order of magnitude lower than estimates derived from the three-point velocity measurements given in Table 8. As a result, the Grant-Madsen model predicts, for all the alongshore runs, velocity profiles far steeper than the characteristic profiles shown in Figure 27. This observation suggests that the current does not "feel" the drag generated in the WBL, at least to the degree predicted by the model. One explanation for this result may be that WBL turbulence, while important for maintaining sediment in suspension close to the bed, is largely destroyed before it reaches the interior of the flow. An assumption of neutral stability, or the lack of significant turbulence damping features in the flow, is implicit in both boundary layer models introduced thusfar. The increased slope (decreased u*) and lower z0 which

characterize the post-frontal velocity profile (run 8) in Figure 27, relative to those from the two earlier runs is strongly suggestive of a change from a stably stratified flow to one that is more neutral. Gradients in suspended sediment concentration, like those due to salinity and temperature, can produce a stable density field capable of inhibiting the vertical eddy diffusive flux of turbulence (Smith and McLean 1977a). The suspended sediment profiles shown in Figure 22, and the more complete data in Table 4, clearly indicate the presence of a significant near-bottom concentration gradient during runs 4 and 5, which was developed to a far lesser degree during run 8. Gravity acts on the suspended sediment particles to oppose the upward flux of turbulence produced by shear at the bed. The flux Richardson

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 90

number provides a dimensionless measure of the relative magnitudes of these forces. Smith and McLean (1977a) give the appropriate form of this ratio as:

(ps-p) dCs du Rf=-g r~~■ —1(~)-2 (8) p dz dz

where g is the force of gravity, dCs/dz is the volumetric sediment

concentration gradient (z positive up), du/dz is the horizontal velocity gradient, and ps and p are the respective densities of the sediment and the

fluid. It can be seen that the gravity force, which acts on the excess mass of sediment, represented by the bracketed term, is balanced against turbulence produced by the velocity gradient. A flow characterized by a positive Rf number is stably stratified, and acts as a sink for turbulent energy. Rf values

calculated for the velocity and suspended sediment gradients between 13 and 30 cm elevations during the alongshore runs range from +0.01 for run 8 to +0.10 in run 5. At values of Rf greater than approximately +0.20,

laboratory experiments show that turbulent flux is completely suppressed (Tennekes and Lumley 1972, p. 99). Einstein and Krone (1962) found that differential consolidation of fluid mud, a process which reduces water content in the lower part of a layer at a more rapid rate than in the upper part, results in a characteristic increase in sediment concentration with decreasing elevation above the bed. If fluid mud was present below the 10 cm elevation of the bottom-most sampler during runs 4 and 5, and turbulence associated with the wave motion is generated within a WBL 1 to 2 cm thick, the magnitude of the Rf number

might be expected to increase at least to the top of the WBL. The effect of

/ Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 91

such a stable density gradient will be to limit the vertical flux of turbulence generated at the bed and thereby restrict sediment dispersal into the flow interior. One consequence of a restriction on the vertical flux of turbulence is the emergence of a mechanism whereby a coherent fluid mud layer might be transported in the direction of the mean flow. As Einstein and Krone (1962) argued cogently 24 years ago, the vertical density and yield strength gradients which characterize such suspensions render their transport en masse by a uni-directional current alone improbable. Shear stress applied only at the upper surface would initiate movement there first, with the result that sediment would be stripped off and dispersed, while the denser suspension closer to the bed remained stationary. Returning once more to the B-L-S model (Fig. 28), however, we note that a sorting mechanism which acts to segregate silt particles from a much larger population of clay floes in the mixed suspension above the high shear zone must cause a break in the sediment concentration gradient close to the bed. Given the relationship between sediment concentration and yield strength, a concentration drop imposed by shear at the bottom of a fluid mud layer would also be a point of reduced yield strength. Transport of an intact fluid mud layer might then occur, under the influence of shear exerted at its upper margin, along a shear plane defined in 2 dimensions by the drop in yield strength at the top of the high shear zone, in this case the WBL.

KINEMATICS OF FLUID MUD DEPOSITION AT THE SHOREFACE Having developed the potential for transport of sediments contained within a fluid mud layer under the influence of a unidirectional flow, the task is now to identify such flows nearshore, particularly those with an onshore

/ Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 92

orientation. A re-examination of the characteristic alongshore velocity profiles (Fig. 27) suggests, however, that near-bottom current velocities in these steady flows are so low that significant transport of sediment in a fluid mud layer would be unlikely. All nearshore flows monitored during the cold front experiment were not steady, however, and would not be expected to reflect this equilibrium condition. This is particularly true of the shore- normal flows observed, which, though sometimes logarithmic, continually undergo acceleration and deceleration. This evolution is apparent in the sequence of velocity profiles from the low-frequency flow event captured during run 3 (Fig. 25). Unlike the characteristic alongshore profiles, velocity profiles associated with shore-normal flow suggest a potential for non-equilibrium sediment transport close to the bed. As noted previously, the low-frequency fluctuations in water level, which are correlated with significant unsteady onshore-offshore flow, have characteristics of a standing wave with an antinode at the shoreline and an offshore node. Suhayda (1972) has shown that, for a linear wave field on a sloping bottom, the frequency of a standing wave which has a node X meters offshore is given by:

2.4 (g tan )l/2 f ------( 9 ) 4 tcX1/2

where tan O is the bottom slope. The period of the low frequency oscillation in run 3 (Fig. 24) is estimated to be 600 s (0.0017 cps). The bottom slope is 0.004. Equation (9) can be rearranged to solve for the distance of the node from shore,

with permission of the copyright owner. Further reproduction prohibited without permission. 93

5.76 g tan O X = ------(10)

(4 7i f ) 2

which, in this case, is found to be 400 m. It was observed that the amplitude of the low frequency oscillations increases in the direction of the shoreline or antinode. In the present example, the amplitude (H/2) of the low-frequency oscillation at sensor B was 4.7 cm. The amplitude at sensor A, 10 m seaward, was 0.17 cm less, a value which yields a maximum between-sensor slope of the still water surface, tan N, of 0.00017. When this slope is linearly extrapolated offshore, the node position is predicted to lie 369 m from the shoreline, a value in reasonable agreement with the result from Equation (10). The shoreface geometry at station 6 is shown schematically in Figure 29. A control volume approach was adopted to estimate the depth and time- averaged velocities required to accomodate standing wave-induced flux through water column cross-sections spaced at 25 m intervals along the shore-normal transit shown in this figure. Onshore/offshore velocities between the shoreline and 800 m offshore were calculated for 3 standing waves, each with a 4.7 cm amplitude at sensor B, but with node positions ranging from 200 to 800 m offshore, and the corresponding periods determined from Equation (9). Results are plotted in Figure 30a for conditions of rising water at the shoreline antinode. Except for a change in sign, the results are identical for falling water conditions.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1.0 ANTINODE ANTINODE TAN N NODE

0.5

200 400 600 800I P ressu re S en so rs 5.76g tancp - ° - 5 r ( b ) x^ V ( a ) X= (4nf)

//S. TAN0

-1 .5

- 2.0

-2 .5

Figure 29. Generalized shoreface geometry, showing slopes used in determining standing wave node position, and changes in the cross- section affecting cross-shore velocities.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 95

(A)

200m NODE

400m NODE 800m NODE i"'*"‘* ......

> H O o Ul-I > ONSHORE FLOW OFFSHORE FLOW

-2

-4

l i t * '

.1*'*“— SEAS (T: 3.1s)

— SWELUT: 6.8s)

0 " _i _____ .— .— i 100 200 300 400 500 600 700 800 X (m) DISTANCE FROM SHORE

Figure 30a-b. Cross-shore trends in low-frequency flow under standing waves (30a), and bed shear stress due to waves attenuated at a constant H/d ratio of 0.2 (30b).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 96

The model predicts that a standing wave with the amplitude observed during run 3, and a node located 400 m offshore, would generate a mean depth-averaged onshore velocity during the rising water phase of 3.8 cm-s-1 at sensor B. This is twice the actual depth- and time-averaged low frequency onshore velocity monitored at this point between 600 and 900 s in run 3 (Fig. 24), but, given the many simplifications, the result is encouraging. It can be seen in Figure 30a that, although the higher frequency, shorter wavelength standing waves generate greater shore-normal velocities for the same amplitude, cross-shore current speeds associated with a 750 s oscillation (node 800 m from shore) would still be significant. The shape of these curves is more important than the velocities calculated in this example. For all three standing waves, the onshore velocity peak is shifted significantly shoreward of the node position it would occupy if the bottom did not slope. As a result, deceleration of onshore, and acceleration of offshore flows, due to standing waves of any period, will be most pronounced within 100 m of the shoreface. The existence of waxing and waning flows at a position fixed by geometry is of great interest with respect to nearshore sediment transport and deposition. It does not completely explain preferential deposition at the shoreface, however, as both onshore and offshore velocities under a standing wave are of the same magnitude. Waves observed during the frontal and post-frontal periods were attenuated completely before reaching the shoreface. Wells (1978) found that attenuation across a Surinam mud bank was a linear function of depth. The characteristic ratio of H/d identified in Surinam (0.2) also appears to act as a limit to wave heights monitored in this study (Fig. 20). If this ratio is assumed to be constant across the nearshore profile shown in Figure 29, peak

/ Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 97

bed shear stresses generated by waves alone may be estimated at any point using an equation derived by Krone (1976) from LWT:

pjc flHrms/T Tmax = 1-7 p (— ------(11) T \x sinh (2jcd/L)

where ji is the coefficient of viscosity. The empirical coefficient of 1.7 is included to account for laboratory results which indicate that, without it, this equation underestimates shear stress based on significant wave height (H j /3 = 1.4 Hrms) by 20 percent (Krone 1976). It should be noted that, for shallow

water waves, both the period, T, and the depth, d, enter Equation (11) through the wavelenth, L = T (gd)l /2} as well as where explicitly stated. Results of shear stress determinations at 25 m intervals are plotted in Figure 30b for the characteristic sea and swell wave periods, 3.1 and 6.8 s, respectively. Two points arise from this analysis. First, it is clear that shorter wavelength seas generate greater peak bed shear stresses than do swells of equal amplitude, attenuated at the same rate. Second, shear stresses due to waves of both periods increase with distance from the shoreline. The presence of a shore-normal wave shear stress gradient which is the reverse of that found on surf-dominated coasts has clear implications for sediment deposition, particularly when considered together with the possibility for cross-shore transport of fluid mud. If a fluid mud layer formed and maintained at one level of WBL shear is carried shoreward, the potential exists for deposition when its yield strength exceeds the lower

/ Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 98

nearshore WBL shear. Mud transported offshore remains supported and is

less likely to deposit. The effect of wave period is less apparent. Though hardly conclusive on this point, the results suggest that high-frequency seas may inhibit development of a fluid mud layer. Wells (1978) has indicated that the converse also appears to be true; that is, that fluid mud filters high-frequency waves from the spectrum. Data from the chenier plain do not permit resolution of this question. The thickness of the WBL is, however, known to be a function of frequency (Eq. 7). It is probable that other factors related to the state of development of the WBL may also influence the ability of a flow to support a coherent fluid mud layer, particularly, if, as in most real systems, several wave trains are superposed. The theoretical difficulties in characterizing a simple system, with only two appropriate boundary layer length scales, are significant, but they are magnified dramatically if multiple WBL length scales must be considered. A range of frequencies then become important, and, particularly for locally generated seas, a variety of

directions. Mud deposition through boundary layer dumping is triggered in the modified B-L-S model whenever the yield strength of the fluid mud suspension attains equivalence with wave generated shear stress. An abrupt reduction in near-bottom suspended sediment concentration would occur immediately after each dumping event. This reduction would permit reformation of the WBL above the new deposit. The cycle of sediment sorting, concentrating, and dumping might then be repeated, and give rise to the stacked sequences noted in mudflat cores. It is important to note that accretion can occur, however, only if sediment continues to be added

/ Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 99

primarily from above the WBL, rather than from resuspension of sediment already deposited. High rates of mudflat sedimentation through WBL dumping are possible if suspended sediment is continuously advected into the vicinity of the shoreface. No high-resolution measurements of the rate of mudflat sedimentation have been made. The best available observation is that of

Morgan et al. (1958)who noted that a layer of mud 18 inches thick (46 cm)

was deposited over a span of3 days. Based on this, a lower bound may be placed on the sedimentation rate if it is assumed.that the deposit has a density

of 1.3 g-cm-3, and, less realistically, that sedimentation was continuous

throughout the 3 day period. A deposition rate is calculated on the order of 1 g-m-2-s-l, 2000 times greater than the average rate for continental shelf

sedimentation estimated by McCave (1984). Instantaneous rates several orders of magnitude higher could apply during the boundary layer dumping events envisioned, such that a single bed 10 cm thick might be deposited in a matter of seconds. Sedimentation rates

ranging between 2 and 50 kg-m-2-s-l are calculated if deposition is assumed

to occur over a period of 1 to 30 s, and the density of the deposit formed is

again estimated at 1.3 g-cm-3. Deposition at this rate from a im thick water

column implies a depth-averaged sediment concentration of nearly 50 g -H . If the nearshore introduction of sediments at these concentrations is required for mudflat sedimentation, it is clear why mudflat formation is such an intermittent process, and, more specifically, why deposition did not occur during the cold front monitored. The highest depth-averaged concentrations found during this experiment, estimated by logarithmically extrapolating the mean concentration data from the frontal period to the bed, were less than

/ Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 100

2.5 g-l"l. One interpretation of this scaling is that, during the periods when mudflat sedimentation is active, the flux of mud advected to the coast must be an order of magnitude higher than was observed during the cold front experiment. The stated purpose of this investigation is to examine the processes by which a coast incorporates influxes of mud intercepted from a shelf dispersal system. A purely kinematic model has been proposed to explain preferential deposition of mud at the shoreface. This is an important first step. Results of the long-term shoreline monitoring program, historical data, and the sedimentation rate scaling just concluded, indicate, however, that factors affecting sediment supply operate at far greater temporal and spatial scales than have been considered thusfar. Attention is directed now to these regional influences.

SHELF SEDIMENT DISPERSAL The high abundances of illite and illite-smectite mixed-layer clays in mudflat sediment suggest an origin in modem Atchafalaya flood discharge, rather than reworked older shelf deposits (Johns and Grim 1958, Brooks 1970), or low-flow Atchafalaya input (Van Heerden 1983), both of which tend to be richer in smectite. A close temporal connection between mudflat sedimentation on the chenier plain and the flux of Atchafalaya sediments to the shelf seems likely. Littoral mud deposition in the study area occurred only during late winter and in the spring when the River was in flood. The high winds, waves, and water level fluctuations accompanying front passages are also important, but frontal weather systems traverse this region from September through April, a far longer period, which includes, but is not

/ Reproduced with permission of the copyright owner. Further reproduction prohibited without permission 101

limited to, the 5 month window identified for coastal deposition. The fact that mud sedimentation did not occur during the December experiment, and does not occur generally, prior to peak flow on the Atchafalaya in late winter and early spring, suggests the need for a volume or availability of advected sediments nearshore which is attained only at that time. The dynamical environment on the shelf is quite different from that identified nearshore along the muddy coast. Turbulent shelf currents have the potential to transport high concentrations of fine-grained sediments dispersed through the full depth of the flow, rather than confined close to the bed. Sediment concentrations at the surface within the shelf "mud stream" offshore of Atchafalaya Bay range up to 800 m g-H during river flood (Wells and Kemp 1981), and are higher than any measured near-surface on the muddy coast (Table 4). This sediment appears to become concentrated in near-bottom fluid mud layers where the characteristic turbulent eddy length scale within the shelf flow is restricted, perhaps by a decrease in depth, and WBL shear stress prevents deposition. Unsteady onshore-offshore flows, associated with the wind-forced, shore-amplified water level oscillations which precede front passage, then capture fluid mud for transport across the shear stress gradient developed under progressively attenuated waves. Stacked graded and laminated mud beds form accretionary deposits at the shoreface through repeated dumping of fluid mud suspensions by the wave boundary layer. Coastal progradation occurs when these mudflat deposits are stabilized by marsh vegetation. While the kinematics remain the same, a direct temporal link with the period of river flood is not required in the special case of mudflat deposition during hurricanes. Under normal cold front conditions, deep-water waves

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 102

offshore are too small to exert significant shear stress at the bed (Adams et al. 1982). Hurricane-generated waves, however, would be capable of reworking and suspending large volumes of mud from deep-water shelf storage sites. Fluid mud might form farther offshore, and be transported a greater distance cross-shelf, by the "standing wave" flow associated with a storm surge several meters in magnitude.

FUTURE CHENIER PLAIN PROGRADATION The spread of clastic beaches in the area east of Freshwater Bayou Canal provides an indication that littoral mud deposition is modulated by changes in sediment flux over periods far longer than a single year. Mudflat deposition and shoreline progradation west of the Canal continues at approximately the same rate today as during the 1950’s and 1960's (Morgan and Larimore 1957, Coleman 1966, Morgan and Morgan 1983). The lateral extent of the muddy coast has decreased, however, as the eastern margin of the zone of active mudflat formation has shifted 10 km west without any corresponding extension on its western boundary. It is clear that less mud is being deposited in the study area today than 20 or 30 years ago. The subaerial delta in Atchafalaya Bay is now 13 years old, and covers nearly 30 km^ (Wells et al. 1983). Sediments entering the bay in river flow are dispersed across the delta through an intricate system of bifurcating channels which did not exist 20 years ago. Prior to the emergence of sandy islands, scour by waves maintained a constant depth and limited sedimentation in the shallow, open bay (Thompson 1955). Now, fine grained sediments which formerly reached the shelf, particularly during low flow periods, are deposited in channel fills and on extensive algal flats

/ Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 103

formed between natural levee ridges (Van Heerden et al. 1981, Van Heerden 1983). The amount of mud trapped relative to total fine-grained sediment flux to the shelf is unknown, but evidence from the chenier plain suggests that the presence of the new delta may be exerting an influence on downdrift deposition. Wells et al. (1983) have projected that the subaerial delta will continue to grow within the confines of Atchafalaya Bay through the end of this century, but will emerge on the inner shelf by 2030. Mudflat deposition on the eastern chenier plain may, therefore, be expected to continue at approximately the rate observed in this study for the next 50 years. When delta development begins on the open coast outside the shell reef, it will be exposed to greater erosion by waves, and can be expected to expand at a slower rate. More fine-grained sediments will be delivered to the coastal mud stream, particularly under low flow conditions, and mudflat sedimentation downdrift should increase dramatically.

IMPLICATIONS FOR OTHER COASTS AND THE GEOLOGIC RECORD Observations made during this study indicate that the only strict requirement for mud sedimentation at the shoreface is that a large flux of fine-grained sediments must be delivered close to shore. A high wave energy regime, for example, is no barrier, as waves, and die boundary layers they generate, play an important role in the formation of fluid mud layers nearshore. A large tidal excursion, likewise, poses no restriction on fine grained sedimentation at the coast. Rapid changes in water level are critical to the onshore transport of sediment contained in fluid mud layers, therefore semi-diurnal, rather than diurnal tides should be more effective for the same range. A 3.2 m semi-diurnal tide almost certainly provides the same forcing

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 104

for cross-shore flow along the Surinam coast (Wells and Coleman 1981b, Rine and Ginsburg 1985) as the wind-generated water level oscillations observed on Louisiana's chenier plain. Where a turbid coastal current is not developed, large storms can cause intermittent mud deposition at the beach. Storm waves suspend large volumes of fine-grained sediments stored offshore, and storm surges provide the cross-shore flows necessary for transport to the shoreface. Such deposition has been observed outside of the period of river flood along the Louisiana coast (Morgan et al. 1958) and on the Rio Grande do Sul coast of southern Brazil (Delaney 1963, Martins et al. 1979). Large-scale coastal progradation associated with mud deposition at the shoreface requires a continuous, rather than intermittent, supply of large volumes of mud, as sediment-starved littoral deposits are quickly destroyed by normal wave action. Continuity of fine-grained sediment provenance is dependent on the steadiness of sediment introduction to the shelf, and on the spatial stability of shelf dispersal. Seasonal and longer-term changes in these factors characterize all real systems to a greater or lesser degree. Thus, all muddy coasts can be expected to exhibit a mix,proportionate to this variability, of coarse-grained and muddy facies at the receiving shoreface. Onshore transport of fluid mud by unsteady flow under standing waves is developed only on the foreshore, where the significant change in bottom slope plays an important role. This restricts mud deposition to shallow water in the immediate vicinity of the shoreface, and thereby limits the maximum thickness of the deposit. Intermittent littoral mud deposits associated with storms will have poor preservation potential, but mudflats that form part of a major

B-«rr*iWW

/ Reproduced with permission of the copyright owner. Further reproduction prohibited without permission 105

progradational sequence may survive if regional subsidence or uplift produce gradual changes in the relative position of sea level. In cross- section, all shoreface mud deposits should exhibit the characteristic laminated and graded bedding which Stow and Bowen (1980) ascribe to boundary layer sorting. Lower beds will be little disturbed by marine organisms, because of the high rate of deposition associated with boundary layer dumping. Layers close to the upper, subaerially exposed surface will, however, display cracks, burrows, trails, and evidence of rooting. Walker and Harms (1971,1976) have described in detail what they interpret as coastal mudflat deposits from an ancient chenier plain setting. This facies is found in more than 20 predominantly muddy, marine- nonmarine sequences in the Upper Devonian Catskill Formation of central Pennsylvania. The sequences represent episodes of coastal progradation on a subsiding coast, which, in some cases, can be traced for 30 km down basin. Transgressive sands define the boundaries of each cycle. Indeed, the shales which lie at the top of the regressive sequences display precisely the structures which are predicted for an accreting littoral mudflat deposit. Rootlets and dessication cracks in some cases occur less than 2 m above marine faunas.

/ Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CONCLUSIONS

Fieldwork conducted between December, 1981, and May, 1983, along a 30 km section of Louisiana's chenier plain coast disclosed widespread shoreline retreat, averaging 5 to 10 m-yr-1. Intermittent sedimentation of shore-welded mudflats was observed, however, to cause localized reversal of this erosional trend on a 15 km long muddy coast extending west from Freshwater Bayou Canal. This deposition takes place only in late winter and early spring, synchronous with peak flood on the Atchafalaya River, which discharges into Atchafalaya Bay 100 km to the east of the study area. A four day investigation of nearshore dynamics, conducted over the broad range of wave and current conditions which accompanied passage of a winter cold front showed that waves incident on the muddy shoreface were well described by shallow water linear wave theory, but exhibited the progressive nearshore attenuation and absence of breaking described by Coleman and Wells (1981a) from Surinam. This attenuation gives rise to an onshore decrease in wave shear stress at the bed which is the reverse of that noted on clastic coasts dominated by surf zone processes. A significant vertical gradient in suspended sediment concentration was observed under low-frequency swell which was not developed when higher-frequency seas formed an important component of the wave energy spectrum. Concentrations 10 cm above the bed ranged up to the lower limit for fluid mud (Einstein and Krone 1962) when this gradient was present. The existence of a concentrated mud suspension near bottom at this time was also apparent in elevated estimates of bed roughness identified in logarithmic velocity profiles of the steady alongshore flow.

106

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 107

The formation of fluid mud layers appears to occur when sediments suspended through the full depth of the coastal boundary current are introduced into a less turbulent environment nearshore, but are prevented from depositing by high shear stresses close to the bed within a wave boundary layer (WBL). An important property of fine-grained sediment suspensions at fluid mud concentrations, which was demonstrated for the mudflat sediments in this study, is the development of yield strength, which is the ability to resist deformation up to some critical level of shear stress (Einstein 1941). Concentration of the sediment load near-bottom in a gel like fluid mud layer decouples the steady alongshore flow from turbulence generated within the 1 to 2 cm thick WBL. Adjustment of the steady current velocity profile to the presence of a fluid mud layer restricts further transport of this sediment alongshore. Cross-shore transport of sediment contained in this layer is proposed, however, under the influence of unsteady onshore-offshore flows which were observed to be associated with low- frequency, shore-amplified water level oscillations. These oscillations, apparently forced by pre-frontal winds, have characteristics of a standing wave with a shoreline antinode. Because of the slope of the foreshore, onshore and offshore flows associated with standing wave-like motion experience maximum acceleration and deceleration within 100 m of the shore. Intact transport of the fluid mud layer in the direction of an accelerating flow is proposed as a consequence of the wave-induced oscillatory shear stress exerted at its lower margin. Boundary-layer sorting of cohesive and non-cohesive sediments at this point, of the type described by Stow and Bowen (1980), implies the existence of a break in the vertical

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 108

sediment concentration gradient at the top of the WBL. Because of the relationship between sediment concentration and yield strength in fine grained sediment suspensions in the fluid mud range, yield strength should also exhibit a significant reduction at this point. Expanded to two dimensions, this point of weakness forms a plane along which the fluid mud layer can move in response to shear exerted by an accelerating flow at its upper margin. Rapid deposition of the fluid mud layer is proposed through a boundary layer dumping process when the yield strength of the fluid mud layer supported above the WBL reaches equivalence with WBL shear stress. The onshore decreasing shear stress gradient developed under attenuating waves gives rise to preferential deposition close to shore. Repeated sequences of boundary layer sorting, followed by "freezing" of the fluid mud layer, give rise to the vertically stacked 2 to 10 cm thick silt/shell + mud beds which characterize mudflat cores. Sedimentation rates associated with this process are estimated to be as high as 50 kg-m-2-s-l. Shoreface accretion by boundary layer dumping gives rise to a characteristic internal mudflat structure which exhibits little disturbance by burrowing marine organisms. Littoral mudflats deposited as part of a major coastal progradational episode have good preservation potential on a subsiding coast, and are well displayed in the regressive sequences described by Walker and Harms (1971,1976) from the Middle Devonian. The spread of clastic beaches into an area of former mudflat deposition in the eastern half of the chenier plain study area indicates that the flux of mud to this coast has decreased over the past 20 years. Trapping of fine grained sediment in the subaerial delta which has formed in Atchafalaya Bay

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 109

since 1973 is proposed as one possible cause for this reduction. Because of the close temporal relationship between mudflat formation and fine-grained sediment flux from Atchafalaya Bay to the shelf, no increase in littoral mud sedimentation on the chenier plain can be expected until 2030, when the subaerial delta is projected to expand onto the inner shelf (Wells et al. 1983). Regional coastal retreat along the chenier plain is, therefore, likely to continue for the next 50 years. Renewed large-scale advance of the marsh shoreline across mudflats into the Gulf of Mexico, the progradational process which has characterized this coast generally over the past 3000 years, is predicted when the new delta emerges from Atchafalaya Bay, and is exposed to increased wave action on the shelf. Some variation of the dynamics proposed for fluid mud formation and transport in the littoral setting investigated are likely to be manifested in the many shelf and estuarine environments where highly concentrated near bottom suspensions of fine-grained sediment are observed. Evidence acquired nearshore, in this study, of the boundary layer sorting originally proposed by Stow and Bowen (1980) to explain shelf margin turbidite sedimentation suggests that this mechanism may be important wherever multiple boundary layer length scales are involved. The dynamical inferences drawn by Stow and Bowen (1980) from sedimentological evidence, coupled with analysis of time-series measurements made on muddy coasts, begun by Wells and Coleman (1981a, b), and continued in this study, have led to proposal of a new mode for fine grained sediment transport in the marine environment. At the heart of this progress is a recognition that properties of fine-grained sediment suspensions, yield strength, for example, which do not emerge from

/ Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 110

consideration of the behavior of individual particles, play an important role in mud sedimentation. The importance of continued field research is clear, as some processes, like wave attenuation, which are critical to fluid mud formation, transport, and deposition, are poorly developed at laboratory scales. The results of this work suggest that the world's many muddy shorelines, which remain largely untouched by the most basic research efforts, are appropriate "laboratories" for the investigation of a broad spectrum of fine-grained sediment problems which also arise in other environments less accessible for study.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. BIBLIOGRAPHY

Adams, C.E., and Weatherly, G.L., 1981a, Suspended-sediment transport and benthic boundary-layer dynamics: Marine Geology, v. 42, p. 1-18.

______, 1981b, Some effects of suspended- sediment stratification on an oceanic bottom boundary layer: Jour. Geophys. Res., v. 86, p. 4161-4172.

Adams, C.E., Wells, J.T., and Coleman, J.M., 1982, Sediment transport on the central Louisiana continental shelf: implications for the developing Atchafalaya River delta: Contrib. Mar. Sci., v. 25, p. 133-148.

Alishahi, M.R., and Krone, R.B., 1964, Suspension of cohesive sediments by wind-generated waves: Hydraulic Engr. Laboratory, Univ. Cal., Berkeley, Tech. Rept. HEL-2-9, 24 p.

Allersma, E., 1971, Mud on the oceanic shelf off Guiana, in Symposium on Investigations and Resources of the Carribbean Sea and Adjacent Regions: UNESCO, Paris, p. 193-203.

Angelovic, J.W., 1976, Physical oceanography, in NOAA Final Report, Environmental Studies of the South Texas Continental Shelf: U.S. Dept, of Commerce, Gulf Fisheries Center, Galveston, TX. 111

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Bailey, A.M., 1934, Chenier au Tigre: La. Cons. Rev., v. 4(2), p. 9-11.

Beall, A.O., 1968, Sedimentary processes operative along the western Louisiana shoreline: Jour. Sed. Petrology, v. 38, p. 869-877.

Bendat, J.S., and Piersol, G.G., 1980, Measurement and Analysis of Random Data: New York, John Wiley and Sons, 302 p.

Bingham, E.C., 1922, Fluidity and Plasticity: New York, McGraw-Hill, 440 p.

Boussinesq, J., 1872, Theorie des ondes et de remous qui se propagent le long d'un canal rectangulaire horizontal, en communiquant au liquide contenu dans ce canal des vitesses sensiblement paralleles de la surface au fond: J. Math. Pures et Appliquees, v. 17, p. 55-108.

Brooks, R.A., 1970, Geochemical and mineralogical relationships in the sediments on the Louisiana continental shelf [unpubl. Ph.D. dissert.]: Louisiana State Univ., Baton Rouge, 97 p.

Byrne, J.V., LeRoy, D.O., and Riley, C.M., 1959, The Chenier Plain and its stratigraphy, southwestern Louisiana: Trans. Gulf Coast Assn. Geol. Soc., v. 9, p. 237-260.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 113

Carter, T.G., Liu, P.L-F., and Mei, C.C., 1973, Mass transport by waves and offshore sand bedforms: ASCE, Jour. Waterways Harbors Coastal Engr. Div., v. 99(WW2), p. 165-184.

Chuang, W.S., and Wiseman, W.J., Jr., 1983, Coastal sea-level response to frontal passage on the Louisiana-Texas shelf: Jour. Geophys. Res., v. 88, p. 2615-2620.

Cochrane, J.D., and Kelly, F.J., 1986, Low-frequency circulation on the Texas-Louisiana continental shelf: Jour. Geophys. Res., v. 91, p. 10645-10659.

Coleman, J.M., 1966, Recent coastal sedimentation: central Louisiana coast: Coastal Studies Series No. 17, Louisiana State Univ., Baton Rouge, 73 p.

Cooley, J.W., Lewis, P.A.W., and Welch, P.D., 1969, The fast Fourier Transform and its applications: IEEE Trans, on Education, v. 12, p. 27-34.

Cornish, V., 1898, On sea beaches and sand banks: Geog. Jour., v. 11, p. 528-559.

Cratsley, D.W., 1975, Recent deltaic sedimentation, Atchafalaya Bay, Louisiana [unpubl. M.S. thesis]: Louisiana State Univ., Baton Rouge, 142 p.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 114

Crout, R.L., 1983, Wind-driven, near-bottom currents over the west Louisiana inner continental shelf [unpubl. Ph.D. dissert.]: Louisiana State Univ., Baton Rouge, 117 p.

______, Wiseman, W.J., Jr., and Chuang, W.S., 1984, Variability of wind-driven currents, west Louisiana Inner Continental Shelf: Contrib. Mar. Sci., v. 27, p. 1-11.

Csanady, G.T., 1977, The coastal jet conceptual model in the dynamics of shallow seas in Goldberg, E.D., McCave, I.N., O'Brien, J.J., and J.H. Steele, eds., The Sea: New York, Wiley-Interscience, v. 6, p. 117-144.

Daddio, E., Wiseman, W.J., Jr., and Murray, S.P., 1978, Inertial currents over the inner shelf near 30° North: Jour. Phys. Oceanogr., v. 8, p. 728-733.

Das, M.M., 1970, Study of Bingham shear strength of a deep marine sediment by capillary viscometer in Einstein, H.A., and Wiegel, R.L., eds., A Literature Review on Erosion and Deposition of Sediment Near Structures in the Ocean: Hydraul. Engr. Lab., Univ. Calif., Berkeley, Final Rept. HEL 21-6, p. 2-1 - 2-13.

Dawson, T.H., 1978, Wave propagation over a deformable sea floor: Ocean Engr., v. 5, p. 227-234.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Delaney, P.J.V., 1966, Geology and geomorphology of the coastal plain of Rio Grande do Sul, Brazil and Northern Uruguay: Coastal Studies Series No. 15, Louisiana State Univ., Baton Rouge, 58 p.

Delft Hydraulics Laboratory, 1962, Demerara coastal investigation: Delft, The Netherlands, 240 p.

Dinnel, S.P., and Wiseman,W.J., Jr., 1986, Fresh water on the west Louisiana and Texas shelves: Cont. Shelf Res., v. 6, p. 765-784.

Dolan, R, Hayden, B., Homberger, G., Zieman, J., and Vincent, M., 1972, Classification of the Coastal Environments of the World, Part I, the Americas: Dept. Environmental Sciences, Univ. Virginia, Tech. Rept 1,163 p.

Doyle, E.H., 1973, Soil-wave tank studies of marine soil instability: Fifth Offshore Tech. Conf., Houston, Preprint 1901, p. II-753 - 11-766.

Drake, D.E., 1976, Suspended sediment transport and mud deposition on continental shelves, in Stanley, D.J., and Swift, D.J.P., eds., Marine Sediment Transport and Environmental Management, New York, Wiley Interscience, p. 127-158.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 116

Einstein, H.A., 1941, The viscosity of highly concentrated underflows and its influence on mixing: Trans. Am. Geophys. Union, Hydrology Section, p. 597-603.

______, and R.B. Krone, 1962, Experiments to determine modes of cohesive sediment transport in salt water: Jour. Geophys. Res. v. 67, p.1451-1461.

Eisma, D., and van der Marel,H.W., 1971, Marine muds along the Guyana coast and their origin from the Amazon basin: Contrib. Mineral. Petrology, v. 31, p. 321-334.

Faas, R.W., 1981, Rheological characteristics of Rappahannock estuary muds, southeastern Virginia, U.S.A., in Holocene Marine Sedimentation in the North Sea Basin, Int. Assn. Sediment. Spec. Publ., v. 5, p. 505-515.

Femandez-Partagas, J., and Mooers, C.N.K., 1975, A subsynoptic study of winter cold fronts in Florida: Monthly Weather Rev., v. 101, p. 742-744.

Fisk, H.N., 1948, Geology of lower Mermentau River basin and control structure sites: Definite Project Report, Mermentau River, Louisiana, App. 2, U.S. Army Corps of Engineers, Miss. River Comm., Vicksburg, MS., 41 p.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Folk, R.L., 1954, The distinction between grain size and mineral composition in sedimentary rock nomenclature: Jour. Geol., v. 62, p. 345-351.

, 1980, Petrology of Sedimentary Rocks, Austin, TX., Hemphill, 183 p.

Gade, H.G., 1958, Effects of a nonrigid, impermeable bottom on plane surface waves in shallow water: Jour. Mar. Res., v. 16, p. 61- 82.

, 1959, Notes on the effect of elasticity of bottom sediments to the energy dissipation of surface waves in shallow water: Arch. Math. Natur., Oslo, v. 55, p. 69-80.

Gibbs, R.J., 1970, Circulation in the Amazon River estuary and adjacent Atlantic Ocean: Jour. Marine Res., v. 28, p.l 13-123.

______, 1976, Amazon River sediment transport in the Atlantic Ocean: Geology, v. 4, p. 45-48.

Gould, H.R., and McFarlan, E, Jr., 1959, Geological history of the Chenier Plain, southwestern Louisiana: Trans. Gulf Coast Assn. Geol. Soc., v. 9, p. 261-270.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Grant, W.D., and Madsen, O.S., 1979, Combined wave and current interaction with a rough bottom: Jour. Geophys. Res. v. 84, p. 1797-1808.

Griffin, G.M., 1971, Interpretation of X-ray diffraction data, in Carver, R.E., ed., Procedures in Sedimentary Petrology, New York,Wiley Interscience, p. 541-569.

Ho, C., and Coleman, J.M., 1969, Consolidation and cementation of Recent sediments in the Atchafalaya basin: Bull. Geol. Soc. Am., v. 80, p 183-192.

Hsiao, S.V., and Shemdin, O.H., 1980, Interaction of ocean waves with a soft bottom: Jour. Phys. Oceanog., v. 10, p. 605-610.

Huntley, D.A., and Bowen, A.J., 1975, Field measurements of nearshore velocities: Proc. 14th Conf. Coastal Engr., Copenhagen, p. 538-549.

Hydraulics Research Station, 1979, Port of Brisbane Siltation Study, Fifth Report: Properties of Brisbane Mud: Hydraul. Res. Station, Wallingford, Rep. E X 860,18 p.

Jenkins, G.M., and Watts, D.G., 1968, Spectral Analysis and its Applications: San Francisco, Holden-Day, 525 p.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Johns, W.D., and Grim, R.E., 1958, Clay mineral composition of Recent sediments from the Mississippi River delta: Jour. Sed. Petrology, v. 28, p. 186-199.

Johnson, A.M., 1970, Physical Properties in Geology: San Francisco, Freeman, Cooper & Co., 577 p.

Kaczorowski, R.T., and Gemant, R.E., 1980, Stratigraphy and coastal processes of the Louisiana Chenier Plain and Louisiana Chenier Plain microfossil associations, in The Sedimentary Environments of the Louisiana Coastal Plain: Field Trip Guidebook for the 30th Annual Convention of the Gulf Coast Assn. of Geol. Soc., Lafayette, LA., p. 1- 72.

Kana, T.W., 1979, Surf zone measurements of suspended sediment: Proc. 16th Conf. Coastal Engr., Hamburg, p. 1725-1743.

Keulegan, G.H., and Krumbein, W.C., 1949, Stable configuration of bottom slope in a shallow sea and its bearing on geological processes: Trans. Am. Geophys. Union, v. 30, p. 855-861.

Kolb, C.R., and Van Lopik, J.R., 1958, Geology of Mississippi River deltaic plain, southeastern Louisiana: U.S. Army Corps of Engineers, Waterways Exper. Sta., Vicksburg, MS., Reports 3- 483, 3-484.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Komar, P.D., 1976, Beach Processes and Sedimentation: Englewood Cliffs, New Jersey, Prentice-Hall, 429 p.

Krone, R.B., 1962, Flume studies of the transport of sediment in estuarial shoaling processes: Hydraulic Engr. and Sanitaiy Engr. Res. Laboratory, Univ. Calif., Berkeley, 110 p.

______, 1963, A study of rheologic properties of estuarial sediments: Sanitaiy Engr. Res. Laboratory Report 63-8, Univ. Cal., Berkeley, 91 p.

• 1976, Engineering interest in the benthic boundary layer, in McCave, I.N., ed., The Benthic Boundary Layer, New York, Plenum Publ., p. 143-156.

Lanesky, D.E., Logan, B.W., Brown, R.G., and Hine, A.C., 1979, A new approach to portable vibracoring underwater and on land: Jour. Sed. Petrology, v. 49, p. 654-657.

Lhermitte, P., 1960, Mouvements des materiaux de fond sous faction de la houle: Proc. 7th Conf. Coastal Engr., The Hague, p. 211 - 261.

Longuet-Higgins, M.S., 1953, Mass transport in water waves: Phil. Trans. Roy. Soc. London, v. 245, p. 535-581.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 121

Maa, P.-Y, 1986, Erosion of soft muds by waves [unpubl. Ph.D. dissert.]: Univ. Fla., Gainesville, 275 p.

______, and Mehta, A. J., in press, Considerations on bed response in wave resuspension of muds, in Kuo, A.Y., ed., Estuarine Circulation, Clinton, New Jersey, Humana Press.

Macpherson, H., 1980, The attenuation of water waves over a nonrigid bed: Jour. Fluid Mechanics, v. 97, p. 721-742.

Mallard, W.W., and Dalrymple,R.A., 1977, Water waves propagating over a deformable bottom: Proc. 9th Offshore Tech. Conf., Houston, TX., v. 3, p. 141-146.

Manner, H.A., 1954, Tides and sea level in the Gulf of Mexico in U.S. Fish. Wildl. Serv., ed., Gulf of Mexico, its Origin, Waters and Marine Life, Fish. Bull. v. 89, p. 101-118.

Martins, L.R., Martins, I.R., Villwock, J.A., and Calliari, L.J., 1979, Occorrencia de lama na praia do cassino (RS): Separata Dos Anais 1978,22 p.

McCave, I.N., 1972, Transport and escape of fine-grained sediments from shelf areas, in Swift, D.J.P., Duane, D.B., and Pilkey, O.H., eds., Shelf Sediment Transport: Process and Pattern, Stroudsburg, PA, Dowden, Hutchinson, and Ross, p. 225-248.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. McCave, I.N., 1984, Erosion, transport and deposition of fine-grained marine sediments, in Stow, D.A.V., and Piper, D.J.W., eds., Fine-grained sediments: Deep-water Processes and Facies: Geological Soc. Spec. Publ. No. 15, Oxford, U.K., Blackwell Scientific, p. 35-69.

McCowan, J., 1894, On the highest wave of permanent type: Phil. Mag., Ser. 5, v. 38, p. 351-357.

Meade, R.H., Sachs, P.L., Manheim, F.T., Hathaway, J.C., and Spencer, D.W., 1975, Sources of suspended matter in waters of the Middle Atlantic Bight: Jour. Sed. Petrol., v. 45, p. 171-188.

Meadows, G.A., 1976, Time-dependent fluctuations in longshore currents: Proc. 15th Coastal Engr. Conf., Honolulu, p. 660-681.

Meie, R., Renshun, Z., and Juhai, Y., 1983, Sedimentation on the tidal mud flat of China: with special reference to Wanggang area, Jiangsu Province, in Qingming, J., and Milliman, J.D., eds., Proc. of Int. Symp. on Sedimentation on the Continental Shelf with Special Reference to the East China Sea, Beijing, China Ocean Press, p. 1-17.

Migniot, C., 1968, Etude des proprietes physiques de differents sediments tres fins et de leur comportement sous des actions hydronamiques: La Houille Blanche, v. 7, p. 591-620.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Mobbs, P.M., 1981, Clay mineralogy of deltaic sediments, Atchafalaya River, Louisiana [unpubl. M.S. thesis]: Louisiana State Univ., Baton Rouge, 70 p.

Moore, R.C., 1929, Environment of Pennsylvanian life in North America: Bull. Am. Assn. Petr. Geol., v. 13, p. 459-487.

Morgan, J.P., Van Lopik, J.R., and Nichols, L.G., 1953, Occurrence and development of mudflats along the western Louisiana coast: Coastal Studies Institute, Tech. Rept. 2, Louisiana State University, Baton Rouge, 34 p.

______, and Larimore, P.B., 1957, Changes in the Louisiana shoreline: Trans. Gulf Coast Assn. of Geol. Soc., v. 7, p. 303- 310.

______, and Morgan, D.J., 1983, Accelerating Retreat Rates Along Louisiana's Coast: Center for Wetland Resources, Louisiana State Univ., Baton Rouge, 41 p.

Murray, S.P., 1976, Currents and circulation in the coastal waters of Louisiana: Center for Wetland Resources, Louisiana State Univ., Baton Rouge, Sea Grant Publ. No. LSU-T-76-003, 33 p.

Nair, R.R., 1976, Unique mud banks, Kerala, southwest India: Bull. Am. Assoc. Petrol. Geologists, v. 60, p. 616-621.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. NEDECO, 1965, A Study on the Siltation of the Bangkok Port Channel: Netherlands Engr. Consultants, The Hague, v. II, 474 p.

, 1968, Surinam Transportation Study: Netherlands Engr. Consultants, The Hague, 293p.

NOAA, NOS, 1981, Tide Tables 1982, East Coast of North and South America: U.S. Dept, of Commerce, Natl. Ocean. Atm. Admin., Natl. Ocean Survey, Rockville, MD., 285 p.

Owen, M.W., 1970, A detailed study of the settling velocities of an estuary mud: Hydraulic Res. Station, Wallingford, Rep. INT 78, 25 p.

______, 1975, Erosion of Avonmouth mud: Hydraul. Res. Station, Wallingford, Rep. INT 150, 17 p.

Postma, H., 1961, Transport and accumulation of suspended matter in the Dutch Wadden Sea: Neth. J. Sea Res., v. 1, p. 148-190.

Powers, M.C., 1957, Adjustment of land derived clays to the marine environment: Jour. Sed. Petrology, v. 27, p. 355-372.

Price, W.A., 1955, Environment and formation of the Chenier Plain: Quatemaria, v. 2, p. 75-86.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Rhodes, E.G., 1982, Depositional model for a chenier plain, Gulf of Carpentaria, Australia: Sedimentology, v. 29, p. 201-221.

Rine, J.M., and Ginsburg, R.N., 1985, Depositional facies of a mud shoreface in Suriname, South America - a mud analogue to sandy, shallow-marine deposits: Jour. Sed. Petrology, v. 55, p. 633-652.

Roberts, H.H., Cratsley, D.W., and Whelan, T., 1976, Stability of Mississippi delta sediments as evaluated by analysis of structural features in sediment borings: Proc. 8th Ann. Offshore Tech. Conf., Houston, TX., p. 9-28.

______, Adams, R.D., and Cunningham, R.H.W., 1980, Evolution of a sand-dominant subaerial phase, Atchafalaya delta: Bull. Am. Assn. Petrol. Geol., v. 64, p. 264-279.

Russell, R.J., and Howe, H.V., 1935, Cheniers of southwestern Louisiana: Geog. Review, v. 25, p. 449-461.

Schlemon, R.J., 1975, Subaqueous delta formation - Atchafalaya Bay, Louisiana, in Broussard, M.L., ed., Deltas: Models for Exploration, Houston Geol. Soc., Houston, TX., p. 209-221.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Schuckman, B., and Yamamoto, T., 1982, Nonlinear mechanics of sea bed interactions, part 2: wave tank experiments on water wave damping by motion of clay beds: Rosenstiel School of Marine and Atmospheric Science, Univ. Miami, Miami, Tech. Rept. TR 82-1.

Schultz, L.G., 1964, Quantitative interpretation of mineralogical composition from X-ray and chemical data for the Pierre Shale: U.S. Geol. Survey, Washington, D.C., Prof. Paper 391-C.

Sheldon, J.E., and Parsons, T.R., 1967, A Practical Manual on the Use of the Coulter Counter in Marine Science: Coulter Counter Electronics, Inc., Toronto, 66 p.

Sikora, W.B., Sikora, J.P., and Prior, A. McK., 1981, The environmental effects of hydraulic dredging for clam shells in Lake Pontchartrain, Louisiana: Report of Center for Wetland Resources, Louisiana State Univ., Baton Rouge, to U.S. Aimy Corps of Engineers, New Orleans, Contract DACW29-79-C- 0099, 214 p.

Smith, J.D., and McLean, S.R., 1977a, Boundary layer adjustments to bottom topography and suspended sediment, in Nihoul, J.C.J., ed., Bottom Turbulence, New York, Elsevier, p. 123-151.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Smith, J.D, and McLean, S.R., 1977b, Spatially averaged flow over a wavy surface: Jour. Geophys. Res., v. 82, p. 1735-1746.

Stow, D.A.V., and Bowen, A.J., 1980, A physical model for the transport and sorting of fine-grained sediment by turbidity currents: Sedimentology, v. 27, p. 31-46.

______, and D.J.W. Piper, 1984, Deep-water fine-grained sediments; history, methodology, and terminology, in Stow, D.A.V., and Piper, D.J.W., eds., Fine-grained Sediments: Deep- water Processes and Facies, Oxford, U.K., Blackwell Scientific, p. 3-14.

Suhayda, J.N., 1972, Experimental study of the shoaling transformation of waves on a sloping bottom [unpubl. Ph.D. Dissert.]: Scripps Inst. Oceanogr., Univ. Calif., San Diego, 106 p.

, 1974, Standing waves on beaches: Jour. Geophys. Res., v. 79, p. 3065-3071.

Tennekes, H., and Lumley, J.L., 1972, A First Course in Turbulence: Cambridge, MA., MIT Press, 300 p.

Thiers, G.R., and Seed, H.B., 1968, Cyclic stress-strain characteristics of clays: Jour. Soil Mech. and Foundations Div., Am. Soc. Civil Engrs, v. 94, p. 555-569.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Thompson, E.F., 1980, Interpretation of wave energy spectra: Coastal Engr. Tech. Aid No. 80-5, U.S. Army Corps of Engineers, Coastal Engr. Res. Center, Fort Belvoir, VA., 21 p.

Thompson, R.W., 1968, Tidal flat sedimentation on the Colorado River delta, northwestern Gulf of California: Geol. Soc. Am. Mem. 107, p. 1-133.

Thompson, W.C., 1955, Sandless coastal terrain of the Atchafalaya Bay area, Louisiana, in Finding Ancient Shorelines, Spec. Publ. No. 3., Soc. Econ. Paleont. Mineral., Tulsa, OK., p. 52-76.

Todd, T.W., 1968, Dynamic diversion: influence of longshore current- tidal flow interaction on chenier and barrier island plains: Jour. Sed. Petrology, v. 38, p. 734-746.

Tubman, M.W., 1978, The interaction of surface waves and a linear viscoelastic sea floor [unpubl. M.S. thesis]: Louisiana State Univ., Baton Rouge, 39 p.

______, and Suhayda, J.N., 1976, Wave action and bottom movements in fine sediments: Proc. 15th Coastal Engr. Conf., Honolulu, v. 2, p. 1168-1183.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Turcotte, B.R., Liu, P.L-F., and F.H. Kulhawy, 1984, Laboratory evaluation of wave tank parameters for wave-sediment interaction: J.H. DeFrees Hydraul. Laboratory, Cornell Univ. Ithaca, N.Y., Rept. 84-1.

USACOE, 1974, Preliminary draft environmental impact statement, Atchafalaya Basin Floodway: U.S. Army Corps of Engineers, New Orleans, LA.

Van Heerden, I. LI., Wells, J.T., and Roberts, H.H., 1981, Evolution and morphology of sedimentary environments, Atchafalaya delta (south-central Louisiana): Trans. Gulf Coast Assn. Geol. Soc., v. 31, p. 399-408.

______, 1983, Sedimentation in eastern Atchafalaya Bay, Louisiana [Publ. Ph.D. dissertation]: Center for Wetland Resources, Louisiana State Univ., Baton Rouge, 117 p.

van Straaten, L.M.J.U., 1954, Composition and structure of Recent marine sediments in The Netherlands: Leidse Geol. Mededel, v. 19, p. 1-110.

______, and Kuenen, Ph. H., 1958, Tidal action as a cause of clay accumulation: Jour. Sed. Petrol., v. 28, p. 406- 413.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Waddell, E., 1973, Dynamics of swash and implication to beach response: Coastal Studies Inst., Louisiana State Univ., Baton Rouge, Tech. Rept. No. 139, 49 p.

Walker, R.G., and Harms, J.C., 1971, The "Catskill Delta": a prograding muddy shoreline in central Pennsylvania: Jour. Geol., v. 79, p. 381-399.

______, 1976, Shorelines of weak tidal activity: Upper Devonian Catskill Formation, central Pennsylvania, in Ginsburg, R.N., ed., Tidal Deposits, New York, Springer-Verlag, p. 103-108.

Wells, J.T., 1978, Shallow-water waves and fluid-mud dynamics, coast of Surinam, South America: Coastal Studies Inst., Louisiana State Univ., Baton Rouge, Tech. Rept. No. 257, 56 p.

______, Coleman, J.M., and Wiseman, W.J., Jr., 1979, Suspension and transportation of fluid mud by solitary-like waves: Proc. 16th Coastal Engr. Conf., Hamburg, p. 1932-1952.

______, and Coleman, J.M., 1981a, Physical processes and fine grained sediment dynamics, coast of Surinam, South America: Jour. Sed. Petrology, v. 51, p. 1053-1063.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Wells, J.T., and Coleman, J.M., 1981b, Periodic mudflat progradation, northeastern coast of South America: a hypothesis: Jour. Sed. Petrology, v. 51, p. 1069-1075.

______, and Kemp, G.P., 1981, Atchafalaya mud stream and recent mudflat progradation: Louisiana Chenier Plain: Trans. Gulf Coast Assn. of Geol. Soc., v. 31, p. 409-416.

______, and Kemp, G.P., 1985, Interaction of surface waves and cohesive sediments: field observations and geologic significance, in Mehta, A.J., ed., Cohesive Sediment Dynamics, Lecture Notes on Coastal and Estuarine Studies No. 14, New York, Springer-Verlag.

______, Chinburg, S J., and Coleman, J.M., 1983, Development of the Atchafalaya River deltas: generic analysis: Final report for U.S. Army Corps of Engineers, Waterways Experiment Station, Vicksburg, MS., Contract No. DACW 39-80-C-0082, Coastal Studies Inst., Louisiana State Univ., Baton Rouge, 98 p.

Whitehouse, U.G., Jeffrey, L.M., and Debbrecht, J.D., 1960, Differential settling tendencies of clay minerals in saline waters: Proc. 7th Nat. Conf. Clays Clay Mineral., New York, Pergamon Press, p. 1- 79.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Wilkinson, R.H., 1984, A method for evaluating statistical errors associated with logarithmic velocity profiles: Geo-Marine Letters, v. 3, p. 49-52.

Woodroffe, C.D., Curtis, R.J., and McLean, R.F., 1983, Development of a chenier plain, Firth of Thames, New Zealand: Marine Geology, v. 53, p.1-22.

Zenkovitch, Y.P., 1967, Processes of Coastal Development: New York, Wiley Interscience, 738 p.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. A PPEN D IX A

BEACH PROFILES The eight Chenier Plain beach stations shown in Figure 2 were reoccupied 9 times after establishment in the winter of 1980/1981. The last surveys were completed in May, 1983, thus the morphodynamic record extends over 2.5 years. The results of these surveys are compiled in Figures A-l through A-8. All profiles are tied to mean sea level (MSL) which lies 0.24 m above Gulf Coast Low Water, the standard datum referenced in local charts and tide tables. Vertical exaggeration is 50 times.

133

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. F IG U R E A -l

17 OEC'eO

LA. CHENIER PLAIN

PROFILE STA. 1

28 JUL'S 1 17 DEC'80-25 MAY'83

20 MAY 81

A VmffHPJQIT . tl -- *8 JUL Q1 • 14 DEC'81 V

6 OCT 61

DEC'81? 1.0m _ 6 MAR 82

\ 12 JUL'82 100m M8L' 200m 250m rTTT»U*l)‘u|l|iMMmiiff

18 DEC 82

20 OCT’82

SrwyMiyuiw,^.^ .... w1 ------~

16 DEC 82

26 MAY'83 — s

...... nil......

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. FIGURE A-2

29 MAY'S 1 LA. CHENIER PLAIN

PROFILE STA. 2

17 DEC'80-25 MAY’83 20 JUL‘81

I.IIIM^i^lMAWAWIAHIAWAIIIAIBiW

, 8 OCT'81 ■^VMlVMilUWUlWr 14 DEC'S 1 1.0 m

100m MSI* 50m 150m 200m 250m DEC'82 RW.lMl.i1 ■ iw ra w im m ^ — 14 DEC'81

12 JUL'62

8 DEC'82

^IH IIIim niiiiH H im in,.,.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 136

FIGURE A-3

.A . CHENIER PLAIN

PROFILE STA. 3

1.0m

MtL 2110m

|IHIIHHHlHHHininilH|,|

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 137

FIGURE A-4 LA. CHENIER PLAIN

PROFILE 6TA. 4 20 MAY'S 1 17 DEC'80 17 DEC'60-25 MAY'83

20 JUL'S 1

10 ocrei

1.0m

MSL SOm 100m 150m 200m 300m

13 JUL'82

10 DEC'82 13 JUL'82 15 MAR'82

14 DEC'81

26 MAY'83 10 DEC'82

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. FIGURE A-5

UL CMENIER PLAIN

PROFILE ETA. 6

14 JAN'01-26 MAY'83 t t MAY 81

JUL 6 1

14 DEC'S 1

16 JUL'S 1

1.0m / 10 OCT'S 1 _____

M8L 200m SOOm IS MAR 62

13 JUL’82

14 DEC 01

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 139

FIG U R E A-6 LA. CHENIER PLAIN

PROFILE STA. 6

1 4 JAN '81 “ 2 5 MAY '8 3

20 MAY'81

14 JAN '81

15 DEC'81

......

1 •On*

15 DEC'81

USL. 50nf 100m 150m 200m 250m 300m

i^OOCT'8^

10 DEC '82

10 DEC’82

lC s a s B ==,

/ Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 140

FIGURE A-7

LA. CHENIER PLAIN

PROFILE ST A. 7

13 FEB’e i “26MAY’83

.'81 15DEC81

13 OCT‘81 1.Qm ^ ....

24 JUN '82

16 MAR *82

MSL 50nv 100m 150m 200m 250m 300r 15 DEC *81

'"""••IIIM h i i IHIII

150EC *82

'{Immy^iyjMIMHIIHHly 20 SEP '82 H>HlllllHHl||nt„ )|

15 DEC'82 28 MAY'83

......

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 141

FIGURE A-8

LA. CHENIER PLAIN

PROFILE ST A. 8

13 FEB ’8 1 - 2 6 MAY'83

20 MAY '81

20 JUL '61

j? " " 11..... 130CT*81 \ 15DEC81 1.0m

23JUN '82

8 APfl *82

MSL 100m — i s o m — '« « .7 > 2 0 0 m 250m 3 0 0 m ^ -iLt. *""1111111 """•••liUlllllMMIliiimmnTTii

23 JUN'82

15DEC82

26 MAY *63 •’ M ill,

15 DEC '82

Utu

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. A PPEN D IX B

PROGRAM FOR GRANT-MADSEN BOUNDARY LAYER CLOSURE The FORTRAN program listed on the next five pages was used to perform the boundary layer closure referenced in the text. Documentation is provided in the comments on the first 3 pages. References of the form "GM EQ X" are to equations which appear in the journal article from which this program was developed (Grant and Madsen 1979).

142

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. C * PROGRAM (GMMODEL) * C * THIS PROGRAM PERFORMS GRANT-MADSEN(1979) CLOSURE FOR * C * WAVE-CURRENT BOUNDARY LAYER PROBLEM. ANGULAR DEVIATION * C * OF MEAN CURRENT FROM DIRECTION OF WAVE PROPAGATION PHICB)* C * IS ASSUMED FIXED AND MUST BE POSITIVE AND LESS THAN 90 * C * DEGREES. A NIKURADSE BOTTOM ROUGHNESS(XKB) MUST ALSO BE * C * SPECIFIED. INITIALLY, GUESS A VALUE FOR UA/UB(RATIO) AND* C * WAVE-CURRENT FRICTION FACTOR(FCW). OBTAIN A NEW VALUE * C * FOR FCW THROUGH ITERATION IN NESTED LOOP. RETURN TO MAIN* C * LOOP TO CALCULATE AN APPARENT BOTTOM ROUGHNESS(XKBC1) * C * WHICH IS COMPARED TO VALUE CALCULATED FROM FIELD DATA * C * (XKBC2) FOR MEAN CURRENT VELOCITY(UCR) MEASURED AT * C * REFERENCE ELEVATION(ZR) ABOVE BED. IF A SPECIFIED * C * TOLERANCE(TOL2) IS EXCEEDED, MAIN LOOP IS RE-EXECUTED * C * WITH AN ADJUSTED VALUE FOR UA/UB(RATIO). THIS PROCESS IS* C * REPEATED UNTIL CONVERGENCE CRITERION IS SATISFIED. * C ************************************************************ C INPUTS C FIELD VALUES FOR WAVE CHARACTERISTICS(LINEAR WAVE THEORY) C TITLE RUN NO. C FREQ WAVE FREQUENCY IN RADS C UB MAX NEAR-BOTTOM ORBITAL VELOCITY(CM/S) C AB NEAR-BOTTOM OSCILLATORY EXCURSION AMP(CM) C C C C C C C FIELD VALUES FOR CURRENT CHARACTERISTICS C PHICB DEVIATION OF MEAN STRESS VECTOR FROM DIRECTION OF C WAVE PROPAGATION IN RADS(LESS THAN 90 DEG) C PHIC DEVIATION OF MEAN CURRENT IN WBL FROM DIRECTION OF C WAVE PROPAGATION IN RADS. GM EQ 18. C UCR MEAN CURRENT VELOCITY AT REFERENCE ELEVATION IN CM/S C ZR REFERENCE ELEVATION-POSITIVE UP IN CM. C XKB NIKURADSE ROUGHNESS IN CM C C C C C C TOLERANCES FOR ITERATIONS C TOL1 PERMISSABLE ERROR FOR FCW ITERATION C TOL2 PERMISSABLE ERROR FOR KBC ITERATION C C C C C VARIABLES FOR WHICH STARTING ESTIMATES ARE NEEDED C RATIO RATIO OF UNKNOWN CURRENT REFERENCE VELOCITY TO MAX C NEAR-BOTTOM ORBITAL VELOCITY (UA/AB). USE (UCR/AB)

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. c TO START. c FCW WAVE-CURRENT FRICTION FACTOR. CHOOSE START VALUE FROM c GM FIG.4. c c c c VARIABLES FOR WHICH NO STARTING ESTIMATES ARE NEEDED c VARIABLES DEPENDENT ON RATIO AND PHIC ONLY c V2 GM EQ 16 c ALPHA GM EQ 20 c c c c c TERMS DEPENDENT ON V2, ALPHA, AND PHICB ONLY c Q1 LAST TERM LFT SIDE GM EQ 54 c Q2 RT SIDE GM EQ 54 c c c c c c TERMS CALCULATED IN FCW ITERATION c UCW CHARACTERISTIC TURBULENCE INTENSITY IN WBL(U*CW) c GM EQ 21 c XL PRANDTL MIXING LENGTH(L) c GM EQ 29 c ZETAO NO-SLIP WBL BOTTOM BOUNDARY PARAMETER c GM EQ 31 c X ARGUMENT OF KELVIN FUNCTION (2*SQRT(ZETAO)) c c c TERMS CALCULATED IN IMSL SUBROUTINE MMKELO c BER BER X (ORDER 0) c BEI BEI X (ORDER 0) c XKER KER X (ORDER 0) c XKEI KEI X (ORDER 0) c IER ERROR PARAMETER C C C C XK BRACED TERMS IN GM EQ 53 C GFCW NEW ESTIMATE OF FCW FROM TRANSPOSED GM EQ 54 C Q3 SECOND TERM LFT SIDE TRANSPOSED GM EQ 54 C Q4 DUMMY VARIABLE USED IN CALCULATING GFCW C XDIF1 (GFCW - FCW) C C C C C

EI LO I N CNEGNE FE 2 ITERATIONS CONVERGENCE NO 20 IF AFTER LOOP EXIT C oooo oono o o o o o o o o 0 F ''2 ’EM DPNET N AI, HC AD HC ONLY'/ AND PHICB PHIC, DEPENDENT ,’TERMS ONRATIO, ('l',2X T A M R FO 600 500 FORMATC'O',2X,'STARTING ESTIMATES'/2X,'RATIO'.F10.5/2X,'FCW', FORMATC'O',2X,'STARTING 500 400 F0RMAT('0',2X,'CONVERGENCE TOLERANCES'/2X, 'FCW ITERATION*,F11.5/ TOLERANCES'/2X, ITERATION*,F11.5/ 'FCW F0RMAT('0',2X,'CONVERGENCE 400 300 F0RMAT('0',2X,'CURRENT PARAMETERS’/2X, PARAMETERS’/2X, 'PHICB=',F10.5,15X, F0RMAT('0',2X,'CURRENT 300 200 FORMATC'O',2X,'WAVE PARAMETERS’/2X,'FREQ =',F10.5,2X,'UB =', =',F10.5,2X,'UB PARAMETERS’/2X,'FREQ FORMATC'O',2X,'WAVE 200 100 F0RMAT('1',2X,2A4,10X,’INITIAL VALUES') 100F0RMAT('1',2X,2A4,10X,’INITIAL 0 O J=l,21 , l = J 1 DO 30 20 F0RMAT(7F10.5) 20 10 F0RMAT(2A4,2X,3F10.5) _ _ 10F0RMAT(2A4,2X,3F10.5) 1 2X,'ITERATION = ’ ,I3 /2 X ,'V 2 = ' ,F10.5/2X,'ALPHA = ' ,F 1 0 .5 /2 X ,' Q2 =' =' Q2 ,' X /2 .5 0 1 ,F ' = ) .5 0 1 ,F10.5/2X,'ALPHA ,F ' ' = = Q3 2 ,' X ,'V 2 X , /2 5 . 0 ,I3 1 F ’ = 1 2X,'ITERATION 1 1F12.5) 2,XB ITERATION',F10.5) 2X,'XKBC 1 1 'PHIC ='.F10.5/2X,'UCR =',F10.5,2X,'ZR =',F10.5,2X,'XKB =', 1F10.5) =',F10.5,2X,'XKB =',F10.5,2X,'ZR 'PHIC='.F10.5/2X,'UCR 1 1 F10.5,2X,'AB =',F10.5) 1F10.5,2X,'AB 2 (2(.SR(QTAPA)))*COS(PHICB) .)) ) 2 * ))* IC H (P IN =(V2/(2.*SQRT(SQRT(ALPHA)) (S Q2 .* .-(3 (4 T R Q S I)* )/P I0 T ALPHA =1.+(RATI0**2.)+2.*RATIO*COS(PHIC) A R .* ((2 = V2 3 QTSR(LH) )/4.)-((V*2 *SQRT(ALPHA))) . 4 ( / ) V2**2. ( ( - ) . 4 / .) 3 * * ) SQRT(SQRT(ALPHA)) ( ( ( = Q3 XKAPPA =0.4 =0.4 XKAPPA PI =3.14159 PI F( E. 1 G T 210 TO GO 21) .EQ. (J IF DIMENSION TITLE(2) DIMENSION WRITE(6,500) RATIO,FCW RATIO,FCW WRITE(6,500) T0L1.T0L2 WRITE(6,400) PHICB,PHIC,UCR,ZR,XKB WRITE(6,300) WRITE(6,200) FREQ,UB,AB FREQ,UB,AB WRITE(6,200) TITLE WRITE(6,100) READ(5,20) RATIO,FCW RATIO,FCW READ(5,20) T0L1.T0L2 READ(5,20) PHICB,PHIC,UCR,ZR.XKB READ(5,20) READ(5,10) TITLE,FREQ,UB,AB TITLE,FREQ,UB,AB READ(5,10) WRITE(6,600) J,V2,ALPHA,Q2,Q3 J,V2,ALPHA,Q2,Q3 WRITE(6,600) EM EEDN NRSLSO C ITERATION FCW OF RESULTS ON DEPENDENT TERMS DF XKBC2-XKBC1 XDIF2 KC FEDBSDETMT FXB RMG Q 49 BASED EQ GM FROM FIELD OFXKBC ESTIMATE XKBC2 XKBC1 ESTIMATE OF GM TRANSPOSED ESTIMATE FROM A9 EQ XKBC XKBC1 UC CHARACTERISTIC SHEAR VELOCITY ABOVE ABOVE VELOCITY WBL 15 EQ GM SHEAR IN A9 EXPONENT CHARACTERISTIC EQ GM BETA UC .... - -' - - 40 DO 2 K=1,21 C EXIT LOOP IF NO CONVERGENCE AFTER 20 ITERATIONS. IF(K .EQ. 21) GO TO 220 UCW =SQRT(.5*FCW*ALPHA*UB**2.) UC =SQRT(0.5*FCW*V2*(UB**2.)) XL =(XKAPPA*UCW)/FREQ ZETAO =XKB/(30.*XL) X =2.*SQRT(ZETA0) WRITE(6,700) K ,UCW,UC,XL,ZETAO,X CALL MMKELO CX,BER,BEI,XKER,XKEI,IER) WRITE(6,800) BER,BEI,XKER,XKEI,IER XKEI =ABS(XKEI) XK =1./(X*(SQRT((XKER**2.)+(XKEI**2. )))) Q1 =.097*SQRT(XKB/AB)*(XK/FCW**(3./4.)) Q4 =XK* v Q.9J*SQRTCXKB./AB.)#( ( Ql+( 2 .*Q2 -) ) /Q3 )...... -...... GFCW =(Q4**(4./3.)) WRITEC6,900) FCW,GFCW,XK,Q1,Q2,Q3,Q4 XDIF1 =(GFCW-FCW) IF(ABS(XDIF1) .LE. T0L1) GO TO 50 FCW =GFCW 2 CONTINUE 700 FORMATCl','TERMS USED IN MMKELO. ITER = ',I3/2X,'UCW =',F10.5, 1 2X,'UC =' ,F10.5/2X,1 XL =',F10.5,2X,'ZETAO =',F10.5,2X,'X 1 F10.5) 800 FORMATC'O',2X,'MMKELO OUTPUT1/2X,'BERX =',F10.5,2X,'BEIX =', 1 F10.5/2X,'XKER =',F10.5,2X,'XKEI =',F10.5,2X,1IER =',F10.5) 900 FORMATC'O',2X,'FACTORS USED TO FIND GFCW'/2X,'FCW =',F10.5,2X, 1 'GFCW =' ,F10.5,2X,'XK = ',F10.5/2X,'Q1 =',F10.5,2X,'Q2 1 F10.5,2X,'Q3 =',F10.5,2X,'Q4 = ’,F10.5) 50 FCW =GFCW UC =SQRT(0.5*FCW*V2*CUB**2.)) UCW =SQRTC.5*FCW*ALPHA*UB**2. ) BETA =1.-(UC/UCW) WRITEC6,910) K,UC,UCW,BETA,FCW 910 FORMATC'1',2X,'CONVERGENCE REACHED ON FCW AFTER',13, 1 ' ITERATIONS'/2X,'TERMS NEEDED TO FIND XKBC’/2X,'UC =',F10.5, 1 'UCW =' ,F10.5,2X,'BETA =',F10.5,2X,'FCW =',F10.5) XKBC1 =XKB*CC24.*(UCW/UB)*(AB/XKB))**BETA) XKBC2 =30.*ZR/EXP(UCR*XKAPPA/UC) XDIF2 =XKBC2-XKBC1 WRITEC6,920) J,RATIO,XKBC1,XKBC2,XDIF2 920 FORMATC'O',2X,'RESULTS OF ITERATION',13,' ON KBC'/2X,'RATIO =', 1 F10.5,2X,'XKBC1 =',F10.5,2X,'XKBC2 =',F10.5,2X,'XDIF2 1 F10.5) IFCABSCXDIF2) .LE. T0L2) GO TO 60 IFCABSCXDIF2) .LE. 5.) GO TO 55 RATIO =RATI0*SQRTCXKBC1/XKBC2) GO TO 1 55 RATIO =RATI0*SQRT(SQRTCXKBC1/XKBC2)) 1 CONTINUE 210 WRITEC6,310) GO TO 999 220 WRITEC6,320) GO TO 999

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 310 FORMATC' 1 ' , ' NO CONVERGENCE ON RATIO AFTER 20 ITERATIONS. ABORT') 320 FORMATC’1 NO CONVERGENCE ON FCW AFTER 20 ITERATIONS. ABORT') 60 XKBC =XKBC2 WRITE(6,930) J,TITLE 930 FORMATC'l',2X,'CONVERGENCE REACHED ON RATIO AFTER',13, 1 1 ITERATIONS'/2X.2A4,10X,' INITIAL VALUES') WRITEC6.200) FREQ, UB, AB WRITEC6,300) PHICB,PHIC,UCR,ZR,XKB WRITEC6,400) T0L1,T0L2 WRITEC6,940) WRITE C 6,9 5 0) RATI0 , FCW,XKBC, UCW,UC, XL, ZETAO, V2, ALPHA 940 FORMATC'O1,22X,'FINAL RESULTS') ...... 950 FORMATC'O',' RATIO = ',F10.5,2X,'FCW = ',F10.5,2X,'XKBC =’,F10.5/ 1 2X,'UCW = ',F10.5,2X,'UC =’,F10.5/2X,'XL =',F10.5, 1 2X,'ZETAO = '.F10.5/2X,'V2 =' ,F10.5,2X. '^fHA =',FI0.5) ...... : • w. r_ : g gj g 'CONTI NT)E r - -; ...... -...r - - - - -..... STOP END

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. VITA

George Paul Kemp was bom on May 2,1954, in Port Jefferson, New York. He graduated cum laude from Saint Paul's School in Garden City, New York, in May, 1971. He enrolled in the Fisheries Science program at Cornell University in Ithaca, New York, and received the B.S. degree in May 1975. Mr. Kemp began graduate studies later that year in the Department of Marine Sciences at Louisiana State University. Following award of the M.S. degree in May, 1978, he pursued his interest in estuarine geochemistry as a Research Associate in the Department of Engineering Research, also at L.S.U., through the fall of 1979. He then re-entered the Department of Marine Sciences to begin a graduate program in coastal sedimentology through the Coastal Studies Institute. Mr. Kemp was awarded the Joseph Lipsey, Sr., Scholarship in 1981, and spent 1984 in Washington, D.C., where he served on the legislative staff of Senator Edward M. Kennedy as a Sea Grant Congressional Fellow. Since returning to Louisiana, Mr. Kemp has been employed by Groundwater Technology, Inc., in Mandeville. Mr. Kemp is currently a candidate for the Ph.D. degree in the Department of Marine Sciences.

148

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. DOCTORAL EXAMINATION AND DISSERTATION REPORT

Candidate: G. Paul Kemp

Major Field: Marine S cien ces

Title of Dissertation: MUD DEPOSITION AT THE SHOREFAGE: WAVE AND SEDIMENT DYNAMICS ON THE CHENIER PLAIN OF LOUISIANA

Approved:

Major Professor and .Chairman

Dean of the Graduatee School

EXAMINING COMMITTEE: n "7—7*

Date of Examination:

November 26, 1986

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.