• TABLE OF CONTENTS
HIDE
 Front Cover
 Half Title
 Title Page
 Acknowledgement
 Table of Contents
 List of Tables
 List of Figures
 Abstract
 Introduction
 Method of analysis
 Cedar River estuary
 Assessment of sediment trapping...
 Conclusion
 Appendix
 Reference
 Biographical sketch














Title: Efectiveness of fine sediment trapping in an estuary
CITATION THUMBNAILS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00091383/00001
 Material Information
Title: Efectiveness of fine sediment trapping in an estuary
Series Title: Efectiveness of fine sediment trapping in an estuary
Physical Description: Book
Language: English
Creator: Paramygin, Vladimir A.
Publisher: Coastal & Oceanographic Engineering Dept. of Civil & Coastal Engineering, University of Florida
Place of Publication: Gainesville, Fla.
 Record Information
Bibliographic ID: UF00091383
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.

Table of Contents
    Front Cover
        Front Cover
    Half Title
        Half Title
    Title Page
        Page i
    Acknowledgement
        Page ii
    Table of Contents
        Page iii
        Page iv
    List of Tables
        Page v
    List of Figures
        Page vi
        Page vii
    Abstract
        Page viii
        Page ix
    Introduction
        Page 1
        Page 2
        Page 3
        Page 4
    Method of analysis
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
    Cedar River estuary
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
    Assessment of sediment trapping efficiency
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
        Page 47
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
    Conclusion
        Page 61
        Page 62
    Appendix
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
        Page 68
    Reference
        Page 69
        Page 70
    Biographical sketch
        Page 71
Full Text



UFL/COEL-2002/002


EFFECTIVENESS OF FINE SEDIMENT TRAPPING IN AN
ESTUARY


by




Vladimir A. Paramygin


THESIS





2002


LJNWERSITY OF
F L OR ID


Coastal & Oceanographic Engineering Program
Department of Civil & Coastal Engineering
433 Weil Hall -P.O. Box 116590 Gainesville, Florida 32611-6590


pptpt~faf$i~re~i~f
-L.








UFL/COEL-2002/002


EFFECTIVENESS OF FINE SEDIMENT TRAPPING IN AN
ESTUARY







by




Vladimir A. Paramygin


THESIS


2002

















EFFECTIVENESS OF FINE SEDIMENT TRAPPING IN AN ESTUARY


By

VLADIMIR A. PARAMYGIN
















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

UNIVERSITY OF FLORIDA


2002

















ACKNOWLEDGMENT


I wish to express my gratitude to Dr. Ashish J. Mehta, chairman of my graduate

committee, for his guidance in my education and research, and for his patience, support

and understanding. I also wish to thank the members of my graduate committee, Dr.

Robert J. Thieke and Dr. Daniel M. Hanes, for their great teaching effort and assistance.

Special thanks go to Dr. Earl J. Hayter for his assistance with my research effort, and

Sidney L. Schofield for help in the laboratory and field work.

I wish to thank my parents and sister for their love and trust, my many friends for

their support, encouragement and help, and the faculty and staff of the Coastal and

Oceanographic Engineering Program of the Department of Civil and Coastal Engineering

for their help during my stay. Data used in this study are in part derived from field work

conducted for the St. Johns River Water Management District (SJRWMD) of Palatka,

Florida. Thanks are due to Dr. Chandy John of the SJRWMD for his technical assistance

throughout this study.

Finally, very special thanks go to my friends Oleg Mouraenko, Leonid Parshukov,

Angela Chulkova, Yuri Melentyev and Ulyana Merkulova for their help and support.

















TABLE OF CONTENTS

page
ACKNOWLEDGMENT............................................................................................... ii

L IS T O F T A B L E S .................................................................................... ..................... v

L IST O F F IG U R E S ........................................................................................................... vi

EFFECTIVENESS OF FINE SEDIMENT TRAPPING IN AN ESTUARY.................. viii

1 IN T R O D U C T IO N .............................................................................. ...................

1.1 P rob lem Statem ent .......................................................................... ................... 1
1.2 C ed ar R iv er ............................................... ......................................................... 2
1.3 O objective and T asks ....................................................................... .................... 3
1.4 O outline of C chapters ................................................................................................. 3

2 M ETH O D O F A N A LY SIS .................................................................................. ........ 5

2 .1 In tro du action ............................................ ........................... ............................... 5
2 .2 F low F field ........................................................ ....................... ..... .... 5
2.3 Sedim ent Transport ................................................................. .................... 10
2.4 Settling V elocity C alculation ........................................................... ..... ........... 11
2.4.1 B background ............ .. .. ..................................... .................... 11
2.4.2 Particle Density and Fractal Representation ............................ ......... 15
2.4.3 Settling Velocity.............................................. 15
2.4.4 Floc Growth and Breakup Functions .................... ......... .......... .. 17

3 CED A R RIV ER ESTU A RY ........................................................................... 20

3.1 Description of the Estuary................................................................................. 20
3.2 Tide, Waves, Current, Salinity and Wind Data..................... ......................... 22
3.2.1 T ide data ........................................................... .......................................... 22
3.2.2 Current data ........ ....................................23
3.2.3 W ind data ... ...................................................... ......................................... 24
3 .2 .4 W av e data ..................................................................................................... 2 6
3.2.5 Salinity data........................................ .............................. ........................... 27
3.3 D ischarge D ata ..................................................................................................... 28
3.4 Sedim ent C concentration ........................................................................................ 29
3.5 Bed Sediment Distributions ......................... ................. 30



iii











4 ASSESSMENT OF SEDIMENT TRAPPING EFFICIENCY ............................... 37

4.1 Flow Model Setup, Calibration and Validation .............................................. ...37
4.1.1 Cedar/Ortega/St. Johns Rivers Model Setup.......................... ........ .. 37
4.1.2 Cedar River Model Setup.............................................. .....................43
4.2 Sediment Transport Model Setup and Calibration...................... .......... .. 50
4.2.1 Sediment Transport Model Setup.............................. ..................... 50
4 .2 .2 B ed E erosion .................................................................. ........................... 52
4.2.2 Settling Velocity and Deposition ................................... ....................... 53
4.3 Trapping Efficiency Analysis.................................................... 57
4.3.1 Treatm ent Plan .......................................... .......................... 57
4.3.2 Sediment Trap Setup .................................................. 57
4.3.3 Effect of Trap Efficiency on Settling Flux Downstream ............................. 58

5 C O N C LU SIO N S ........................................................... .................................. 61

5 .1 S u m m ary .............................................................................. ................. .......... 6 1
5.2 C onclu sion s ............................................................. ............. . ...... ............... 6 1
5.3 Recommendations for Further Work.............. ............................................. 62

APPENDICES

A WATER DISCHARGE ESTIMATION BASED ON ADCP................................... 63

B SETTLING VELOCITY AND FLOC SIZE CALCULATIONS ............................65

B Introduction .......................................................................... .................... 65
B .2 Settling V elocity Calculations ............................................ ...................... 65
B .3 Particle Size Calculations .................................................... ... .................. 67

R EFE R EN C E S ........................... ...................................................................................... 69

BIOGRAPH ICA L SKETCH ........................................................... ..................... 71


















iv


















LIST OF TABLES


Table Page

3-1: Wind speed/direction distribution. Dominant speed/directions are highlighted... 25

3-2: Cedar River cross-section discharges, May 17, 2001 .......................................29

3-3: Sediment concentrations from water samples, May 17, 2001........................... 30

3-4: Statistical values associated with bed sediment distribution............................... 32

4-1: TSS removal efficiencies of treatment systems in Florida (after Harper, 1997)... 57

4-2: Comparison of sites with different removal efficiencies with a no-trapping
scene ario ........................................................................... ..... ................ ......... 5 9

4-3: Summary of the effect of treatment on sediment load in the confluence area...... 60

B-l: Data from settling column tests with Ems-Dollard mud..................................... 68


















LIST OF FIGURES


Figure Page

2-1: Effect of sediment concentration and fluid shear stress on the median floc
diam eter (D yer, 1989). .......................................................... ..................... 13

2-2: Relationship between settling velocity and floc diameter in still water, based on
equation (2.29). .................................................................... ..................... 17

3-1: Cedar/Ortega Rivers estuary, (within the light rectangular area), aerial photo,
M ay 19 9 8 .......................................................... ..................... ............ ..... ..... 2 0

3-2: Cedar/Ortega Rivers data collection and sediment treatment (Wet Detention
System ) sites 1 and 2 .................................................. ............................... 2 1

3-3: Tidal ranges at stations TG -TG3. Cumulative frequency distribution based on
record obtained during 09/29/00-10/18/01............................ .................... ... 23

3-4: Current speed at station WGC (mouth of the Ortega in St. Johns River).
Cumulative frequency distribution based on record obtained during 02/05/01-
03/08/01 .................... ... ...... .............. ...................................... ................ .. 24

3-5: Wave spectrum based on 10/02/01 -25/04/01 record at WGC............................. 26

3-6: Significant wave height based on spectral analysis. .......................................... 27

3-7: Cumulative distribution of the significant wave height at the mouth of the Ortega
River in the St. Johns River.................... ........................................................... .. 27

3-8: Cumulative distribution of salinity at stations TG -TG3; 10/27/00-11/26/00...... 28

3-9: Cedar River cross-section discharge and Cedar/Ortega tide data, May 17, 2001. 29

3-10: Moisture content distribution (%). Based on 1998 sampling................................ 33

3-11: Organic content distribution (%). Based on 1998 sampling. ............................ 34

3-12: Solids content distribution (%). Based on 1998 sampling................................ 35

3-13: Thickness of soft deposit in the study area based on core thicknesses in 1998
sam pling. ............................................................................... 36











4-1: Areas covered by the two (coarse grid and fine grid) models........................... 38

4-2: Cedar/Ortega/St. Johns River grid with open boundary locations...................... 39

4-3: Cedar/Ortega/St. Johns River bathymetry. Depths are in meters........................ 40

4-4: Cedar/Ortega/St. Johns River open boundary condition (BC7) during 2001
showing water surface elevation time-series and cumulative distribution.......... 41

4-5: Cedar/Ortega/St. Johns River tributaries discharge, cumulative distribution....... 42

4-6: Cedar River m odel grid. ........................................................ .................... 44

4-7: Cedar River bathymetry. Depths are in meters. ............................ ............. 45

4-8: Measured and simulated water level variations at the downstream boundary of the
C edar R iver. ........................................ ........................................................ 47

4-9: Water surface elevation at three control points in Cedar River ........................ 48

4-10: Measured and simulated discharges through the Cedar River cross-section........ 48

4-11: Measured and simulated discharges through the Ortega River cross-section (north
cross-section of the confluence)......................... ................ ....................... 49

4-12: Measured and simulated discharges through the Ortega River cross-section (south
cross-section of the confluence) .................................................. ....................... 49

4-13: Depth averaged TSS concentrations at the Cedar River cross-section near the
Cedar/Ortega confluence simulated by the coarse grid model (May 17, 2001).... 51

4-14: Depth averaged TSS concentrations from the water sample data, collected at the
Cedar River cross-section near the Cedar/Ortega confluence (May 17, 2001)..... 51

4-15: Bed erosion rate function obtained from laboratory experiments on mud from the
Cedar/O rtega Rivers ................................... ....................... ........................ 53

4-16: Calculated floc size as a function of shear stress and concentration................... 55

4-17: Settling velocity curve based on laboratory tests in a settling column using
sedim ent from the Cedar River and vicinity. ..................................................... 56

B-1: Settling velocity calculation test results, and comparison with data of Wolanski et
al. (1992) using sediment from Townsville Harbor, Australia ........................... 66

B-2: Floc growth with time measured and predicted for River Ems-Dollard mud....... 68

















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

EFFECTIVENESS OF FINE SEDIMENT TRAPPING IN AN ESTUARY

Vladimir A. Paramygin
May 2002

Chairman: Ashish J. Mehta
Major Department: Civil and Coastal Engineering

A study of the effectiveness of fine sediment trapping in the Cedar River estuary

in north Florida is presented. A combined hydrodynamic and sediment transport model

has been used to simulate discharge, water level variation and suspended sediment flux at

various sections in the river. A settling velocity sub-model accounting for the effects of

floc growth and breakup due to turbulence in the water column has been incorporated in

the model.

The simulated discharges and surface elevations at the downstream open

boundary of Cedar River at its confluence with the Ortega River are found to agree

reasonably well with measurements. It is shown that trapping sediment upstream in Cedar

River would have only a minor effect on sediment deposition in the downstream region,

where contaminated fine sediment has accumulated. This is so because the majority of

the sediment is apparently derived from creeks entering the river within its middle reach.

On the other hand, sediment entrapment closer to the confluence of the Cedar and Ortega

rivers appears to be able to measurably reduce sediment transport to the confluence and,











therefore, can be expected to lower the flux of sediment-bound contaminants out of Cedar

River.
















CHAPTER 1
INTRODUCTION

1.1 Problem Statement

Sediment shoaling in estuarine environments can create significant problems such

as degradation of water quality and concentration of organic matter and contaminants.

Accumulated organic-rich sediment can increase contaminant loads in these waters,

because contaminants such as polychlorinated biphenyls (PCBs) and polyaromatic

hydrocarbons (PAHs) are preferentially bound to organic (National Research Council,

2001).

A commonly implemented solution to reduce sedimentation is the construction of

a sediment trap. Such traps can be of different types; however, all of them rely the same

basic mechanism decreasing the speed of the flow, thus allowing the larger portion of

the suspended sediment load to settle out in the trap. Traps can be on-line or off-line. One

example of an on-line trap is a dredged trench along the submerged bottom, which

reduces the flow velocity and causes the material to settle there. An off-line trap is made

by artificially diverting part of the flow into a natural/artificial pond, which would reduce

flow velocity and increase deposition. Selecting the Cedar River in northern Florida as a

case study, Stoddard (2001) examined the efficiency of a trap trenched at the bottom of

the river. In the present study, the efficiency of a trap created by ponding along the side

of the same river is explored.

For the present purposes, trap efficiency will be determined by the sediment

removal ratio, i.e., the percentage by which the effluent sediment load is reduced with








2


respect to the influent load (Ganju, 2001). By creating efficient traps, much of the

detrimental effects of excess sediment and unwanted contaminants entering the system

can be curtailed.


1.2 Cedar River

Cedar River estuary occurs in northeast Florida. Trapping contaminants in this

river has become essential due to elevated concentrations of sediment-bound PCBs in

water resulting from leaching of sediment and runoff from the site of a chemical

company since January 1984. The site is located approximately 0.5 km east of the Cedar

River near its headwaters, adjacent to municipal storm drains and drainage ditches.

There, fire destroyed several tanks storing high concentrations (4,425 ppm) of PCB-laden

oils and other materials. It is believed that a combination of damage to the storage tanks

and the fire-fighting effort caused PCBs to enter the Cedar River basin. The surrounding

groundwater and soils were sampled extensively in 1989, and the concentrations were

still significantly above the regulated amount of 50 ppm.

Estuaries characteristically trap significant quantities of particulate matter through

a wide variety of physical and biochemical processes. Fine-grained sediments play an

important role in these processes. Due to relatively strong currents, fine sediments, which

are mixtures of clay- and silt-sized material, are usually very mobile. In the Cedar River

they are admixed with organic matter derived from local terrestrial and aquatic sources.

Fine sediment transport is mainly defined by the hydrodynamic action, which

advects the suspended matter and provides the bed erosion force. Also, turbulence plays a

major role in the flocculation of fine, cohesive sediments. Flocs are formed by the joining

of individual particles and can greatly affect the settling velocity of particulate matter.











The St. Johns River Water Management District of Palatka, Florida is considering

the possibility of establishing off-line sediment traps upstream along the Cedar River, and

would like to have an estimate of the influence of this entrapment on sediment influx at

the downstream end of Cedar River, at its confluence with the Ortega River, where heavy

accumulation of PCB-laden, organic-rich sediment has occurred.


1.3 Objective and Tasks

The objective of this study was to determine the effect of traps with different

efficiencies at selected locations upstream in Cedar River to sediment flux at the

confluence of Cedar and Ortega Rivers downstream. Several tasks were undertaken to

achieve this objective including:

1) Use of data to characterize the nature of flow. Data included tidal elevations,
current velocities, wind speed and direction, salinity, streamflows at the major
tributaries of the Cedar River and the Cedar River itself, and suspended
sediment concentrations.

2) Modeling the flow field using a numerical code to determine water velocities,
water surface elevations and bed shear stresses.

3) Modeling fine sediment settling velocity as a function of the local flow
conditions.

4) Use of a sediment transport model (with implemented settling velocity model)
to determine suspended sediment concentrations within the modeled domain.

5) Using the calibrated flow and sediment models, modeling flow and sediment
transport in the estuary with the sediment traps (with three assumed
efficiencies 30%, 60% and 90%) and without the traps.

6) Comparison of the results of above modeling in terms of sediment transport at
the downstream end of Cedar River to assess the effects of above traps.

1.4 Outline of Chapters

Chapter 2 describes the flow and sediment transport model used to evaluate trap

efficiency. Chapter 3 contains the field data collected for this study. Chapter 4 describes








4


the calibration and validation of the model using measured data. Finally, Chapter 5

contains the summary and conclusions, followed by bibliography.
















CHAPTER 2
METHOD OF ANALYSIS

2.1 Introduction

This chapter provides a description of the hydrodynamic and sediment models

that were used to model flow and the sediment transport in the Cedar River. The chapter

gives basic equations, numerical method used to solve these equations, and the

capabilities and limitations of the models for problem analysis.

The Environmental Fluid Dynamics Code (EFDC) used herein implements a

numerical algorithm for estuarine flows (Hamrick, 1996). It contains a three-dimensional,

hydrostatic flow model, as well as a compatible sediment model.


2.2 Flow Field

The coordinate system of the model is curvilinear and orthogonal in the horizontal (x,y)

plane. In the vertical, z direction, which is aligned with the gravity vector, it is stretched

to follow the bottom topography and free surface displacement (o-grid). A level 2.5

turbulence closure scheme (Mellor and Yamada, 1982) in the hydrodynamic model

relates the turbulent viscosity and diffusivity to the turbulence intensity and a turbulence

length scale. An equation of state relates density to pressure, salinity, temperature and

suspended sediment concentration (Hamrick, 1992).










The momentum equations in the model are

D,(m,m,,Hu)+ ,. (m,Huu)+ a ,(myHvu)+ a. {mm.iwu)- f.mm, Hv

=-m,Hm (p +p ,,,,, + )+ m,,(xz;. +z, H)p+ l mm,, u) (2.1)
H

+. HA x x HAa uj mm,cD,(u +v2 )12
mx ) \ m, y

(mm ,,Hv)+ 9,. m,.Huv)+ 0 (mrHw)+ (mxm,,wv)- fmm Hu

=-m xH,.I(p+p,,,,, +p )+ m, (.,z', + z.,,H).p+. nim m,Am .v (2.2)

+a., HAH.d v +a."| HAHaYV _nm) cPD P(u2 +v2)12 V


m,m,,f, = mxm,,f Ul,mx + vaxm, (2.3)

)x:>,)= A,.H -'a(u,v) (2.4)

where u and v are the horizontal velocity components in the x, y coordinate directions,

respectively, and w is the vertical velocity; m, and nm, are the scale factors of the

horizontal coordinates; z and z, are the vertical coordinates of the free surface and

bottom bed, respectively; H is a total water column depth; 0 = gz, is a free surface

potential; f, is the effective Coriolis acceleration and incorporates curvature acceleration

terms with the Coriolis parameter, f as in equation (2.3); A, and A,, are the horizontal

and vertical turbulent viscosities, respectively, where A,. relates the shear stresses to the

vertical shear of the horizontal velocity components; the last terms in equations

(2.1) and (2.2) represent vegetation resistance, where c is a resistance coefficient and

D is the dimensionless projected vegetation area normal to the flow per unit horizontal










area; and p.,,, is the kinematic atmospheric pressure referenced to water density. The

excess hydrostatic pressure in the water column is

a p = -gHb = -gH(p o )po' (2.5)

where

p = p(p,S,T) (2.6)

and p and p, are the actual and reference water densities, respectively, and b is

buoyancy.

The three-dimensional continuity equation in the model is

a, (rnm, H)+ a,(m,,Hu)+ ,,(m_,Hv)+ a (mwm,,w)= QH (2.7)

and the corresponding vertically-integrated form of the continuity equation is

a,(mx mH)+ a(m,,/-u)+a y (mx v)= QU (2.8)

where Q, represents volume sources and sinks including rainfall, evaporation,

infiltration and lateral inflows and outflows having negligible momentum fluxes.

Transport equations for temperature and salinity are

a,(mHS)+ (mHuS)+ ,(mHvS)+ (mwS)= a:(mH 'A,zS)+Q, (2.9)

a,(mHT)+ a,(mHuT)+3a,(mHvT)+ ,(mwT)= (mH 'A,,aT)+ Q, (2.10)

where Q, and Q, are source and sink terms, respectively, which include sub-grid scale

horizontal diffusion and thermal sources and sinks, and A, is the vertical turbulent

diffusivity.

Two transport equations determine the turbulent intensity and turbulent length scale as

follows:









M, m mY.Hq2)+3, (mHuq2)+ ,, (mxHvq2)+ (mxm,wq2)
A, 2 Hq3
=3, mxmyI- q 2-2mxm, B (2.11)
H B1l
+2mm -((dyu)2 +(. v)2)+qc7pD(lu 2 +v gK,.+:b)+Qq


3, (mm,.mHq2 1)+ 3 (m Huq2l)+ a, (mxHvq2l)+ 3 (m.m,,wq21)

= nmxm AdL (q2Hq1)- mm r+E2( H 2 +E3 kH( 1) (2.12)
H B, kHz kH( -(z)

+mxmyE, l ((3u)2 +(a3v)2)+gK,3b+7cpD,(u2+v2) +Q,

where (E,,E,,E3)= (1.8,1.33,0.25) are empirical constants, Qq and Q, represent

additional source-sink terms, and the third term in the last line of both equations
represents net turbulent energy production by vegetation drag with a production

efficiency factor of rq,.

Equation (2.4), which specifies the kinematic shear stress at the bed and free
surfaces, provides the vertical boundary conditions for the solution of the momentum
equations. At the free surface, the shear stress boundary conditions are given by the water
surface wind stress

(r,..,r, )= (r,,,)= c JU,, + V, (U,,,V,,.) (2.13)

where U,,, and V,, are the x and y components of the wind velocity, respectively, 10 m

above the water surface. The wind stress coefficient for the wind velocity components is

c = 0.001 (0.8+0.065 U2 + V,) (2.14)
AOW












where P'i and P,, are the air and water densities. At the bed, the shear stress components

are considered to be related to the near-bed or bottom layer velocity components as

follows:

( --. ,,; )= rh., r,,)= cC, U ` +v (u,, v,) (2.15)


where ul and v' are the bottom layer velocity components and the bottom stress

coefficient is




ch,= (2.16)
InA


which assumes that the near-bottom velocity profile is logarithmic. In equation (2.16) K

is the von Karnan constant, A, is the dimensionless thickness of the bottom layer, and


zo = is the dimensionless roughness height.
H

The vertical boundary conditions for the turbulent kinetic energy and length scale

equations are

2
q 2=BT, : z Z=1 (2.17)

2
q2 =B13 T : z =0 (2.18)

I=0 : z=0,1 (2.19)

The above set of equations forms a closed system that is solved by a numerical

method (Hamrick, 1992).










The model uses the finite volume method to bring the partial differential equation

into a discrete form. The Smolarkiewicz (1983) scheme is used to solve for the 2D

advection problem. An external/internal mode splitting procedure is implemented to

increase the numerical efficiency of the code.

2.3 Sediment Transport

The transport equation for a dissolved or suspended material having a mass per

unit volume concentration C is

(mm,,HC)+ (m,,HuC)+ a, (mHvC) + (mm, ,wC)- a, (m,m1,w,,C)
m m, mm K,, K (2.20)
=a-, f H ,.K C +aY xHKa ",v +la mxm C+,c Y
mx. 'm ) H

where KH and K,. are the horizontal and vertical turbulent diffusion coefficients,

respectively, w,, is (a positive) settling velocity when C represents suspended matter,

and Q. represents external, and reactive internal, sources and sinks.

Due to a small numerical diffusion that remains inherent in the scheme used to

solve the sediment transport equation, the horizontal diffusion terms are omitted from

equation (2.20). This results in

a, (mxmHS,)+ a, (m HvS,)+ a, (m.,HvS,)+ a (mm, mwS,)
-aK(m rniwS= as + i, (2.21)
a (..( m ,w s ,)= a m, m a S + Q,'; + Qo;

where S, represents the concentration of the j-th sediment class. Source-sink are

represented by two terms: an external part, which would include point and non-point

source loads, and an internal part, which could include reactive decay of organic











sediment, or exchange of mass between sediment classes when floc growth and breakup

are simulated.

The vertical boundary conditions for equation (2.21) are

K
K- K S -wS=J :z=0
H o
H (2.22)
K
a.S -wS- = 0: z = 1


where Jo is the net water column-bed exchange flux (Hamrick, 1992).


2.4 Settling Velocity Calculation

2.4.1 Background

A settling velocity algorithm was implemented, as part of the present study, in the

sediment transport code. The algorithm calculates the settling velocity of the particles by

accounting for the floc growth and breakup processes that occur for fine-grained

sediment in estuarine and coastal waters due to different mechanisms. As a result, instead

of using the settling velocity measured in a laboratory settling column in still water

directly, the model is merely calibrated using laboratory data.

There are a number of models in which the settling velocity is expressed as an

analytical function of the shear rate and the sediment concentration. Also, there are some

models that take a different approach in which the settling velocity only depends on the

properties of the primary particles. Generally, the former use the median (or mean) size

of the particles and the latter use a multi-class approach, in which particle sizes are

defined by a discrete distribution function. Both approaches have their advantages and

disadvantages. The former models usually depend on empirical coefficients, about which

a very little is known in most cases. The latter, however, require significant











computational time and, even though they are useful in simulating simple cases, high

memory and processing power requirements make them almost unusable for simulations

with large grids and for significant periods of time. Accordingly, in this study an attempt

has been made to implement a simple settling velocity model dependent on empirical

coefficients, but which includes the basic physics of floc growth and breakup.

Floc growth/breakup can be triggered by different mechanisms Brownian

motion, turbulent motion and differential settling. The following model only accounts for

the turbulence effect on growth/breakup. By analyzing theoretical results presented in the

literature and carrying out their own experiments in a settling column, Stolzenbach and

Elimelich (1994) concluded that the effect of the differential settling is minor and may be

practically absent in many turbulent flow situations. This would be the case because the

probability of the event when a particle with a large settling velocity overtakes a particle

with a smaller settling velocity is small, due to the fact that the trajectories of the two

particles tend to be deflected from one another. Also, the Brownian motion effect on

growth/breakup can be considered to be negligible in estuaries (Winterwerp, 1998).

Dyer (1989) presented a schematic description of the dependence of floc (median)

diameter on both turbulence and sediment concentration, as shown in Figure 2-1.













FLCC DIAM

200

SI \' CONCENTRATION
t10 0 104 \ '- .01-,






SHEAI STRESS
dynes mn -2




Figure 2-1: Effect of sediment concentration and fluid shear stress on the median floc
diameter (Dyer, 1989).

At very low concentrations and shear stresses, collisions are rare and the floc

growth rate is very small. Increasing fluid shear increases the number of collisions, thus

forming larger particles. A further increase in shear stress, however, causes the floc

breakup process to dominate over the floc growth, thus decreasing floc size. Also,

increasing sediment concentration increases the particle collision frequency, thus causing

the median floc size to increase. However, above approximately 10 kg/m3, flocs start to

disaggregate due to the low shear strength of the larger particles, which makes them

fragile.

A study of growth/breakup of fine sediment based on multi-class grain size

distribution, two- and three-body collisions, Brownian motion and differential settling

was done by McAnally and Mehta (2001). They present a set of equations that

characterize the growth/breakup mechanisms. These equations depend on primary

particle properties, flow properties and two empirical parameters: a collision diameter

function and collision efficiency. (However, collision efficiency can also be related to the










collision diameter, thus the system only contains one empirical, heuristic parameter).

They propose expressions for both, collision diameter and collision efficiency functions,

based on dimensionless arguments, and provide fitted forms of those functions using

experimental data:

Collision diameter function:


F2 [ Ap (D,+D,,,) S D, ( T, j CEC (2.23)
Ap, ,,,' v So D, T, CECo

Collision efficiency function:
D, S 1-0.875 T CEC ____
,, n AD, so To CECo [,,,,,,,, T,
a = )e (2.24)
Ap, D, u, (D, + D,,, ) (2.24)
Ap,,,D,, ) v

where TI, is some function of the bracketed non-dimensional terms; indices i and m

denote the colliding particle size classes; D = diameter of primary grain; D, = reference

particle size; S = salinity; So = reference salinity; T = fluid temperature, deg Celsius;

To= reference temperature; D,, D,, =particle diameters, Ap, and Ap,, = differential

densities of the particles; v = kinematic viscosity of the fluid; u, = velocity of particles;

r,, = maximum flow induced shear stress in a spherical particle; ,,,,, = shear stress

experienced by the m aggregate and CEC, CECo = actual and reference cation exchange

capacities, respectively.

Parshukov (2002) presents a multi-class grain size settling velocity model, which

implements the collision mechanisms presented in McAnally and Mehta (2001), and

combines them with other turbulence related parameters and settling velocity expressions











to represent growth/breakup. Testing of the model was done using laboratory data

involving grid-generated turbulence and its effect on the settling of flocculated clays.

2.4.2 Particle Density and Fractal Representation

For estuarine flocs the relation between the volumetric and mass concentrations,

0 and c, respectively, and between 0 and the number of particles per unit volume, n, is

given by:


0 = P C = jfnD (2.25)


where p, = is the sediment density; p, = floc density; p,, = water density; c = sediment

concentration; f = is a shape factor (for spherical particles f/ = r/6); and D = particle

diameter.

It is has been shown elsewhere that mud flocs can be treated as fractal entities

(Krone, 1984, Huang, 1994). Kranenburg (1994) shows that the differential density Apl

can be related to the fractal dimension of the particles using the formulation


Ap, = p, p ,, -P,) (2.26)


where D is the particle diameter, Dp is the diameter of the primary particles and nf is

the fractal dimension. The fractal dimension for strong estuarine flocs is found to be in a

range of 2.1-2.3 (Winterwerp, 1998).

2.4.3 Settling Velocity

A settling velocity function including the effects of both concentration and fluid

shear rate was proposed by van Leussen (1994):










(1+aG)
Wb =W ( G) (2.27)
(l+bG )

where W, and W,o are the actual settling velocity and reference settling velocity,

respectively, G is the dissipation parameter or the rate of flow shear, and a,b are

empirical constants.

Teeter (2001) proposed a more advanced functional relationship:

C I+a2GC
W= C -tl+aG exp (-a4 C 1 (2.28)
Cul 1+aG 3 C/

where C,, and C,, are mass-weighted average upper and lower reference concentrations,

respectively; and n, a, a2, a3 and a4 are the empirical parameters.

Using the force balance for a settling particle, one can obtain an implicit formula

for the settling velocity in still water, which depends on the fractal dimension

(Winterwerp, 1998):


W = (,-P.")g3, D"' (2.29)
S18/8 p "P 1+0.15 Re0687 o

where a, ,f are the coefficient that depend on particle sphericity, these coefficients will

be taken as 1 (spherical particles) here; up =dynamic viscosity and Re = settling

p,,.W, D1,
Reynolds number,


Figure 2-2 shows the relationship between the floc diameter and the settling

velocity in still water, as described by equation (2.29).












D vs Ws
10,






102


:102' ..;*. : : .f 3 *y . ... .. : .: .

E .. nf=2.
.. . ... : . : ... . . ... . . n f= 1 : . .

100









10...--2. . . ..... ... . . ...
101 102 103 104
D, mu

Figure 2-2: Relationship between settling velocity and floc diameter in still water, based
on equation (2.29).

2.4.4 Floc Growth and Breakup Functions

In the following equations the effect of turbulence is expressed through the energy

dissipation parameter G, defined as




du

G = l-dz (2.30)



where e = energy dissipation rate of flow, v = kinematic viscosity of the fluid, 1t =


dynamic viscosity of the fluid, z = fluid shear stress and u = mean flow velocity.

Levich (1962) determined the rate of coagulation of particles in a turbulent fluid

by integrating the diffusion equation over a finite volume n:










dn 3
= --e eceGD3n2 (2.31)
dt 2

where e,.,e, are the collision and diffusion efficiency parameters, respectively.

Combining it with equations (2.25) and (2.26) yields the expression for the rate of growth

of particle:

dD=3 ere, CGD"'-3D44/ = k cGD4-''
dt 2 fnf p (2.32)

where

3e, ret D'- D i-3
k e =k, f (2.33)
2Th nfp,, nfpu ,

The rate decay of particle due to breakup is suggested as:


dD ae D -G(2.34)
dD ae, DG D-D uI 1 = -k,G"+ (D-Dr)P D2q+ (2.34)
dt nf D, F



k, aeD DPj-/D D,' (2.35)
"/ [F,) nf 1 F,-

where p and q are the empirical parameters; e, is a floc breakup efficiency parameter

and F, is the floc strength (assumed to remain constant, due to the fractal structure of

flocs and determined by the number of bonds in a plane of failure) (Winterwerp, 1998).

Combining the rates due to floc growth and floc breakup yields the

expression for the net rate of change of rate due to turbulence:

D k c GD"' -3D4-"n /k GI+ D-D2q+1 (D Y (2.36)
dt nf p, nf F,











Winterwerp (1998) assumed that the parameters k and k( remain constant,

noting that the empirical coefficients they depend on are poorly known. In the present

study, k' is assumed to remain constant, since it depends only on floc strength (which as

noted remains unchanged due to the fractal nature of the particle). However, since the

growth process is more complicated than breakup, we will allow k' to remain a variable,

enabling a differentiation between the flocculation and hindered modes of floc settling

during the calibration process. In the former mode the settling velocity increases with

increasing concentration, whereas in the latter mode, which occurs at higher

concentrations, the settling velocity decreases with increasing concentration. Note,

however, that limitations are imposed on the choice of k,, based on the reported ranges

of e. = 010-2} and e,, =0.5-1.0 (Levich, 1962; O'Melia, 1985). Winterwerp (1998)


estimated k0 = 0{10-5 .
















CHAPTER 3
CEDAR RIVER ESTUARY

3.1 Description of the Estuary

The Cedar River estuary is contained within Duval County in northeast Florida

Figure 3-1.
























Figure 3-1: Cedar/Ortega Rivers estuary, (within the light rectangular area), aerial photo,
May, 1998.

Both the Cedar River, and the Ortega River into which it flows, together empty

into St. Johns River, which is connected to the Atlantic Ocean. Two main tributaries feed

Cedar River--Butcher Pen Creek and Williamson Creek--along with several smaller

tributaries. Another stream, Fishweir Creek also flow into the Ortega (Figure 3-2).














D
Site 1






Site 2
IQ


St Johns
River
WGC
o


O TG1
*0Se
oP


TG2
0


Ortega
River


ADCP
Transects

0 L>
/'


Ortega
River


1 kilometer


Figure 3-2: Cedar/Ortega Rivers data collection and sediment treatment (Wet Detention
System) sites 1 and 2.











Depths within the Cedar/Ortega system vary from 0.5 to 3 m with an average

depth of just over 1 m. Depths at the Cedar River vary from 0.3 m to 1.5 m with an

average of 0.5 m.


3.2 Tide, Waves, Current, Salinity and Wind Data

3.2.1 Tide data

Three ultrasonic recorders (Infinities USA Inc.) were installed to measure tidal

elevations; data included here are for the period of November 29, 2000 through May 17,

2001. The locations were chosen so as to cover a relatively large area of the estuary, and

also to facilitate gauge installation/removal and data retrieval. All gauges were placed

against bridge piers. Gauge locations are shown in Figure 3-2 (TG1-TG3). Tidal

elevations were measured relative to the National Geodetic Vertical Datum (NGVD).

Sampling interval was set at 30 min. From the ogive curves presented in Figure 3-3 we

note that the median range was around 30 cm. Station TG2, being the closest to the St.

Johns River, in the wide portion of the Ortega River, responded to the tide the most. TG1,

in Cedar River responded significantly less than the other two, possibly due to fresh

water outflows from Butcher and Fishweir Creeks, which would have opposed the tide.





































0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

Figure 3-3: Tidal ranges at stations TG1-TG3. Cumulative frequency distribution based
on record obtained during 09/29/00-10/18/01.

3.2.2 Current data

A tethered current meter (Endeco 174) was installed at location WGC (Figure 3-

2). The data reported here is from an approximately 1 month-long record (speed and

direction) collected from February 5 through March 8, 2001, until the device

malfunctioned. Data sampling interval was 15 minutes. As seen from Figure 3-4, the

current speed was below 30 cm/s 98% percent of the time, and below 25 cm/s 95% of the

time. Hence currents in the estuary are not very strong, and cannot be expected to result a

high level of sediment transport under normal weather conditions.












Current speed, cumulative distribution


0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4
Speed, m/s

Figure 3-4: Current speed at station WGC (mouth of the Ortega in St. Johns River).
Cumulative frequency distribution based on record obtained during 02/05/01-03/08/01.

3.2.3 Wind data

Wind record was obtained from the Jacksonville Naval Air Station for the period

January 1, 1995 through December 31, 1998 (sampling was every 3 hours). Wind

statistics derived from this record (Table 3-1) provide information on the dominant wind

speeds and directions. These data indicate a potentially complex dependence of wind on

wind-driven currents in the estuary, especially because portions of the waterway reaches

are lined by trees, while others have been cleared and developed.















Table 3-1: Wind speed/direction distribution. Dominant speed/directions are highlighted.


0 1


2 3 4 5 6 7 8 9 10 11 12 13 14 15 %


Speed,
m/s
24

48

72

96

120

S144

168
0
S 192

S216

240

264

288

312

336

360

%
occurrence


7s.i 11.5 I 20.8 I 22.7 18.2 I 9.6 6.6 3.2 0.0 0.0


15 171 382 486 332 174 78 32 10 4 1 1 1 4.9

21 184 389 413 287 129 82 32 19 9 4 1 3 4.6

47 413 722 778 627 343 236 154 109 59 17 8 2 4 10.3

43 328 579 575 402 211 135 96 53 24 5 5 5 1 1 7.2

25 164 310 362 288 137 122 49 22 3 4 4.3

23 213 419 461 373 213 146 111 31 7 3 4 5.8
4.8
23 148 298 377 321 181 145 97 27 10 4 1 1 4

1368 916 836 766 542 318 300 112 40 15 14 2 2 15.2

30 178 458 453 380 243 145 89 31 13 13 10 1 5 6.0

24 162 371 414 412 273 193 109 55 16 8 2 1 5.9

29 272 527 513 519 283 183 55 9 6.7

24 195 380 354 308 150 104 20 8 3 8 3 1 4.5

33 257 576 617 463 252 186 51 16 12 8 3 2 7.2

11 139 330 403 308 129 69 25 7 2 5 2 4.2

27 214 548 814 687 274 142 50 15 5 4 1 1 8.1


11.5 20.8 22.7 1 18.2 9.6


6.6 1 3.2


0.0 1 0.0


1 5.1













3.2.4 Wave data

A pressure transducer (Transmetrics Inc.) at WGC was used to obtain wave data.

The sampling interval was 6 hours. Data included here are for the February 10 through

April 25,2001. The modal period was found to be 2 s. Figure 3-5 is a spectral


representation of data. The variation of the significant wave height H,,, corresponding to


the modal period is shown in Figure 3-6.


The cumulative distribution of H,,,, shown in Figure 3-7, indicates that it did not


exceed -0.2 m. This in turn implies a mild wave climate, due to the limited wind fetches

in the St. Johns River. Wave action in the Cedar River is believed to be even milder, and

is unlikely to contribute much to sediment transport except under severe conditions when

comparatively large waves may break along the banks.






0.16-

0.14-

0.12,

0.1
E
L-0.08

a 0.06,

0.04-

0.02

160
140
120
0.4
100 0.3
80 0.2
60 0.1
Time, Days 40 Frequency, Hz


Figure 3-5: Wave spectrum based on 10/02/01 -25/04/01 record at WGC.





























0 15-
015




0.05



40 50 60 70 80 90 100 110 120 130 140 150
Time, days


Figure 3-6: Significant wave height based on spectral analysis.

Wave data. HmO, from spectral analysis. Cumulative distribution.
100

90-

80-

70-

60

50











Hm0, m




Ortega River in the St. Johns River.
40

30

20

10


0 0.05 0.1 0.15 0.2 0.25 0.3 0.35
HmO, m


Figure 3-7: Cumulative distribution of the significant wave height at the mouth of the
Ortega River in the St. Johns River.


3.2.5 Salinity data


Salinity data were collected at stations TG1-TG3 during October27-


November 26, 2000. Figure 3-8 shows the cumulative distribution of these data for each


station. The water is generally brackish; at TG1 in Cedar River the salinity is low due to


freshwater outflows from the river itself as well as the creeks that flow into it.














Salinity. Cedar/Ortega Rivers. Cumulative distribution.
100 .- -
TG1
TG2
90- TG3
I /
80

70 -

60

E 50
SI I

40 -

30

20

10 /


0 2 4 6 8 10 12 14
Salinity, ppt


Figure 3-8: Cumulative distribution of salinity at stations TG1-TG3; 10/27/00-11/26/00.


3.3 Discharge Data


Discharge measurements were obtained using an ADCP (Acoustic Doppler


Current Profiler) Workhorse 1200 kHz (RD Instruments Inc.), on May 17, 2001. Four


cross-sections of the Cedar/Ortega River confluence area were selected for data


collection for almost a full semi-diurnal tidal cycle. In addition to the flow data, water


samples were collected for determination of sediment concentration.


Due to the shallow nature of the estuary and poor performance of ADCP in


shallow waters, the data were found to have a somewhat qualitative significance. A large


fraction of the total discharge had to be estimated, Appendix A describes the estimation


algorithm, which closely follows the estimation implemented in WinRiver (RD


Instruments software for ADCP systems). Table 3-2 presents the analyzed data for the


Cedar River cross-section (all transects were made at the cross-section of the Cedar River


near the confluence; see Figure 3-2). Positive discharge is directed west, and negative is


directed east.













Table 3-2: Cedar River cross-section discharges, May 17, 2001.
ADCP transect # Time Discharge, m3/s
6 12:18 +90
14 13:25 -23
24 15:46 -69
32 17:13 -14
40 19:31 -3
50 20:29 +1

The mean depth at the cross-section was 1.0 m and maximum depth 1.4 m.

Figure 3-9 shows discharge plotted on the same scale as tide data at TG2. It can be seen

from the figure that as expected, the discharge curve precedes the water elevation curve.

Discharge data. Cedar River cross-section.
100 1 I I I


14 15 16 17 18
Time, hr
Water level at Main Bridge Cedar/Ortega River


19 20 21


Figure 3-9: Cedar River cross-section discharge and Cedar/Ortega tide data,
May 17, 2001.

3.4 Sediment Concentration

As noted, during the ADCP survey on May 17, 2001, a set of water samples was

also collected. All samples were filtered, dried and weighed to determine the sediment


-10nnL













concentrations. Table 3-3 gives the data thus obtained. The concentrations are

characteristically low, and range between 8 and 101 mg/1.

Table 3-3: Sediment concentrations from water samples, May 17, 2001
e # Concentration, # Concentration, e Concentration,
Sample # Sample # Sample #
mg/l mg/1 mg/1
1 14 31 14 61 16
2 17 32 14 62 15
3 16 33 13 63 16
4 17 34 17 64 20
5 8 35 18 65 15
6 22 36 17 66 15
7 14 37 17 67 14
8 13 38 15 68 16
9 35 39 11 69 16
10 23 40 11 70 26
11 15 41 15 71 15
12 15 42 14 72 14
13 20 43 18 73 101
14 15 44 12 74 19
15 14 45 19 75 17
16 15 46 37 76 16
17 19 47 13 77 15
18 16 48 17 78 57
19 13 49 16
20 27 50 19
21 33 51 17
22 17 52 21
23 19 53 16
24 16 54 18
25 14 55 13
26 16 56 13
27 17 57 16
28 17 58 15
29 16 59 16
30 13 60 16


3.5 Bed Sediment Distributions


In order to represent bed sediment distribution patterns, bed-sampling data

supplied by the St. Johns River Water Management District were used to generate the

maps showing the distribution of solids content, moisture content and organic content in












the Cedar/Ortega Rivers. This set of maps, based on data obtained during March 3 -

October 2, 1998, were generated using approximation methods as follows.

First, maps were generated using Matlab routines for surface fitting (meshgrid,

griddata). These routines generate a rectangular grid covering the data set supplied to

them. Grid values are then approximated by fitting the surface to the data points and

determining the values at the grid points. This approximation caused two problems:

Due to the lack of adequate spatial coverage of data, values at the river boundaries

were automatically assigned zero values;

River boundary presence was not considered when generating the surface.

In order to avoid these problems, first the boundary points were approximated.

The method used for this approximation was as follows.

The distance from each data point to each boundary point was calculated (the

shortest distance possible following the river). Boundary points (V,) were then evaluated

as:


V = -, v, (3-1)
I =1



S= (3-2)

e
/-1

in which v,= data point value, a = standard deviation of the dataset, and r,= distance to

data point i from boundary point j.

After generating the boundary points they were merged with the measured data

points, and surface fitting functions were applied to the combined data. In this way











contour maps were produced for moisture content (Figure 3-10), organic content

(Figure 3-11), solids content (Figure 3-12) and thickness of the deposit (Figure 3-13).

Measurement points are displayed as black dots and have values besides them. Also,

equal-percent contours are drawn to identify the areas where the percentage of

moisture/organics/solids is approximately the same.

Table 3-4 contains relevant statistics: minimum, maximum and the mean values

of moisture content, organic content and solids content. We observe that the upstream

reach of the Ortega River is characterized by the high organic content (30-35%), whereas

the upstream reach of the Cedar River has a high solids percentage (25-30%). These

trends reflect the more natural, vegetated surroundings of the Ortega versus more

developed reaches of the Cedar. A thickness of the deposition layer is more or less

constant in the Ortega/St. Johns River and is much smaller in the Cedar River.

Table 3-4: Statistical values associated with bed sediment distribution
Statc Moisture content Organic content Solids content
Statistic
(%) (%) (%)
Minimum 54 6 16
Maximum 84 51 46
Mean 76 21 24





















































Figure 3-10: Moisture content distribution (%). Based on 1998 sampling.






















































Figure 3-11: Organic content distribution (%). Based on 1998 sampling.





















































Figure 3-12: Solids content distribution (%). Based on 1998 sampling.
























































Figure 3-13: Thickness of soft deposit in the study area based on core thicknesses in 1998
sampling.
















CHAPTER 4
ASSESSMENT OF SEDIMENT TRAPPING EFFICIENCY

4.1 Flow Model Setup, Calibration and Validation

Model setup was carried out in two steps. The area of interest lies within the

Cedar River, while some of the data for calibration were available for various sites within

the much larger Cedar/Ortega/St. Johns River estuary. It was therefore decided to

calibrate and run the model with a coarse grid covering the Cedar/Ortega/St. Johns River

estuary. Running the model in this way generated the downstream boundary conditions

for the Cedar River model, which was then run for sediment trapping efficiency

assessment. This procedure allowed for the use of a finer grid in the Cedar River, without

significantly increasing the simulation time.

Figure 4-1 shows the areas covered by the two model setups. Both, the coarse and

fine grids are Cartesian because, as noted, the sediment transport model was found not

conserve mass when run with a curvilinear grid.

4.1.1 Cedar/Ortega/St. Johns Rivers Model Setup

The Cedar/Ortega/St. Johns River (Cartesian) grid and has dimensions of 160 by

300 cells, with a cell size of 50 by 50 m (Figure 4-2). The boundary conditions are

labeled BC1-BC7. The grid was o -stretched in the vertical direction with six horizontal

layers.

Figure 4-3 shows the bathymetry of the modeled domain. The Cedar/Ortega River

portion of the domain is typically shallow with 1.5 2 m depth in the channel thalweg,

and an average depth of m.



















Area covered by
the Cedar River
Sgrid only


Area covered by the
Cedar/Ortega/St. Johns
grid only


Area covered by
both grids


Figure 4-1: Areas covered by the two (coarse grid and fine grid) models.














BC2
BC4

x


BC6

BC5





BC3


BCl


S

Figure 4-2: Cedar/Ortega/St. Johns River grid with open boundary locations.


BC7
gk^





















































Figure 4-3: Cedar/Ortega/St. Johns River bathymetry. Depths are in meters.

Hourly water level and salinity data from the St. Johns River (supplied by the

St. Johns River Water Management District) were used to define the boundary conditions

at the north and south open boundaries of the St. Johns River (BC7, BC8). As an













example, Figure 4-4 shows the time-series and the cumulative distribution of the

measured water surface elevation data. The mean tide range is approximately 0.4 m.

0.6

0.4
0.2
0
0 0
-0.2
M -0.4
-0.6

-0.8
90 100 110 120 130 140 150 160
Time, days


100

80-

6 60

B 40-

20-


0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Water elevation, m

Figure 4-4: Cedar/Ortega/St. Johns River open boundary condition (BC7) during 2001
showing water surface elevation time-series and cumulative distribution.

Small creeks (Williamson Creek, Butcher Pen Creek, Big Fishweir Creek and


Fishing Creek denoted on Figure 4-2 as BC6, BC5, BC3 and BC4, respectively)

minimally affect the flow in the larger estuary. Hence, instead of using "open" boundary

conditions there, they were defined in terms of sink/source cells for specifying the flow

and sediment flux conditions at the heads of these creeks. The relevant boundaries BC2-

BC6 are shown in Figure 4-5. The boundary condition time-series were supplied by the

St. Johns River Water Management District.

























E 50-

40 /

30

20

10 ^
2 I

0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5
Discharge, m3/s

Figure 4-5: Cedar/Ortega/St. Johns River tributaries discharge, cumulative distribution.

Measured salinity time-series at open boundaries BC7 and BC8 were used. The

model was run for a thee-week period to establish the salinity field (model "spin-up"),

defined by these time-series. The bottom roughness z0 was chosen as 0.04 m throughout.


The kinematic viscosity and molecular diffusivity were set to 10-6 and 108 T2/S,

respectively [equations (2-1) (2-4); Section 2.2]. The period of simulation corresponded

to April 25, 2001 through May 30, 2001. A time-step of 3 s was used.

The purpose of the simulation run was to generate the flow, salinity and

suspended sediment time series at the downstream boundary of the Cedar River model

(near the Cedar/Ortega confluence), and also to establish a (conservative) salinity field

over the estuary. These outputs were then used to generate the initial and boundary

conditions for the Cedar River model.








43


4.1.2 Cedar River Model Setup

The Cedar River grid was also horizontally Cartesian and vertically o-stretched.

Its horizontal dimensions were 160 by 450 cells, each cell representing an area of 15 m

by 15 m, and also used six horizontal layers. The grid and bathymetry are shown in

Figure 4-6 and Figure 4-7, respectively.
















EFDC Cedar River grid


Tap 1 (CP1)


Hilliamson Creek


Trap 2 (CP2)


Butcher Pen Creek


Fishing creek


Figure 4-6: Cedar River model grid.


Mrap 3 (CP3)




























































02 03 04 05 06 07 08 Q9


1 1.1 1.;


Figure 4-7: Cedar River bathymetry. Depths are in meters.


I I I __











The boundary conditions at the upstream ends of the Butcher Pen Creek, Fishing

Creek, Williamson Creek and the Cedar River were forced by establishing sink/source

cells with the given discharge time-series (Figure 4-5). The downstream open boundary

was represented by the water surface elevation forcing time series (Figure 4-8) generated

by the Cedar/Ortega/St. Johns River model. The same physical boundary also served as a

boundary condition for sediment concentration, which was defined based on the water

sample data (May 17, 2001). The initial salinity field was generated by approximating the

salinity field from the Cedar/Ortega/St. Johns model, which helped in decreasing the time

needed for model "spin-up" required to establish a conservative salinity field. The bottom

roughness coefficient, viscosity and diffusivity were unchanged from the values used in

the Cedar/Ortega/St. Johns River model.

Figure 4-8 shows a reasonable agreement between the measurement and

simulation of tide. Note that the "measured" time-series was derived by averaging the

tides at stations TG2 and TG3 by taking the time lags into consideration, in order to

represent tide at the open boundary, which occurred in-between the two tide stations. The

mean range was 0.52 m for the measured tide and 0.50 m for the calculated one. Figure 4-

9 shows the predicted water surface elevation plotted over a short period of time at three

control stations (cells): 1) at the confluence of Butcher Pen Creek and the Cedar River),

2) at the upstream-most cell of the Cedar River grid, and 3) at the cross-section in Cedar

River at its confluence with the Ortega. The time lag between the upstream control

station and the confluence station is equal to approximately 7.5 min, which is consistent

with the registered time lag in the measured water level data, considering the distance

(5.4 km) between the stations.













The simulated discharge data were compared to discharge obtained by the ADCP.


The measured and simulated discharges in the Cedar River at the confluence are plotted

on Figure 4-10. Figures 4-11 and 4-12 show similar results for the north and south cross-


sections of the Ortega River, respectively. The latter two simulations were derived from


the Cedar/Ortega/St. Johns River model. In general, the simulated discharge appears to be


in a reasonable agreement with measurement, especially considering measurement errors


(see Appendix A).


0.5
measured (avg of 2 stations)
calculated
0.4


0.3


0.2 -


0.1


E 0


-0.1


-0.2


-0.3 -


-0.4


-0.5
122 124 126 128 130 132 134


Figure 4-8: Measured and simulated water level variations at the downstream boundary
of the Cedar River.











48



0.15
- Butcher Pen creek
upstream
confluence



0.1



E .

-- -.
0.05














-0.05

122.3 122.305 122.31 122.315 122.32 122.325 122.33 122.335 122.34 122.345 122.35
Time, days


Figure 4-9: Water surface elevation at three control points in Cedar River.


100
x- Simulated discharge
x ADCP measured discharge
x ADCP estimated total discharge


50 -




E \*






S-50





-100





-150
137 137.1 137.2 137.3 137.4 137.5 137.6 137.7 137.8 137.9 138
Time, days


Figure 4-10: Measured and simulated discharges through the Cedar River cross-section.





























E 20-
















80
-60-


-80 L----L---L----L---L---L----L-------L-
137 137.1 137.2 137.3 137.4 137.5 137.6 137.7 137.8 137.9 138
Time, days


Figure 4-11: Measured and simulated discharges through the Ortega River cross-section

(north cross-section of the confluence).


80 r r-r
Simulated discharge
ADCP measured discharge
60 x ADCP estimated total discharge





20


E 0
2O

20 *


-40


-60


-80


-100 L
137 137.1 137.2 137.3 137.4 137.5 137.6 137.7 137.8 137.9 138
Time, days


Figure 4-12: Measured and simulated discharges through the Ortega River cross-section
(south cross-section of the confluence).











4.2 Sediment Transport Model Setup and Calibration

4.2.1 Sediment Transport Model Setup

For running the sediment transport model the initial suspension concentration was

set to 5 mg/l, the average value for the sediment concentration in the

Cedar/Ortega/St. Johns River estuary. For the upstream boundary conditions in Cedar

River, Williamson Creek and Butcher Pen Creek supplied by the St. Johns River Water

Management District were used. It should be pointed out that at the downstream

boundary of the Cedar River, the outputted values from the Cedar/Ortega/St. Johns River

were significantly lower (-5-12 g/l), than the values, ranging between 8 and 57 mg/1

(with an additional, exceptional value of 101 mg/1 in one case), obtained from water

sampling on May 17, 2001. Depth-averaged TSS concentration series simulated by the

coarse grid model at the Cedar River cross-section near the confluence is shown in

Figure 4-13 and commensurate collected samples at the same location are shown in

Figure 4-14.
















Sediment concentration at the Cedar River cross-section near the confluence


137 137.1 137.2 137.3 137.4 137.5 137.6 137.7 137.8 137.9 138
Time, days


Figure 4-13: Depth averaged TSS concentrations at the Cedar River cross-section near
the Cedar/Ortega confluence simulated by the coarse grid model (May 17, 2001).


137.5 137.55 137.6 137.65 137.7 137.75
Time, days


137.8 137.85 137.9


Figure 4-14: Depth averaged TSS concentrations from the water sample data, collected at
the Cedar River cross-section near the Cedar/Ortega confluence (May 17, 2001).


*
*














*





*
*


137.95











The above discrepancy between the simulated and measured concentrations was

found to be due to the low concentrations predicted at the head boundaries of the Cedar

River and the creeks. These boundary conditions, supplied by the St. Johns River Water

Management District, were not verified. The problem was unfortunately realized towards

the end of the present study. It was however felt that rerunning the sediment transport

calculations for the Cedar River was not necessary, because the trapping efficiency

results, described later, rely on relative rather than absolute values of the sediment flux.

Thus the conclusions of the study were not affected.

Since variation in the sediment concentration with time was small (8-9 mg/1),

compared to the increase in concentration with depth (as found from the water sampling

analysis), the sampled data were averaged, and a representative vertical profile of

concentration with linearly distributed values from 14 g/1 in the top layer to 27 g/1 in the

bottom layer, was used to set the open boundary condition at the in the Cedar River.

4.2.2 Bed Erosion

The bed erosion function (lines representing erosion rate as a function of the bed

shear stress) required for sediment model code is shown in Figure 4-15. It was based on

laboratory experiments by Gowland (2002) using mud samples collected from the Cedar

and the Ortega Rivers. This function was used for both models, i.e., coarse and fine grid.













x10-4
Measured data
Linear fit
*

0.8


Z*

S0.6 -



0
0.4 -


*

0.2





0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45
Bed shear stress, Pa

Figure 4-15: Bed erosion rate function obtained from laboratory experiments on mud
from the Cedar/Ortega Rivers (after Gowland, 2002)

4.2.2 Settling Velocity and Deposition

The settling velocity model (Section 2.4) was calibrated using the data obtained

from laboratory settling column tests using sediment from the site (Gowland, 2002).

Some model tests were also carried out it against data available in the literature, the

results of which are given in Appendix B.

The model did not function for the values of the dissipation parameter G on the


order of magnitude of 10.3 Hz and less, because very low turbulence levels caused the


particle to grow infinitely large. This was due to the model formulation, in which particle

size is dependent on a level of turbulence. It should be noted that in reality the volumetric











c D
concentration oc cannot exceed unity, by definition, hence the constraint
p D1,


D <: D is imposed on the calculation of the diameter.
c

To calibrate the model against the laboratory data (Gowland, 2002) the value of

G = 10-2 Hz was used, in order to satisfy the above constraint and, at the same time, to

simulate a near-quiescent situation (the settling column being a quiescent environment).

The estimated range of G for Cedar River estuary was found to be within 0.5 to 10 Hz

[based on equation (2-30), which gives a relation between the flow velocity, shear stress

and dissipation parameter]. The concentration in the Cedar River (from the water

samples, collected in May 17, 2001) was found to be within 8 to 57 mg/1; however for the

modeling purposes the range of concentrations from 0.1 to 100 was selected for

convenience. Figure4-16 thus obtained qualitatively resembles the Dyer diagram

(Figure 4-16). The dissipation parameter was related to the shear stress using equation (2-

30) considering the mean flow of 10 cm/s.






















E 200-






50-

0





0.8 E 102
2 1 1
Shear stress N/m2 10 Concentration mg/

Figure 4-16: Calculated floc size as a function of shear stress and concentration.

Figure 4-17 shows settling velocity as a function of concentration, based on the

laboratory settling column data using sediment from the site (Gowland, 2002). The curve

is described by the equation

aC"
W = a(4.1)
(b2 +C2)


with the parameters a, b, n and m set to 0.035, 2.0, 3.5 and 2.75, respectively.






















W -
E 10




I 10
0)


10-1 100 101 102
Concentration, kg/m3

Figure 4-17: Settling velocity curve based on laboratory tests in a settling column using
sediment from the Cedar River and vicinity (after Gowland, 2002).

For calculation purposes, the value of the fractal diameter, nf [equation (2-26)],

was taken as 2.3. By fitting the settling velocity predicted by the model to the curve given

by Figure 4-17, the parameters (k,,k,, p and q) for the settling calculation velocity


were found. The exponents p and q were found to be 0.7 and 0.5, respectively, and the


growth/breakup efficiency coefficients k, = 10.3 and k, = 16.8 103 The values of these


coefficients are of the same order of magnitude as those of Winterwerp (1998) (k, = 14.7


and k, = 14.0 10'). Fluid properties were selected as p,w = 1,020 kg/m3 and v = 10-6 m2/S.












4.3 Trapping Efficiency Analysis

4.3.1 Treatment Plan

As described in Section 1.2, the effects of two sediment treatment sites were to be

tested. For the present purpose the locations of the sites (1 and 2) were changed

(Figure 4-6). Each site was tested with four assumed trapping efficiencies: 0% (no

trapping), 30%, 60% and 90%. The maximum efficiency (90%) is in part based on the

estimated 85% for TSS removal by Wet Detention Systems (WDS) in Florida; see

Table 4-1.

Table 4-1: TSS removal efficiencies of treatment systems in Florida (after Harper, 1997)
Treatment system Estimated TSS removal efficiency (%)
Dry Retention 60-98
Off-Line Retention/Detention 90
Wet Retention 85
Wet Detention 85
Wet Detention with Filtration 98
Dry Detention 70
Dry Detention with Filtration 60-70
Alum Treatment 90

For that purpose, the resulting (calculated) settling flux (total mass of sediment

passing the cross-section of the estuary in a unit of time) values at the Cedar River open

boundary were compared to determine the potential effect of trapping sediment near the

upstream end of the Cedar River (Site 3) on deposition downstream, where contaminated

sediments derived from upstream Cedar River tend to deposit. It should be noted that in

the Cedar River the direction of the water flow changes with flood and ebb tides; hence

the ebb tide is the only time when there is a sediment flux out of the river.

4.3.2 Sediment Trap Setup

The sediment trap at the treatment site typically is a water detention (i.e.,

temporary retention) pond. By diverting river flow into the pond where flow velocities

are small, a major portion of suspended sediments will typically deposit. Such systems











can also be effective for storm water treatment when the bulk of the solids is carried with

the first flush, as they can be intercepted and given a sufficient residence time to allow

them to deposit.

While some treatment facilities may require drainage pumps, others are strictly

gravity flow systems. If the water is high in nutrients, the facility may include a vegetated

wetland area that will absorb the nutrients in the water before it is discharged into the

receiving waters. The concern for the Cedar River treatment system was to provide as

much treatment as possible; hence the effectiveness of the facility was defined by the area

available for it.

Due to modeling limitations and related complications in representing the site as a

water body with channelized flow diverted into it, site representation in the model was

simplified. Accordingly, a function was implemented that decreased the sediment flux

bypassing the grid cell by a pre-defined percentage. The channel cross-section, where the

treatment site would be located, was represented by cells having such a sediment removal

function (in terms of the percentage by which the effluent sediment load, leaving the site,

is reduced with respect to influent load entering the site).

4.3.3 Effect of Trap Efficiency on Settling Flux Downstream

Cedar River model runs were run without and with the sites in place (Figure 4-6),

each for the selected four removal ratios (0%, 30%, 60%, and 90%). The model was run

for three days, during May 16-18, 2001. Four output control points (CP1-CP4) were

selected (Figure 4-6). CP1-CP3 corresponded to the sites and were placed just upstream

of a site to measure sediment flux into the site, and CP4 was the control point just

upstream of the open boundary, for monitoring trapping influence at the downstream end.












Net sediment fluxes at the control points averaged over three semi-diurnal tidal

cycles (the second cycle on May 17 and two cycles on May 18, 2001) are presented in

Table 4-2. As seen in this table, it can be inferred that Sites 1 and 2 in the upstream

portion of the Cedar River would have a small effect on sediment transport at the lower

end of the Cedar River. In contrast, Site 3 can be considerably more effective. The reason

for these differences appears to be that the majority of sediment load is derived from

Williamson and Butcher Pen Creeks, rather than the Cedar River.

Table 4-2: Comparison of sites with different removal efficiencies with a no-trapping
scenario.
Trap efficiencies (%) Net sediment flux g/s Resulting efficiency at the
confluence (%)
Trap 1 Trap 2 Trap 3 CPI CP2 CP3 CP4 confluence (%)
0 0 0 6.7 8.9 11.4 14.1 0.0
30 0 0 6.7 7.9 11.0 13.8 2.1
60 0 0 6.7 7.7 9.9 13.6 3.5
90 0 0 6.7 7.0 9.6 13.1 7.1
0 30 0 6.7 8.9 10.1 12.3 12.8
0 60 0 6.7 8.8 8.2 10.9 22.7
0 90 0 6.7 8.7 7.6 9.2 34.7
0 0 30 6.7 8.9 11.4 10.6 24.8
0 0 60 6.7 8.8 11.4 7.9 44.0
0 0 90 6.7 8.8 11.2 5.8 58.9
30 30 30 6.7 7.8 7.2 6.9 51.1
60 60 60 6.7 6.0 4.1 2.9 79.4
90 90 90 6.7 2.8 1.1 0.3 97.8

The above observations are further highlighted in Table 4-3, by taking the cases

of no entrapment and 30% entrapment (which may closer to a realizable efficiency), and

seeing the effect in the confluence area. From the table it appears that: 1) Any treatment

facility upstream of Williamson and Butcher Pen Creeks, as presently envisaged, will not

be effective in reducing sediment loading in the confluence area, 2) treatment

downstream of Butcher Pen will have measurable, but possibly not significant effect, and

3) more than one treatment site may have to be developed. In the event that a downstream

treatment site cannot be constructed, dredging a trap in the river bed at that site should be









60


considered. Such an action should preferably be coupled with a one-time dredging of the

confluence area to remove soft sediment deposit there.



Table 4-3: Summary of the effect of treatment on sediment load in the confluence area
Upstream Mid-stream Downstream Load reduction at the confluence (%)
Cedar Cedar Cedar
Treatment -2
Treatment -13
Treatment -25
Treatment Treatment Treatment -51

















CHAPTER 5
CONCLUSIONS

5.1 Summary

A study of the effectiveness of fine sediment trapping in the Cedar River estuary

in north Florida was carried out. A combined three-dimensional hydrodynamic and

sediment model was set up and calibrated for this estuary. The original model (EFDC)

was improved for fine sediment settling velocity calculations, by accounting for floc

growth and breakup processes due to turbulence. The effect of selected sediment

treatment sites or traps with different efficiencies (and placed in different locations

upstream) on sediment transport downstream was examined.


5.2 Conclusions

The following are the main conclusions of this study:

1. Simulated discharge and tidal variations in the Cedar River were found to agree

reasonably well with measurements.

2. The settling velocity calculation routine was found to be applicable to conditions when

the flow is turbulent, but not in near-quiescent waters, i.e., when the energy dissipation

parameter has low values.

3. It appears that fine sediment trapping in the upstream reach of the Cedar River would

have only a minor effect on sediment transport downstream near the confluence of the

Cedar and Ortega Rivers. This is so because a major part of suspended sediment flux











downstream appears to arrive there from creeks (especially Butcher Pen and Williamson)

that flow into the middle reach of Cedar River.

4. Sediment entrapment closer to the confluence of the Cedar and Ortega rivers appears

to be able to measurably reduce sediment transport to the confluence and, therefore, can

be expected to lower the flux of contaminants out of Cedar River.


5.3 Recommendations for Further Work

Further development of the settling velocity model is required, in order to extend

the calculation to settling in near-quiescent water.

Traps simulation should be made more realistic by incorporating the mechanics of

an actual retention/detention pond in the model.

















APPENDIX A
WATER DISCHARGE ESTIMATION BASED ON ADCP


Water discharge must be calculated for each ADCP transect. Due to the inability

of the ADCP used to record measurements close to the water surface and the bottom, as

well as in the shallow near-bank areas, a method must be used to account for the loss of

coherent signals for these blank zones in an approximate way. The following uses the

method suggested and used in RD Instruments, WinRiver (software designed by

RD Instruments for analysis and visualization of the ADCP data) and is described in a

help system provided with the software

The required total discharge (Q,,) from the instrument consists of measured

(Q,.e,,mre) and estimated (Q,, ) values:

Q1oa, = e, + Qe.a.red (A. 1)

The estimated discharge, which must be added to the measured value, consists of

four components: top (layer close to the surface), bottom (layer close to the bottom), right

and left (discharge in the zones close to the bank, where ADCP data are usually not taken

because of shallow water). The "estimated" discharge is then calculated as:

Q, = Q,,, + ,,,,,, + Q,,,,, + Qle,, (A. 2)

For calculating the top and bottom discharges two (user-definable) methods can

be used: Constant and Power. The Constant Method assumes that the velocity is constant

in the top/bottom layer and is equal to topmost/bottommost successfully measured











acoustic bin. The Power Method assumes a power-law velocity (u) profile in the vertical

(z) direction:

u = r.z (A. 3)

in which the exponent p is user-defined with a default value of 1/6, and the

proportionality constant r is found by fitting the power-law profile to the measured

points.

For calculating the right/left discharge the following formula is used:

Q=c-D-H-u,,, (A. 4)

where c is a user-defined coefficient equal to0.35 for a triangular near bank bottom

shape (default) and 0.91 for a rectangular near bank bottom shape; D is a distance to

bank (defined in a data collection process and obtained from the ADCP data files); H is

a water depth of the leftmost/rightmost measured ensemble; and u,,, is velocity averaged

over the user-defined number of leftmost/rightmost ensembles.


















APPENDIX B
SETTLING VELOCITY AND FLOC SIZE CALCULATIONS


B.1 Introduction

In order to demonstrate the application of the settling velocity model described in

Section 2-4 and to test the model against the data available in the literature, the following

calculation tests were performed.


B.2 Settling Velocity Calculations

Wolanski et al. (1992) presented data on the settling of sediment from Townsville

Harbor, Australia. A Plexiglas cylinder of 10 cm internal diameter and 140 cm height was

used as a settling column. Turbulence could be generated in this column by oscillating

1 cm wide rings along the walls, spaced 2 cm apart. Two sets of data were obtained: in

quiescent water, and with rings oscillating. Quiescent water can be characterized by very

low values of dissipation parameter G.

First, model predicted settling velocity was fitted to the data in oscillating flow

based on equations (2-29), (2-36) and floc aggregation coefficient in a form similar to

C"
equation (4-1), i.e., kA, =k bC2 The parameters k,,,k,, p and q were determined
(b+C2)

in this way; p and q were found to be 0.6 and 0.45, respectively, and. k, =8.7 and

k,=19.1.103 for n=0.87 and b=1.96. A representative value of the dissipation













parameter was found to be G= 1.3 Hz. Then G =102 Hz was used to represent


quiescent water and the corresponding velocity curve was plotted (Figure B-l).

In Figure B-l, the simulated curve based on measurement in oscillating water

indicates a reasonably good match with data points. However, measurements in quiescent

water are not predicted as well. This is believed to be due to the fact that, as noted in

Section 4.2.2, the model does not perform well for low values of dissipation parameter G

(i.e., in the absence of turbulence).


/*


Lab results (quiescent water)




x x
. , X xx .
X




Lab results
(Max. velocity=0.09 mis)


10-31 ; , : ; I ,--- I
10-1 100 101 102
Concentration g/l

Figure B-l: Settling velocity calculation test results, and comparison with data of
Wolanski et al. (1992) using sediment from Townsville Harbor, Australia.











B.3 Particle Size Calculations

In steady flows and with given sediment properties, flocs tend to have a narrow

size distribution and may be assumed to have an equilibrium size defined in terms of, for

example, the median diameter. The equilibrium size condition implies that the growth

and breakup processes balance each other. Thus, flocs that are smaller than the

equilibrium size would have growth dominating over breakup, and for larger flocs the

breakup process would be dominant. As a result the floc size tends to fluctuate around its

equilibrium value.

The model was tested against the floc size data published by Winterwerp (1998)

from two settling column tests using sediment from the Ems-Dollard River area in The

Netherlands. Particle sizes were measured using a Malvern particle sizer. For simulation

purposes the parameter values in Table B-l, plus those provided by Winterwerp, were

used. The initial particle size was taken as 4 pm, as measured by Winterwerp.

Coefficients k,, = 14.7 and k, = 14.0 103 were selected.

Comparisons between simulations and data are shown on Figure B-2. The values

of concentration and dissipation parameter used are given in Table B-1. The resulting

curves, which lead to equilibrium sizes, appear to be the same as those of Winterwerp

(1998). Floe size is seen to grow with time until it reaches an equilibrium value (there is

an equilibrium particle size for given concentration and dissipation parameter) and

remains the same beyond that point.

























E





10 -








10 I
10 10 10 10"
Time, s

Figure B-2: Floc growth with time measured and predicted for River Ems-Dollard mud
(Winterwerp, 1998).

Table B-1: Data from settling column tests with Ems-Dollard mud.
c G
Test No C G
TestNo (kg/m3) (Hz)
T-73 1.21 81.7
T-69 1.17 28.9


















REFERENCES


Dyer, K. R., 1989. Sediment processes in estuaries: Future research requirement. Journal
of Geophysical Research, 94(C 10), 9489-9498.

Ganju, N. K. 2001. Trapping organic-rich sediment in an estuary. M.S. Thesis, University
of Florida, Gainesville.

Gowland, J. E., 2002. Laboratory experiments on the erosional and settling properties of
sediment from the Cedar/Ortega River system. Report UFL/COEL-CR/2002/001,
Coastal and Oceanographic Engineering Program, Department of Civil and
Coastal Engineering, University of Florida, Gainesville.

Hamrick, J. M., 1992. A three dimensional environmental fluid dynamics computer code:
Theoretical and computational aspects. Special Report No 317, Applied Marine
Science and Ocean Engineering, Virginia Institute of Marine Science, Gloucester
Point, VA.

Hamrick, J. M., 1996. User's manual for environmental fluid dynamics computer code.
Special Report Special Report No 331, Applied Marine Science and Ocean
Engineering, Virginia Institute of Marine Science, Gloucester Point, VA.

Harper, H. H., 1997. Pollutant removal efficiencies for typical stormwater management
systems in Florida. Proceedings of the Biennial Stormwater Research Conference,
Southwest Florida Water Management District, Tampa, FL, 6-19.

Huang, H., 1994. Fractal properties of flocs formed by fluid shear and differential
settling, Physics of Fluids, 6(10), 3229-3234.

Levich, V. G., 1962, Physicochemical hydrodynamics, Prentice Hall, Inc.

McAnally, W. H., Mehta, A. J., 2000. Aggregation rate of fine sediment. Journal of
Hydraulic Engineering, 126(12), 883-892.

McAnally, W. H, and Mehta, A. J., 2001. Collisional aggregation of fine estuarine
sediment. In: Coastal and Estuarine Fine Sediment Processes, W. H. McAnally
and A. J. Mehta eds., Elsevier, Amsterdam, 19-40.











Mellor, G. L., and Yamada, T., 1982. Development of a turbulence closure model for
geophysical fluid problems. Reviews in Geophysics and Space Physics, 20, 851-
875.

National Research Council, 2001. A risk-management strategy for PCB-contaminated
sediments. National Academy Press, Washington, DC.

Parshukov, L. N., 2001. Effect of turbulence on the deposition of cohesive flocs. M.S.
Thesis, University of Florida, Gainesville.

RD Instruments, 1994. Transect: User's Manual for Broadband Acoustic Doppler Current
Profilers, San Diego, CA..

Smolarkiewicz, P. K., 1983. A simple positive definite advection scheme with small
implicit diffusion. Monthly Weather Review, 111, 479-486.

Stoddard, D. M., 2001. Evaluation of trap efficiency in an estuarine environment. Report
MRP-2001/003, Department of Civil and Coastal Engineering, University of
Florida, Gainesville.

Stolzenbach, K. D., Elimelich, M., 1994. The effect of density on collisions between
sinking particles: implications for particle aggregation in the ocean. Journal of
Deep Sea Research, 41(3), 469-483.

Teeter, A. M., 2001. Clay-silt sediment modeling using multiple grain classes: Part I:
Settling and deposition. In: Coastal and Estuarine Fine Sediment Processes, W. H.
McAnally and A. J. Mehta eds., Elsevier, Amsterdam, 157-170.

Van Leussen, W., 1994. Estuarine macroflocs and their role in fine-grained sediment
transport. PhD dissertation, University of Utrecht, The Netherlands.

Winterwerp, J. C., 1998. A simple model for turbulence induced flocculation of cohesive
sediment. Journal of Hydraulic Research, 36(3), 309-326.

Wolanski, E., Gibbs, R., Ridd, P., Mehta A., 1992. Settling of ocean-dumped dredged
material, Townsville, Australia. Estuarine, Coastal and Shelf Science, 35, 473-
489.


















BIOGRAPHICAL SKETCH

Vladimir Paramygin was born the first son of Tatyana and Alexander Paramygin

in 1979 in Barnaul, Russia. In 1996 he graduated from Darby High School in Darby,

Montana, and High School 69 in Barnaul, and entered the Department of Mathematics at

the Altai State University in Barnaul. During his studies there he found his interests in

various areas of applied mathematics and computer science. In spring of 2000 he received

his bachelor's degree in applied mathematics and was admitted to the Graduate School of

the University of Florida, to continue his academic work in the Coastal and

Oceanographic Engineering Program of the Department of Civil and Coastal

Engineering.




University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs