Citation
Loxahatchee River, Florida central embayment

Material Information

Title:
Loxahatchee River, Florida central embayment sediment budget and trap evaluations
Series Title:
UFLCOEL-2001008
Creator:
Ganju, Neil K
Place of Publication:
Gainesville FL
Publisher:
Coastal and Oceanographic Engineering Program, Dept. of Civil and Coastal Engineering, University of Florida
Publication Date:
Language:
English
Physical Description:
38 leaves : ill., maps ; 28 cm.

Subjects

Subjects / Keywords:
Suspended sediments -- Florida -- Loxahatchee River ( lcsh )
Sediment transport -- Florida -- Loxahatchee River ( lcsh )
Sedimentation and deposition -- Florida -- Loxahatchee River ( lcsh )
Loxahatchee River (Fla.) ( lcsh )
Genre:
government publication (state, provincial, terriorial, dependent) ( marcgt )
bibliography ( marcgt )
technical report ( marcgt )
non-fiction ( marcgt )

Notes

Bibliography:
Includes bibliographical references (leaves 37-38).
Funding:
Sponsor: Jupiter Inlet District Commission.
General Note:
Cover title.
General Note:
"July, 2001."
Statement of Responsibility:
by Neil K. Ganju ... et al..

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
49575829 ( OCLC )

Full Text
UFL/COEL-2001/008

LOXAHATCHEE RIVER, FLORIDA CENTRAL EMBAYMENT: SEDIMENT BUDGET AND TRAP EVALUATIONS
By
Neil K. Ganju, Ashish J. Mehta, Leonid N. Parshukov and Ray B. Krone
Sponsor:

Jupiter Inlet District Commission 400 North Delaware Boulevard
Jupiter, FL 33458
Coastal and Oceanographic Engineering Program Department of Civil and Coastal Engineering University of Florida, Gainesville, FL 32611

July, 2001




SYNOPSIS
Sedimentation in the central bay area of the Loxahatchee River is investigated in this study. Sources of sediment include Jupiter Inlet, which provides seawater and mainly sand, and the tributaries that introduce primarily fine sediment. Using collected field and historical data as well as flow and sediment transport models, sand and fine sediment budgets are developed for the system. A sand trap west of the central embayment is evaluated for its efficiency in trapping incoming sand, while a trap at the end of the C- 18 canal is evaluated for its performance in trapping the fine sediment fraction introduced into the system. The sand shoal in the central embayment is removed in model simulations to determine what effect removal would have on the local current pattern.
Causes of sedimentation in the central bay in recent decades include natural events, dredging and maintaining Jupiter Inlet and land-use patterns in the watershed. Recommendations for management of sedimentation in the central bay are derived from an analysis of these causes are presented, including: do-nothing, sand and muck trapping, North Fork dredging and capital-dredging in the central bay. Further examination of the potential for implementation of all options, including do-nothing, will require collection of data on present-day inputs of fine sediment from the tributaries, for which available information is scarce. Accordingly, we recommend a measurement program to determine the input of sand and fine sediment applicable to the design of the proposed traps.




TABLE OF CONTENTS
SYN OPSIS .......................................................................................................................... 2
LIST OF FIGURES ............................................................................................................. 5
LIST OF TABLES .............................................................................................................. 6
1 INTRODU CTION ............................................................................................................ 7
1.1 Problem Statem ent ........................................................................................................ 7
1.2 Report Outline ............................................................................................................... 8
2 FIELD AND LABORATORY DATA ANALYSES ....................................................... 9
2.1 Field D ata ...................................................................................................................... 9
2.1.1 Bathym etry .............................................................................................................. 9
2.1.2 Tides ........................................................................................................................ 9
2.1.3 Currents ................................................................................................................... 9
2.1.4 Tributary flows ...................................................................................................... 11
2.1.5 Sedim ent loads ...................................................................................................... 11
2.1.5.1 Sand ................................................................................................................ 11
2.1.5.2 Fine sedim ent ................................................................................................ 12
2.1.6 Sedim entation ....................................................................................................... 13
2.1.7 W ind ...................................................................................................................... 16
2.2 Laboratory D ata ........................................................................................................... 17
2.2.1 Bed density ............................................................................................................ 17
2.2.2 Sedim ent erodibility .............................................................................................. 17
2.2.3 Settling velocity .................................................................................................... 17
3 FLOW AND SEDIMENT ACCUMULATION MODELING ...................................... 18
3.1 Flow M odeling ............................................................................................................ 18
3.2 Sedim ent Transport M odeling ..................................................................................... 18
3.2.1 Sand transport m odeling ....................................................................................... 18
3.2.2 Fine sedim ent transport m odeling ......................................................................... 19
4 SED IM ENT BUD GETS ................................................................................................ 20
4.1 Sand Budget ................................................................................................................ 20
4.2 Fine Sedim ent Budget ................................................................................................. 21
5 TRAP EVALUATIONS ................................................................................................ 23
5.1 Interior Sand Trap ....................................................................................................... 23
5.2 C- 18 Canal Fine Sedim ent Trap .................................................................................. 25
5.3 Sedim ent Inflow M easurem ents .................................................................................. 28




5.3.1 N eed for m easurem ents ......................................................................................... 28
5.3.2 Sand input from the Ocean ................................................................................... 29
5.3.3 Fine sedim ent input from the C-18 canal .............................................................. 29
6 CEN TRA L EM BA YM EN T SH OA L ............................................................................ 32
7 CONCLUSIONS AND RECOMMENDATIONS ........................................................ 34
7.1 Conclusions ................................................................................................................. 34
7.2 Recom m endations ....................................................................................................... 35
8 REFEREN CES ............................................................................................................... 37




LIST OF FIGURES
Figure 1. 1 The Loxahatchee River estuary, Florida ........................................... 7
Figure 2.3 Four sub-regions of the central bay area.......................................... 14
Figure 2.4 Bathymnetry of the central bay area in 1994 ...................................... 14
Figure 2.5 Bathymetry of the enetral bay area in 1996 ...................................... 15
Figure 2.6 Bathymetry of central bay area in 1998........................................... 15
Figure 2.7 Fine sediment deposition flux versus distance from the S-46 structure in the C18 canal ........................................................................................... 16
Figure 4.1 Annual sand budget for the Loxahatchee River (in the absence of the proposed trap) ............................................................................................ 20
Figure 4.2 Annual fine sediment budget for the Loxahatchee River....................... 22
Figure 5.1 Portion of model grid showing interior sand trap. Each grid is 60 m x 60 in.. 23 Figure 5.2 Predicted sand budget with proposed sand trap at the end of the first year after dredging .......................................................................................... 24
Figure 5.3 Interior sand trap bottom elevation (measured from trap initial bottom) and mass trapped versus time ....................................................................... 25
Figure 5.4 Portion of model grid showing fine sediment trap. Each grid is 60 m x 60m. 26 Figure 5.5 Initial trapping efficiency of the fine sediment trap under median C- 18 canal discharge (and concentration) conditions ..................................................... 27
Figure 5.6 Fine sediment trap bottom elevation (measured from trap initial bottom) and mass trapped versus time ....................................................................... 28
Figure 5.7 Schematic design of sand test pit and fine sediment flux measuring tower.... 30 Figure 5.8 Locations of test towers and pit.................................................... 30
Figure 6.1 Dredging in the sand shoal area ................................................... 32
Figure 6.2 Simulated velocity measurement locations for Table 6.1 ...................... 33




LIST OF TABLES

Table 2.1 Measured water discharges in the central bay area, 02/22/01 1.................. 10
Table 2.2 Volumetric changes in the central bay area sub-regions ........................ 13
Table 6.1 Comparison of current velocities in the central embayment, with and without sand shoal......................................................................................... 33




1 INTRODUCTION

1.1 Problem Statement
Sedimentation in the central embayment of the Loxahatchee River (Figure 1.1) has been documented previously (Sonntag and McPherson, 1984, Antonini, et al., 1998). Jupiter Inlet, which connects the river with the Atlantic Ocean, introduces seawater and mainly coarse sediment (sand) into the central embayment, while three main tributaries provide freshwater and fine sediment.

Figure 1.1 The Loxahatchee River estuary, Florida.

Sedimentation in the central embayment poses a navigation hazard as well as threatens native aquatic communities (Ganju, 2001). The objective of this study is to develop a sediment budget for the Loxahatchee River, which will enable managers to determine what areas are immediately threatened by sedimentation, and where dredging is most needed. In addition, proposed sediment traps are evaluated for their ability to capture incoming sediment before it can accumulate in sensitive areas (e.g., navigation channels, seagrass communities). An additional investigation is carried out to determine the effect of removing the central embayment sand shoal. Specific attention is paid to the effect on current velocities in the central embayment.




1.2 Report Outline
The field and laboratory analyses are described in Section 2, followed by the flow and sediment transport modeling in Section 3. The developed sediment budgets are presented in Section 4, and trap evaluations for both sand and fines follow in Section 5. Section 6 addresses simulations concerning the alteration of flow patterns due to central embayment dredging, and Section 7 contains study conclusions and recommendations.




2 FIELD AND LABORATORY DATA ANALYSES

2.1 Field Data
Field data were obtained through excursions to the Loxahatchee River, as well as by compiling the available literature. Data relevant to sediment budget and trap evaluations are referenced here. Detailed reports on the sedimentological evaluation of the central embayment will be issued separately (Jaeger and Hart, 2001; Faas, 2001).
2.1.1 Bathymetry
Surveys supplied by the Jupiter Inlet District (JID) were used to characterize the bathymetry of the central embayment. The average depth in the central embayment is 1 m, with a maximum depths of 5 m in channel sections. Details of the bathymetry used are found in Ganju (2001).
2.1.2 Tides
Three tide gauges were deployed, and water surface elevations were measured over one month's time (09/13/00-10/13/00). For gauge locations and an analysis of the records; see Ganju (2001).
2.1.3 Currents
An acoustic Doppler current profiler (ADCP) was used to estimate the discharge through multiple flow transects in the estuary. Ganju (2001) refers to the discharge data obtained on 08/30/00. Additional data were obtained on 02/22/01 at flow cross-sections A, B and C shown in Figure 1.1. Table 2.1 contains the discharges and times at which they were measured. Figure 2.2 shows the measured water level variation during the measurement period and Figure 2.2 shows plots of discharges based on the data in Table
2.1.




Table 2.1 Measured water discharges in the central bay area, 02/22/01
Cross-section A Cross-section B Cross-section C
Discharge Time Discharge Time Discharge Time
14 13:07 92 13:39 106 13:57
20 14:04 58 14:23 109 14:41
15 14:54 40 15:12 46 15:30
3 15:37 4 16:09 -49 16:23
-35 17:45 -58 18:00 -141 18:23
-45 18:31 -81 18:46 -157 19:04
-42 19:13 -100 19.29 -163 19:48
-39 19:63 -97 20:13 -139 20:32
-27 20:42 -61 21:04 -30 21:24
10 21:35 32 21:54 27 22:10

-0.4'~
360

400 450 500 550 600 650 700 750
Time (h)

Figure 2.1 Water level variation (relative to NGVD) with time in the central bay, 02/22/01.




2.1.4 Tributary flows
Tributary flow data were compiled from U.S. Geological Survey (USGS) records for the three major tributaries, and cumulative flow-frequency distribution curves were constructed (Ganju, 2001).
2.1.5 Sediment loads
2.1.5.1 Sand
Annual net sand load into Jupiter Inlet was estimated by Harris (1991). Of the estimated 38,220 m3 of sand which enters the inlet on a net annual basis, 60% is captured by the JID sand trap, while the remainder is distributed between the Intracoastal Waterway (ICWW) and the central embayment. The ICWW and the three forks contribute negligible sand loads to the central embayment.

/-1

50
0
-50
-100

Figure 2.2 Meas central bay area

400 450 500 550
Time (h)
;ured variation of discharge with on 02/22/01.

S00 650 700 750
time at three cross-sections in the

-150 [
-200
350




2.1.5.2 Fine sediment
Sediment rating curves were constructed using historical suspended sediment data for the three forks. The data for the North Fork were insufficient to represent the influx of fine sediment, since the data collection location was near the central embayment, rather than further upstream (Ganju, 2001). These rating curves provide the suspended sediment concentration as a function of freshwater discharge. The concentration (kg/m3) can be multiplied by the discharge (m3/s) to yield sediment load (kg/s).
Mud accumulation coupled with low flows in the North Fork suggest the possibility of a turbidity current which can bring fine sediment into the North Fork from the other central embayment. Lin and Mehta (1997) provide a method by which to calculate the mass transported into such a canal-like tributary. First the steady-state rate of sediment influx, F, is calculated as
= Gs)J (2.1)
where a' is an empirical proportionality coefficient, g is the acceleration due to gravity, H is the mean water depth at the entrance to the tributary, Pw is the density of the water, G, (= Ps/Pw) is the sediment specific gravity, where Ps is the sediment granular density, and C is the mean suspended sediment concentration at the canal entrance. The mass of sediment deposited due to turbidity current over one tidal period, MT, can be calculated as
FTBH
2 (2.2)
where T is the semi-diurnal tidal period (44,712 s), and B is the tributary width. Sediment is also introduced due to tidal inflow, and the mass deposited due to this effect, Mc, is calculated as
Mc = 2(paoLBC (2.3)
where (p is the fraction of incoming sediment which deposits, and a0 is the tidal amplitude at the canal entrance. Therefore the total mass introduced and deposited in the canal is MT+Mc.




Using a = 0.02, H=1.54 m, ps=2,403 kg/m3 (corresponding to 15% organic content; Ganju, 2001), pw=l,015 kg/m3, C=0.0173 kg/m3 (average influx concentration at median flow from Northwest Fork and C-18 canal), and B=300 m, the mass deposited due to turbidity current is 0.061 Mkg annually. Using a (p = 0.025 (canal traps 2.5% of incoming sediment due to tidal current), and ao=0.43 m (from tide records; Ganju, 2001), Mc-0.079 Mkg annually. The total mass introduced and deposited annually is then 0.14 Mkg.
2.1.6 Sedimentation
Estimates of sedimentation are given by Sonntag and McPherson (1984) for the central embayment as well as the south shore access canals. Sedimentation within the central embayment is due mainly to sand, since the circulation and flushing patterns tend to preclude the deposition of fine sediment. In contrast the lower energy access canal locations allow fine sediment to deposit, resulting in higher deposition rates relative to the central embayment.
We used available surveys (1994, 1996 and 1998) of the central bay area to determine net changes within four, conveniently delineated sub-areas of the central bay area (Figure 2.3). Bathymetry derived from these surveys is shown in Figures 2.4, 2.5 and 2.6. Region 1 shows consistent erosion and 3 shows accretion. Regions 2 and 4 show both erosion and accretion. Overall, the entire central bay area is shown to have eroded between 1994 and 1998, which we believe is inconsistent with anecdotal evidence, and points to the need for a tight survey control and dense bathymetric coverage within the region.
Table 2.2 Volumetric changes in the central bay area sub-regions.
Period Region 1 Region 2 Region 3 Region 4 Overall
1994-1996 -10500 -16800 7900 45800 26400
1996-1998 -1900 15200 4200 -16600 900
1994-1998 -12400 -1600 12100 29200 27300
Note: A negative value implies deposition.




Figure 2.3 Four sub-regions of the central bay area.

Figure 2.4 Bathymetry of the central bay area in 1994.




Figure 2.5 Bathymetry of the central bay area in 1996.

Figure 2.6 Bathymetry of central bay area in 1998.




Estimates of sedimentation in the C-18 canal for the present study (Ganju, 2001) were made by driving a pole through the bed at selected locations until a hard substrate was reached. This substrate was typically to be within the range of the original dredged bottom from 1957, when the C-18 canal was created. The subsequent 42-year accumulation in the C-18 was found to be 0.92 m on average, yielding a 2.2 cm/yr accumulation rate. The deposits in the C-18 were mainly sand/mud mixtures, with sand as the dominant fraction (14% on average). A few sampling sites in the C-18 were primarily mud however, notably before the entrance to the Southwest Fork. At this location, a 2 m-deep mud deposit of 32% sand content was collected.
The poling depths in Ganju (2001) were corrected for percent mud, to identify the mud deposition rate. The mud deposition rate is shown in Figure 2.7, as a function of distance from the S-46 structure. This rate increases greatly just before the confluence of the canal with the Southwest Fork, and then drops rapidly.
(2
0
'E 8
X
~LA
*
61
( 4
LL
2
0 - - - - - -
0 200 400 600 800 1000 1200 1400 1600 1800 2000 Distance from S-46 (m) t
Confluence of C-18 canal
and Southwes Fork
Figure 2.7 Fine sediment deposition flux versus distance from the S-46 structure in the C18 canal.
2.1.7 Wind
The tide record obtained by gauge UFG1 (Ganju, 2001) contained a trend in midtide elevation which seemingly corresponded to the mid-tide trend at Mayport, over 300 km to the north. Wind records from offshore buoys (37 and 222 km east of Cape




Canaveral) were obtained to determine if farfield wind-induced setup was responsible for the trend. A positive correlation was indicated between mid-tide elevation and wind speed, suggesting that onshore winds can raise the water surface elevation in the estuary. The linear trendline (Ganju, 2001) indicates a 0.33 m increase in mid-tide elevation for a
1 m/s increase in wind speed.
2.2 Laboratory Data
2.2.1 Bed density
Bed density properties were determined for the fine sediment samples collected (Ganju, 2001).
2.2.2 Sediment erodibility
Erodibility of the fine sediment bed was determined using the PES (Particle Erosion Simulator) device. The two measures of erodibility (shear strength and rate constant) were shown to be correlated to bed density (Ganju, 2001).
2.2.3 Settling velocity
The settling velocity of the fine sediment was determined using a settling column. The procedure is described in Ganju (2001). These data were used in the procedure for developing a fine sediment budget for the central bay area.




3 FLOW AND SEDIMENT ACCUMULATION MODELING

3.1 Flow Modeling
A flow model was developed and calibrated to model the hydrodynamics of the estuary. The model simulates horizontal, vertically averaged, water velocities and water surface elevations. The flow model is described in Ganju (2001), Chapters 2 and 4. Chapter 2 describes the model formulation, and Chapter 4 contains the procedure for calibration and validation.
3.2 Sediment Transport Modeling
Ganju (2001) contains the formulation of the sediment transport model, although some components have been modified depending on sediment type, i.e., sand or fines. Scenarios for sand and fines are discussed separately in the following sections.
3.2.1 Sand transport modeling
Sand input was accounted for at Jupiter Inlet. A unit stream power function formulated by Yang (1979) provides the total sand load as a function of water velocity. This function was chosen due to its accuracy when applied by Yang to over 1200 data sets, where the predicted load was 3% greater than the measured load, on average. The total load, Cest, is calculated by
log(C,,,) = 5.165-0.1531ogwd -0.2971og'u* + (1.78-0.361ogWd -0.48og--log u (s
v w v w ) ws (3. 1)
where W, is the settling velocity, d is the median sand diameter, v is the kinematic viscosity of water, u. is the shear velocity, U is the flow velocity, and S is the energy slope. The settling velocity of beach sand (for d > 0.135 mm, ps=2,650 kg/m3) can be calculated via Hallermeier (1981)
- 6v (3.2)
With known flow velocity U (obtained from the flow model), Equation 2.12 (Ganju, 2001) can be used to calculate the bottom shear stress Tb. The shear velocity is then




U, = b
(3.3)
where pw is the fluid (water) density. The energy slope can be calculated via S "Tb
PwgH (3.4)
where H is the water depth.
Equation 3.1 accounts for bedload transport (grains "rolling" on the bed) and suspended load transport (grains suspended in the flow), and yields the concentration of sand entering the given boundary. In the sediment transport model, this equation was applied at Jupiter Inlet. It was calibrated so that the annual net sand load (38,220 in3, or 65.5 Mkg) into the inlet was identical to the estimated value by Harris (1991).
The sand trap located immediately west of the inlet, maintained by the JID, captures approximately 60% of this incoming load (Harris, 1991). The JID trap (Figure 1) was incorporated into the model, and a sediment deposition function (Equation 2.14; Ganju, 2001) was used and calibrated to match this annual net trapping rate (22,940 m3, or 39.3 Mkg).
3.2.2 Fine sediment transport modeling
Fine sediment transport was modeled using relations developed for settling velocity, sediment density, and bed erodibility, based on collected fine samples. The modeling of fine sediment transport is given in Ganju (2001), Chapters 2 and 4, and is not repeated here.




4 SEDIMENT BUDGETS

4.1 Sand Budget
A sand budget based on the work of Harris (1991) was formulated for the estuary, shown in Figure 4.1. Of the 65.5 Mkg entering the river, 39.3 Mkg are captured by the JID trap. Of the remaining 26.2 Mkg, 17.0 Mkg are transferred to the south and north arms of the ICWW. The remaining 9.2 Mkg are scattered throughout the central embayment.
To calculate a deposition thickness, mass of sand can be converted into the corresponding volume using the relation:
m
V= M
p9(1-n) (4.1)
where V is the sand volume, m is the sand mass, ps is the sand granular density, and n is the porosity. For these calculations, ps=2,650 kg/m3, and n=0.4. Converting volume of' sand deposited into deposition thickness can be accomplished via D V m
DR =--M
A p,(1-n)A (4.2)
where DR is the deposition rate, and A is the area over which sand is deposited.
Proposed f.,
-- "----- k .-.sand trap#///
rT'Z71. Mkg/y .5 Mkg/y
9.2 Mkg/y Accegs.9k
Canals t JID track
or 3.25 mm/y
depotied
Figure 4.1 Annual sand budget for the Loxahatchee River (in the absence of the proposed trap).




4.2 Fine Sediment Budget
Fine sediment load is considered in the three main tributaries. Since the outflow in the North Fork is typically low, sediment outflow (product of sediment concentration and outflow rate) is considered to be minor. To determine sediment loads from the C- 18 canal and Northwest Fork, sediment rating curves (Ganju, 2001) for the two tributaries can be used to estimate suspended sediment concentration as a function of tributary discharge. Multiplying this concentration value by the tributary discharge yields sediment load.
Ten-year flow records (03/01/81-1/18/91) for the C-18 canal and the Northwest Fork were used to estimate annual sediment loads, by multiplying the flow discharge by the sediment concentration (provided by the respective sediment rating curves). The rating curve for the Northwest Fork, however, was found to underestimate concentrations; for example, at a maximum discharge of 61 m 3/S, the rating curve predicted a concentration of 0.036 kg/in3, which is unrealistically low. Data from feeder tributaries to the Northwest Fork indicate maximum concentrations of over 0.25 kg/in3 during storm flows (Sonntag and McPherson, 1984). The main reason for this inaccuracy is the lack of comprehensive suspended sediment concentration data in the Northwest Fork. In any event, for the present purpose the coefficient a, (Section 3.6, Ganju, 2001) was re-calibrated (from 0.012 to 0.024) to represent the concentration conditions more accurately. Following this calculation, the sediment flux boundary conditions were established for the Northwest Fork and the Southwest Fork. Load to the North Fork was estimated above in Section 2.1.5.2, as 0. 14 Mkg per year.
The rates of accumulation of fine sediment in the Northwest Fork and the Southwest Fork we re estimated from pushcore and vibracore data (Jaeger and Hart, 2001). For each fork, when that rate of accumulation is subtracted from the rate of sediment transport in the fork, the difference equals the rate of sediment that reaches the central embayment from the fork. The fine sediment budget calculated based on this procedure is shown in Figure 4.2.
The deposit volume was calculated from




m
V-
Pd (4.3)
where m is the mass deposited (obtained from the model), and Pd is the sediment dry density. The thickness can then be calculated via Equation 4.2.

1.27 Mkg deposited Unifonn 7.9 nun thickness

Figure 4.2 Annual fine sediment budget for the Loxahatchee River.




5 TRAP EVALUATIONS
5.1 Interior Sand Trap
The interior sand trap proposed by Mehta et al. (1992) was revisited in this study. The trap was simulated in the flow and sediment transport models by deepening the estuary bottom east of the Florida East Coast Railroad (FECRR) bridge by 2.5 m. The simulated trap was 180 m wide, 60 m long, and 2.5 m deeper than the surrounding bed (Figure 5.1). The trap was 30 m wider than the original trap (due to model cell size), while its length was unchanged.

Central embayment

- Interior sand trap
- 1 Z

Figure 5.1 Portion of model grid showing interior sand trap. Each grid is 60 m x 60 m.
The calibrated sand deposition function (Equation 2.14 in Ganju, 2001) used for the JID trap was applied over this trap as well, while using the calibrated total load function (Equation 3.1) to specify input sand load.
The simulation showed that of the 9.2 Mkg that originally entered the central embayment (Figure 4.1), 7.82 Mkg were deposited on an annual basis, which indicates an 85% reduction in sand load to the central embayment. The remaining 1.38 Mkg deposits




in the central embayment, which yields a spatial mean (annual) deposition thickness of
0.5 mm (Figure 5.2).
7.82 M kg p
roposed ,/and trap
J I/
1 ~~1.38 ~~~'~
I
1.38 M~g/y -,
or 0.5 mmly L'a '
deposited A
Figure 5.2 Predicted sand budget with proposed sand trap at the end of the first year after dredging.
It should be noted that as the trap fills, the flow velocity over it increases due to the decreasing flow cross-section. Therefore the trapping rate also decreases. Vicente (1992) provides a method to calculate shoaling rates in a trap (Ganju, 2001). Using the initial (9.20-1.38 =) 7.82 Mkg/yr (0.46 m/yr) sand deposition rate as the rate over the first year, Equation 6.1 (Ganju, 2001) can be developed to estimate long term shoaling rates. For a given one-year deposition rate, there exists one specific "K" value (Equation 6.1) which satisfies this short-term rate. For the calculated 0.46 m/yr short-term rate, the Kvalue is 0.2. As the trap fills, the deposition rate in the trap decreases, and the mass trapped decreases accordingly. Equation 6.1 can also be used to determine the time necessary to fill the trap to 90% of its original depth, i.e., a practically full trap. For this trap and K-value, the trap would fill to 2.25 m (90%) in 12 yr.
The variation of trap depth and mass trapped with time are shown in Figure 5.3. The mass trapped follows an exponentially decreasing trend, the inverse trend of the increase in bottom elevation.




9000000 8000000 7000000

6000000 5000000 4000000 C
,Q
3000000 ~

-*-Trap bottom elevation SMass trapped

0.5 2000000
{1000000
0-l I I
0 5 10 15 20 25 30
Time (yr)
Figure 5.3 Interior sand trap bottom elevation (measured from trap initial bottom) and mass trapped versus time.
5.2 C-18 Canal Fine Sediment Trap Sedimentation at the confluence of the C-18 canal and Southwest Fork could possibly be contained by trapping the sediment before its arrival there. Sedimentation in the access canals on the south shore could also be contained by removing the source by entrapment.
A trap was included in the C-18 canal, at the entrance to the Southwest Fork. A similar trap in Ganju (2001) was 180 m long; however, here a more realistic trap length of 60 m was selected, with a width of 60 m, and a depth of 2.5 m (Figure 5.4).




Freshwater flow with input sediment load
C-18 canal
4- Fine sediment trap
Central embayment
Figure 5.4 Portion of model grid showing fine sediment trap. Each grid is 60 m x 60m.
This trap is designed to capture fine sediment released by the S-46 structure down the C-18 canal, before it reaches the Southwest Fork. This site was chosen because pushcore data suggested that this is the zone of high mud deposition (Figure 2.1).
Simulations were performed with 15% organic content, along with the median C18 canal discharge (1.3 m3/s). Under these conditions, the sediment removal ratio (Ganju, 2001) would be 53%, indicating that approximately half of the incoming sediment to the trap is removed by deposition in the trap, and the remainder passes over the trap (Figure
5.5).
The average simulated trapped load over one year was 0.154 Mkg, which corresponds to a deposit thickness of 0.13 m/yr in the trap (Equations 4.2 and 4.3). This is the initial deposition rate, analogous to the same quantity calculated for the interior sand trap. To predict the long-term performance of the trap, the procedure by Vicente (1992) was used again, resulting in a K-value of 0.054. The short-term rate (0.13 mnyr) corresponds to the depth increase (shoaling) over the first year. However, as the trap fills it will be unable to capture less and less sediment, because the rate of capture is




proportional to the difference between the actual depth and the final depth. This difference continues to decrease with the passage of time. Accordingly, it can be shown that the time required to fill the trap to 90% would be 43 yr. Figure 5.6 shows the mass trapped and trap elevation (from bottom) as a function of time. The mass trapped follows an exponentially decreasing trend, and the bottom elevation increases with an inverse trend. This calculation assumes that the filling sediment will be mud only. However, in reality sand is likely to arrive as well as in the past, although perhaps at as a smaller fraction of the incoming material. The trap efficiency will therefore be lower, so that filling may occur is about a decade.
0.154Mkg to trap
Southwest Fork
C-18 a7,
Figure 5.5 Initial trapping efficiency of the fine sediment trap under median C- 18 canal discharge (and concentration) conditions.
For practical applications, the trap cannot be as wide as the canal, and therefore a more realistic width of 30 m is suggested for the actual trap design. In addition, the simulated trap length of 60 m is excessive, a realistic length of 30 m is also suggested, with the modeled trap depth of 2.5 m. it appears that for convenience of dredging an alternative site for the trap could be just beyond the very end of the canal, south if the Indiantown Road bridge, where a delta has formed. Trap efficiency there is likely to be similar. In either case, the exact location and dimensions of the trap need further investigation.




180000

nano- 160000
140000
120000
Cun
>. 100000
(1 m Trap bottom elevation
E M ~Mass trapped
0 80000 0
0 1 C
.0
Cu 60000 .
0.5 40000
7 20000
0 0
0 5 10 15 20 25 30 35 40 45 50 Time (yr)
Figure 5.6 Fine sediment trap bottom elevation (measured from trap initial bottom) and mass trapped versus time.
5.3 Sediment Inflow Measurements
5.3.1 Need for measurements
Our study has revealed that while the rate of input of sand coming into the bay from the ocean end does not seem to have increased over the past one or two decades, over the same period the rate of input of mud to the Northwest Fork of the river close to the central bay has increased. The specific reason for this increase is not presently clear; however it appears that it is linked in some way to land-use practices in the watershed and a reduction in minimum freshwater flow in the fork, which in turn has impacted the ecology of the river and its floodplain. Since restoration of the river regime to decrease sediment input and increase the ability to flush out sediment present may take years, it appears that any future strategy for managing sediment in the central bay will involve bottom sediment dredging. The main roadblock against evolving a coherent strategy in that regard is that present day sediment inputs via the principal tributaries, the Northwest




Fork and the C-18 Canal/Southwest Fork, are not available, and past data are sparse, hence unreliable for future predictions.
We propose to monitor rates of sediment input to the central bay for a period of one year with the objective to improve upon our cursory understanding of the sediment budget for the central bay, and to ascertain whether sediment input is of such magnitude and frequency as to require the construction of traps to intercept the arrival of sediment into the central bay. The plan of study will be as described next.
5.3.2 Sand input from the Ocean
Immediately westward of the Florida East Coast Railroad Bridge, which is the eastern flow boundary of the central bay, we believe sand is transported primarily as bed load. Accordingly, we propose to measure sand deposition in a 3 m wide by 3 m long and 2.5 m deep test pit shown schematically in Figure 5.7. A series of (wholly submerged) poles will be embedded in and in the vicinity of the trap, and the deposit thickness will be obtained by measuring the height of the poles above the sand bed on a monthly basis with the help of a diver and underwater photography. A self-recording current meter will be installed in the vicinity as well. From these simple measurements we hope to resolve two issues: 1) the annual rate of sand input, and 2) whether that input is quasi-steady or episodic. The tentative site for the test pit is shown in Figure 5.8.
5.3.3 Fine sediment input from the C-18 canal
The tower assembly schernatized in Figure 5.7 will consist of a vertical array of five Infrared backscatter sensors to measure the suspended sediment load, a pressure sensor to measure the water level and a current meter. During selected times we will also carry out flow discharge measurements at the sites by using an Acoustic Doppler Current profiler across the respective flow section of the waterway. In addition, suspended sediment




U.1, ,A
4(111 111T

..... ....

1 ~'~t 4)14

Figure 5.7 Schematic design of sand test pit and fine sediment flux measuring tower.

4"n'- access canals
1 Penn
'z 1 Turner Ouay
3 Pompano
k ,4 Dolphin
S-46 control structure '; Marlill
Figure 5.8 Locations of test towers and pit.

samples will be collected to calibrate the infrared sensors. Tentative sites for the towers are shown in Figure 2. The ultimate objective of this exercise will be to develop: 1) stagedischarge relationships for the Northwest Fork and the C-18 Canal, and 2) corresponding

Mal

d"J'-h+,++




rating curves relating the discharge to suspended sediment concentration. From these data we will be able to: 1) calculate the inflow of sediment to the Central Bay over the data collection period, and 2) to examine the correlation between rainfall, runoff and sediment input via the two tributaries.




6 CENTRAL EMBAYMENT SHOAL

The sand shoal located in the central embayment occupies areas known to harbor seagrass colonies. Simulations were performed in order to determine what changes in current velocities would be observed if the shoal were dredged to improve flow circulation and flushing. The bathymetry of the central embayment was modified by deepening shoal area to the surrounding depth (Figure 6. 1). The maximum increase in depth was about I m. The flow model was then run to determine the new current velocities. It was found that currents are more concentrated in the main navigation channel with the shoal in place, and that removal of the shoal results in more uniform currents. Figure 6.2 shows locations where velocity (from the model) was recorded for point comparisons of velocity change. Table 6.1 compares the velocities with and without the shoal in place. The results show that velocities are reduced overall, though velocity increased at some locations, since the removal of the shoal allowed flow to reach those areas more readily. The highest decrease (-24%) was recorded at location I as a result of the removal of the flow confining effect of the shoal.

Shoal dredged to -1.1lM Figure 6.1 Dredging in the sand shoal area.




1 2 3 4 5 6

Figure 6.2 Simulated velocity measurement locations for Table 6.1.

Table 6.1 Comparison of current velocities in the central embayment, with and without sand shoal.
Location 1 2 3 4 5 6
Velocity with 0.49 0.52 0.56 0.31 0.33 0.39
shoal (m/s)
Velocity without 0.37 0.48 0.56 0.29 0.33 0.43
shoal (m/s)
% change -24 -8 0 -6 0 10




7 CONCLUSIONS AND RECOMMENDATIONS

7.1 Conclusions
Based on our analysis to date we have come to the tentative conclusion that changes presently occurring in the area of the central bay in terms of sedimentation are due to four principal causes as follows:
1. Natural Effects: All rives and especially estuaries are dynamic and change with
time due to ongoing sedimentation, sea level changes, tectonic processes and
changes in weather patterns. The Loxahatchee is no exception.
2. Jupiter Inlet Dredging: The widening, deepening and training of Jupiter Inlet has
had a profound effect on the intrusion of seawater and sand into the estuary.
3. Construction of C-1 8 Canal: Diversion of water from the Northwest Fork into the
C-18 canal, which in turn delivers sediment-laden water to the central bay, in a sense amounts to taking fresh water from the river, adding sediment to it and pumping it in to the estuary. This has led to a change in the discharge hydrograph of the Northwest Fork with associated salinity intrusion, and introduction of muck into the central bay. While on the one hand the C-18 canal project has successfully lowered flood stages in the river, it must be noted that, in general, the wetland-dominated flood plain of a river acts, at times the river is in spate, as a filter for the heavy sediment load by allowing that sediment to deposit over the plain. In the Loxahatchee, the C-18 canal acts as a shunt bypassing part of this filtering process. Reestablishment of minimum flow in the Northwest Fork, which has been assessed on the basis of reducing saltwater intrusion, may have no
tangible effect on the role of the C- 18 canal in importing sediment.
4. Land-Use: Changes in the land-use patterns in the watershed coupled with salinity
intrusion due to construction of Jupiter Inlet and the C- 18 canal has meant ecological changes, e.g., loss of sediment stabilizing flora, with the result that the




Northwest Fork has become more muddy in recent decades. As a result more mud has been arriving in the area of the access canals in the central bay and in the
North Fork.
7.2 Recommendations
The above observations make it evident that the sedimentation issue in the central bay is linked to the behavior of the entire Loxahatchee River and watershed. Keeping that in mind, we offer the following observations regarding options and actions:
1 Do-nothing: Under this "wait and see" option, no action is necessary. However,
we recommend biennial surveys of the North Fork, the Northwest Fork and the Southwest Fork within the JID boundaries. The C-18 canal survey needs to be up to the S-46 control structure. Survey lines should be 100 ft apart. This survey density is needed to determine bottom changes due to sedimentation (and erosion)
in the estuary, which are presently difficult to calculate.
2. Sand Trapping: Further evaluate the need for a sand trap adjacent to the FECRR
bridge, as recommended by Mehta et al. (1992). A change in that recommendation is that in our opinion a trap on the west side of the bridge, as opposed to the east side, will be more appropriate for capturing sand arriving into the central bay area. We did not recommend that location previously (in 1992) because of then likely difficulty in obtaining a permit for dredging west of the bridge, which is an aquatic preserve. Our analysis confirms the finding of that study that a trap at that location can be potentially effective in reducing the influx of sand into the Central Bay. In order to determine the frequency of maintenance dredging we recommend the construction of a test pit at the site and evaluate its performance over one year. The results will allow us to determine the nature of sand transport in that area, which is needed to model the infilling role of the fullsized trap.




3. Muck Trapping : We recommend that a trap for capturing muck at the end of the
C-i18 canal, south of the Loxahatchee River Road bridge be evaluated. Our analysis suggests that such a trap can reduce the rate of accumulation of muck in the vicinity of the five access canals, portions of which should be dredged at the same time as the trap, both to reduce accumulated muck and improve navigation, if desired. As to predicting the frequency of maintenance dredging of the trap, past record of accumulation of muck in the area is helpful but not unequivocal, because of the influx of sand possibly connected to the changes in the waterway that took place in years immediately following the construction of the C-i18 canal in 1957. We therefore recommend that the influx of suspended sediment be monitored downstream of the S-46 structure for a period of one year, in order to determine the rate at which muck arrives. This exercise must be coupled with surveys of the bottom using push cores at selected spots, both at the beginning and at the end of the study period, in order to determine the role of accumulation.
At each of these spots a stratigraphic horizon must be laid out at the outset to
identify accumulation rate above the horizon.
4. North Fork Dredging: We recommend that the North Fork, which was dredged in
1980, be considered for dredging again. The site of the dredging should meet the dual requirements of removing accumulated muck and navigation. The area and the volume to be dredged can be decided on the basis of core data collected as
part of the present study (Jaeger and Hart, 2001).
5. Capital Dredging in Central Bay: Although the sand bar that has developed close
to the eastern end of the central bay is a popular attraction for weekend recreation, its removal must be assessed as a future potential option to reduce sand quantities in the central bay, should a condition occur at some future date whereby flow circulation on the bay is threatened by "sand choking". Presently it is not clear if such an action would help or harm the seagrass community. We recommend that
the role of the sand bar in that regard be studied and quantified.




8 REFERENCES
Antonini, G. A., Box, P. W., Fann, D. A., and Grella, M. J., 1998. Waterway evaluation
and management scheme for the south shore and central embayment of the Loxahatchee River, Florida. Technical Paper TP-92, Florida Sea Grant College,
Gainesville, 47p.
Faas, R. W., 2001. Environmental sedimentology of the lower Loxahatchee River estuary
and Jupiter Inlet revisited. Report under preparation.
Ganju, N. K., 2001. Trapping organic-rich fine sediment in an estuary. M.S. Thesis,
University of Florida, Gainesville, 120p.
Hallermeier, R. J., 1981. A profile zonation for seasonal sand beaches from wave climate.
Coastal Engineering, 4, 253-277.
Harris, P. S., 1991. The influence of seasonal variation in the longshore sediment
transport with special applications to the erosion of the downdrift beach at Jupiter
Inlet, Florida. M.S. Thesis, University of Florida, Gainesville, 138p.
Jaeger, J. M., and Hart, M., 2001. Sediment accumulation within the Loxahatchee
Estuary: 4000 years ago to present. Project report, Department of Geology,
University of Florida, Gainesville (under preparation).
Lin, P. C.-P., and Mehta, A. J., 1997. A study of fine sedimentation in an elongated
laboratory basin. Journal of Coastal Research, S125, 19-30.
Mehta, A. J., Montague, C. L., Thieke, R. J., 1992. Erosion, navigation and sedimentation
imperatives at Jupiter Inlet, Florida. Report UFLICOEL/92/002, Coastal and
Oceanographic Engineering Department, University of Florida, Gainesville.
Sonntag, W. H., and McPherson, B. F., 1984. Sediment concentrations and loads in the
Loxahatchee River estuary, 1980-1982. Water-Resources Investigations Report
84-4157, U.S. Geological Survey, Tallahassee, FL, 30 p.
Vincente, C. M., 1992. Experimental dredged pit of KA-HO: analysis of shoaling rates.
Proceedings of the International Conference on the Pearl River Estuary in the Surrounding Area of Macao, Vol. 2, Civil Engineering laboratory of Macao,
Macao, paper P6.4, 11 p.




Yang, C. T., 1979. Unit stream power equations for total load. Journal of Hydrology, 40,
123-138.