HomeMy WebLinkAbout2009-01 Hilo Water Circulation Water Quality Report FINAL2 January2009 i
EXECUTIVE SUMMARY
By letter dated October 14, 2004, the County of Hawaii (County) requested
planning assistance from the U.S. Army Corps of Engineers, Honolulu District
(POH) to evaluate water circulation and apparent degraded water quality
within Hilo Bay and identify potential solutions. In response to the County’s
request, this study was initiated under the authority of Section 22 of the Water
Resources Development Act of 1974 (Public Law 93-251) as amended,
through a cost-sharing partnership between POH and the County. Technical
assistance was provided by the U.S. Army Engineer Research and
Development Center’s (ERDC) Coastal and Hydraulics Laboratory (CHL) and
Field Research Facility (FRF), and the University of Hawaii at Hilo (UHH) to
collect field data and implement numerical modeling of circulation, wave
transformation, and identify alternatives to improve water quality in Hilo Bay.
The present study by POH investigates the feasibility of modifying the Hilo
Harbor breakwater to increase water circulation within Hilo Harbor. Increased
circulation could potentially provide corresponding improvements in water
quality within the bay thereby providing a more suitable environment for
recreation and a greater aesthetic enjoyment of the area. The resulting
changes to wave energy within the harbor are also investigated to quantify
the relative effects that breakwater modification may have on navigation.
Model results and predictions for five alternative plans are documented in this
technical report. This report does not provide a specific recommendation to
address the water quality issue within Hilo Bay, but rather provides
information to be used by Federal, State and County agencies and other
stakeholders in determining an appropriate course of action.
The criteria for assessing alternative plans in this study were determined by
examining changes in waves, current circulation, water quality, and residence
time, as well as by determining areas subject to stagnant or weak circulation
or focused wave energy resulting from proposed alternatives. The initial
numerical modeling efforts concentrated on quantifying change in circulation
and wave patterns with and without the alternatives in place for a range of
forcing conditions.
All breakwater modifications considered in this report resulted in an increase
in wave energy within the harbor and navigation channel, as evidenced by
numerical wave modeling of each alternative. The increase in wave energy
within the navigation channel varies greatly between alternatives, from a
minimal increase that may be acceptable for safe navigation to a significant
increase that would likely be considered unacceptable. Overall water quality
model predictions indicated little difference in the results for any of the
proposed harbor alternatives. At some locations there were differences in
some constituent values such as particulate organic carbon. However, these
differences appear to be due to phasing in the model response to the
circulation and were relatively small and short lived. Further evaluation of
ii
these effects with input from all stakeholders will be required before initiating
any breakwater modification to improve water quality within Hilo Bay.
Cost estimates for the conceptual alternatives were also prepared by POH to
enable comparison between the various conceptual features. Other
considerations that should be evaluated in future breakwater modification
studies include the effects on the Hilo Bay shoreline, the changes to
breakwater access, and the impacts to Blonde Reef. Evaluation of these
additional impacts was not within the scope of this report.
Water quality in Hilo Harbor and Hilo Bay is dependent on several interrelated
environmental processes which include the effects of the breakwater, as
detailed in this report. Another major contributor to the water quality in Hilo
Bay is the input of pollutants and organic materials from the Hilo Bay
watershed via surface water, ground water, and storm water runoff. In order
to comprehensively evaluate the bay’s water quality and possible methods for
improvement, these sources of contaminants must also be included in an
overall watershed study that encompasses the ancient Hawaiian ahupua’a
concept of “mountain to the sea” stewardship. This approach has been
initiated and led by the Hilo Bay Watershed Advisory Group and Dr. Tracy
Wiegner at the University of Hawaii at Hilo, and should be continued with a
more detailed evaluation of breakwater modifications and their effect on water
quality included as an integral component of the overall study.
iii
HILO BAY
WATER CIRCULATION
AND
WATER QUALITY STUDY
TABLE OF CONTENTS
EXECUTIVE SUMMARY.......................................................................................i
1. INTRODUCTION...........................................................................................1
1.1. Authority ..........................................................................................................1
1.2. Study Purpose .................................................................................................1
1.3. Project Background ........................................................................................1
1.3.1. General Description...................................................................................1
1.3.2. Fresh Water Inflow to Hilo Bay..................................................................3
1.3.3. Wind and Tides..........................................................................................4
1.3.4. Waves........................................................................................................5
1.3.5. Navigation and Recreational Use..............................................................6
1.4. Previous Circulation Studies .........................................................................6
1.5. Study Method...................................................................................................8
2. HARBOR ALTERNATIVES CONSIDERED..................................................9
2.1. Existing Conditions and Additional Dredging ..............................................9
2.2. Alternative 1 .....................................................................................................9
2.3. Alternative 2 ...................................................................................................11
2.4. Alternative 3 ...................................................................................................12
2.5. Alternative 4 ...................................................................................................14
2.6. Alternative 5 ...................................................................................................15
2.7. Assumptions and Limitations of Alternatives ............................................16
3. FIELD DATA COLLECTION.......................................................................16
3.1. Waves and Water Circulation Data ..............................................................16
3.1.1. Wave Data Collection..............................................................................18
3.1.2. Current Profile Data Collection................................................................18
3.1.3. Current Drogue Data Collection ..............................................................22
3.2. Water Quality Data ........................................................................................22
4. HYDRODYNAMIC MODELING...................................................................28
4.1. Circulation Modeling .....................................................................................28
4.1.1. Model Descriptions..................................................................................28
4.1.2. Model Setup and Forcing Data................................................................29
4.1.3. Validation and Calibration........................................................................34
4.2. Wave Modeling ..............................................................................................37
4.2.1. Model Description....................................................................................37
4.2.2. Model Setup and Forcing Data................................................................38
4.2.3. Validation and Calibration........................................................................41
4.3. Combined Wave and Circulation Modeling.................................................45
4.3.1. Radiation Stress Gradients in STWAVE..................................................45
4.3.2. Application of Radiation Stress Gradients in ADCIRC ............................46
4.3.3. Radiation Stress Gradients and Water Levels in CH3D..........................46
4.4. Modeling Simulation of February – March 2007 .........................................46
4.4.1. Forcing Data............................................................................................48
4.4.2. Harbor Alternative Runs ..........................................................................48
iv
5. WATER QUALITY MODELING...................................................................54
5.1. Water Quality Model ......................................................................................54
5.1.1. Model Description....................................................................................54
5.1.2. Model Setup ............................................................................................57
5.2. Modeling Simulation of February – March 2007 .........................................59
5.2.1. Forcing Data............................................................................................59
5.2.2. Flushing Studies......................................................................................60
5.2.3. Validation to UH Water Quality Data.......................................................79
5.2.4. Harbor Alternative Runs ..........................................................................90
6. ESTIMATED COSTS AND OTHER CONSIDERATIONS.........................101
6.1. Construction Costs .....................................................................................101
6.1.1. Alternative 1...........................................................................................102
6.1.2. Alternative 2...........................................................................................102
6.1.3. Alternative 3...........................................................................................102
6.1.4. Alternative 4...........................................................................................102
6.1.5. Alternative 5...........................................................................................103
6.1.6. Cost Summary.......................................................................................103
6.2. Other Considerations ..................................................................................104
7. SUMMARY OF RESULTS AND RECOMMENDATIONS FOR FUTURE
WORK ..............................................................................................................105
7.1. Effects on Waves in Hilo Harbor ................................................................105
7.2. Effects on Hilo Bay Flushing and Water Quality ......................................106
7.3. Summary Matrix ..........................................................................................107
7.4. Recommendations for Future Work ..........................................................107
8. REFERENCES..........................................................................................109
v
LIST OF FIGURES
Figure No. Title Page No.
1-1 Vicinity Map of Hilo Bay and Hilo Harbor 2
1-2 Major fresh water sources at Hilo Bay (Wailoa and
Wailuku Rivers) 3
1-3 Wind rose for 1981 – 2004 (WIS Pacific Station 105) 4
1-4 Wave rose for 1981 – 2004 (WIS Pacific Station 105) 5
2-1 Existing conditions within Hilo Harbor 10
2-2 Alternative 1 11
2-3 Alternative 2 12
2-4 Alternative 3 13
2-5 Alternative 4 14
2-6 Alternative 5 15
3-1 ADCP Instrument Locations 17
3-2 Typical Monthly Wave Rose at ADCP 1 (March 2007) 19
3-3 Typical Current Measurement Time-Series at ADCP1 20
3-4 Typical Current Profile Snapshot at ADCP1 21
3-5 Photo of Floating Surface Drogue with GPS Antenna 23
3-6 Photo of Floating Sub-Surface Drogue with GPS Antenna 23
3-7a Drogue Tracks between 21-22 March 2007 24
3-7b Drogue Tracks between 22-23 March 2007 24
3-7c Drogue Tracks between 5-6 June 2007 25
3-7d Drogue Tracks between 6-7 June 2007 25
3-8 UHH Water Quality Monitoring Stations 26
4-1 ADCIRC Grid for the Hawaiian Islands 29
4-2 ADCIRC Grid for Hilo Bay 30
4-3 ADCIRC Grid for Hilo Harbor 30
4-4 ADECIRC Grid for Existing Breakwater Condition 31
4-5 ADCIRC Bathymetry for Typical Alternative (Alternative 3) 32
4-6 CH3D Grid Coverage (Existing Condition) 33
4-7 CH3D Grid for Typical Alternative (Alternative 3) 34
4-8 NOAA Tide Station 1617760 at Hilo Harbor 35
4-9 ADCIRC Modeled Water Level Compared to Tide Prediction 36
4-10 ADCIRC & CH3D Comparison to Tide Measurement
And Prediction 37
4-11a. STWAVE Coarse Grid 38
4-11b. STWAVE Fine Grid 39
4-12 Detailed Bathymetry of STWAVE Grid for Existing Condition 40
4-13 Detailed Bathymetry of STWAVE Grid for Typical
Alternative (Alternative 3) 40
4-14 Sheltering Angle used for Wave Transformation with WAVTRAN 42
vi
LIST OF FIGURES
(continued)
Figure No. Title Page No.
4-15 Wave Height Comparison (STWAVE vs. ADCP1) 43
4-16 Wave Period Comparison (STWAVE vs. ADCP1) 44
4-17 Wave Direction Comparison (STWAVE vs. ADCP1) 45
4-18 Meteorological Conditions during 19 February-19 March 2007 47
4-19a Typical Wave Height Increase for Alternative 1 50
4-19b Typical Wave Height Increase for Alternative 2 50
4-19c Typical Wave Height Increase for Alternative 3 51
4-19d Typical Wave Height Increase for Alternative 4 51
4-19e Typical Wave Height Increase for Alternative 5 52
4-20 Water Level Validation for 19 February-19 March 2007 Run Period 54
5-1 Hilo Bay Time Series Comparison Stations 61
5-2 Tracer Concentrations Harbor Flushing Test at Station A1 62
5-3 Tracer Concentrations Harbor Flushing Test at Station A2 62
5-4 Tracer Concentrations Harbor Flushing Test at Station A3 63
5-5 Tracer Concentrations Harbor Flushing Test at Station A4 63
5-6 Tracer Concentrations Harbor Flushing Test at Station A5 64
5-7 Tracer Concentrations Harbor Flushing Test at Station A6 64
5-8 Tracer Concentrations Harbor Flushing Test at Station A7 65
5-9 Tracer Concentrations Harbor Flushing Test at Station A8 65
5-10 Tracer Concentrations Harbor Flushing Test at Station A9 66
5-11 Tracer Concentrations Harbor Flushing Test at Station A10 66
5-12 Tracer Concentrations Harbor Flushing Test at Station A11 67
5-13 Tracer Concentrations Harbor Flushing Test at Station A12 67
5-14 Tracer Concentrations Harbor Flushing Test at Station S2 68
5-15 Tracer Concentrations Harbor Flushing Test at Station S3 68
5-16 Tracer Concentrations Harbor Flushing Test at Station S5 69
5-17 Tracer Concentrations Harbor Flushing Test at Station S6 69
5-18 Tracer Concentrations Harbor Flushing Test at Station C1 70
5-19 Tracer Concentrations Harbor Flushing Test at Station C2 70
5-20 Tracer Concentrations in Flushing Test Alt.0 (Base Case)
for Days 0, 1, 2, and 5 71
5-21 Tracer Concentrations in Flushing Test Alt.2 (multiple gaps in
outer breakwater) for Days 0, 1, 2, and 5 72
5-22 Tracer Concentrations in Flushing Test Alt.3 (multiple gaps in
outer breakwater with interior groins) for Days 0, 1, 2, and 5 73
5-23 Tracer Concentrations in Flushing Test Alt. 4 (single gap in
in breakwater) for Days 0, 1, 2, and 5 74
5-24 Tracer Concentrations in Flushing Test Alt. 5 (single gap in
breakwater with external groin) for Days 0, 1, 2 and 5 75
5-25 Harbor Mouth Flow Rate During Water Quality Modeling
Period for Alternative 0 76
vii
LIST OF FIGURES
(continued)
Figure No. Title Page No.
5-26 Harbor Mouth Flow Rate During Water Quality Modeling
Period for Alternative 2 76
5-27 Harbor Mouth Flow Rate During Water Quality Modeling
Period for Alternative 3 77
5-28 Harbor Mouth Flow Rate During Water Quality Modeling
Period for Alternative 4 77
5-29 Harbor Mouth Flow Rate During Water Quality Modeling
Period for Alternative 5 78
5-30 Salinity 80
5-31 Dissolved Oxygen 81
5-32 Temperature 82
5-33 Suspended Solids 83
5-34 Dissolved Organic Carbon 84
5-35 Particulate Organic Carbon 85
5-36 Ammonia 86
5-37 Nitrate 87
5-38 Dissolved Inorganic Phosphorus 88
5-39 Alternative Salinities 91
5-40 Alternative Dissolved Oxygen Concentrations 40
5-41 Alternative Temperatures 93
5-42 Alternative Suspended Solids 94
5-43 Alternative Dissolved Organic Carbon Concentrations 95
5-44 Alternative Particulate Organic Carbon Concentrations 96
5-45 Alternative Ammonia Concentrations 97
5-46 Alternative Nitrate Concentrations 98
5-47 Alternative Dissolved Inorganic Phosphorus Concentrations 99
viii
LIST OF TABLES
Table No. Title Page No.
3-1 ADCP Instrument Identifications and Locations 18
3-2 University of Hawaii at Hilo Sampling Events
through September 2007 27
5-1 Water Quality Model State Variables 56
5-2 Hilo State Variables 58
5-3 CEQUAL-ICM Grid Characteristics 59
5-4 Net Flows at Hilo Harbor Mouth 78
5-5 Water Quality Boundary Conditions 79
6-1 Estimated Dredging Cost for Each Alternative 104
6-2 Estimated Breakwater Cost for Each Alternative 104
6-3 Estimated Total Cost for Each Alternative 104
7-1 Summary of Alternatives 107
1
HILO BAY
WATER CIRCULATION
AND
WATER QUALITY STUDY
1. INTRODUCTION
1.1. Authority
Federal funding for this study was provided through the Section 22 Planning
Assistance to States (PAS) program. The PAS program was authorized by
the Water Resources Development Act of 1974 (Public Law 93-251). It
provides authority for cooperating with any state in preparation of
comprehensive plans for water resources develop, utilization and
conservation. Cost sharing for the PAS program is 50% federal and 50%
non-federal. PAS program is applicable to coastal zone and lake shores as
well as riverine and drainage areas. PAS studies may include collection of
new data, but only in the context of a legitimate planning study (not for large
data sets). Non-federal funding for 50% of the study cost was provided by the
County of Hawaii.
1.2. Study Purpose
Hilo Harbor appears degraded to an undefined degree which does not
provide a suitable environment for recreation and aesthetic enjoyment of the
area. The objective of any conceptual alternative considered is to promote
greater water circulation in the Hilo Harbor to improve its water quality. The
present study by the US Army Corps of Engineers (USACE), Honolulu District
investigates the feasibility of modifying the Hilo Harbor breakwater to increase
water circulation within Hilo Harbor and potentially provide corresponding
improvements in water quality within the bay. The resulting changes to wave
energy within the harbor are also investigated to relatively quantify the effects
that breakwater modification may have on navigation.
1.3. Project Background
1.3.1. General Description
Hilo Bay is located along the east (windward) coast of the island of Hawaii,
extending south from Pepe'ekeo Point, and west from Leleiwi Point
(Figure 1-1). Hilo Harbor encompasses an area approximately 3658
meters (12,000 feet) by 2134 meters (7,000 feet) in the bight of Hilo Bay.
The 3072 meter-long (10,080-foot-long) Hilo breakwater, completed in
1929, extends across the northeastern half of the harbor, constructed on
the relatively shallow Blonde Reef. The Hilo Harbor Federal navigation
2
project consists of an entrance channel that is 2,200 feet long by 440 feet
wide by 39 feet deep, a turning basin 1,800 feet long by 1,400 feet wide by
38 feet deep and the breakwater. The small boat harbor maintained by
the State of Hawaii exists at the easternmost end of the harbor known as
Radio Bay, near the root of the breakwater.
Hilo Airport
Pepe’ekeo Point
Leleiwi Point
HILO BAY
Hilo Harbor
Figure 1-1. Vicinity Map of Hilo Bay and Hilo Harbor
3
1.3.2. Fresh Water Inflow to Hilo Bay
There are two major rivers that empty into Hilo Bay, see Figure 1-2. The
larger is the Wailuku River which has a drainage area of 125 square miles.
The average annual flow of water from the Wailuku River into Hilo Bay is
one million cubic meters. The flow of the Wailuku can vary widely
depending upon rainfall with a range of 40 thousand -7 billion cubic meters
(M & E Pacific, 1980). The other tributary is the Wailoa River which
connects Hilo Bay and Waiakea Pond. The main source of flow is a large
basal compound spring, Waiakea Spring, which provides the single largest
source of groundwater into Hilo Bay (M & E Pacific, 1980). It is estimated
that 1.8 million cubic meters of groundwater enters the Bay annually in this
area (M & E Pacific, 1980).
Wailuku
River
Wailoa River and
Waiakea Pond
Figure 1-2. Major fresh water sources at Hilo Bay (Wailoa and Wailuku Rivers)
4
1.3.3. Wind and Tides
In the Hilo area, the tradewind flow is modified by the presence of Mauna
Loa and Mauna Kea. During typical east-northeast tradewind conditions,
the wind speeds off East Hawaii are relatively lighter than over the open
ocean. This area of minimum wind speed is centered at Hilo. The
temperature differential between land and sea results in the formation of a
land and sea breeze system in the Hilo vicinity, which alternately
reinforces and opposes the already weak underlying trade wind flow.
During the day the onshore sea breeze reinforces the trade winds. At
night, the offshore land breeze dominates, resulting in light southwest
winds (Sea Engineering, 1981). Figure 1-3 shows a wind rose from the
area offshore of Hilo Bay between the years 1981 – 2004.
Figure 1-3. Wind rose for 1981 – 2004 (WIS Pacific Station 105)
5
The tides in Hilo Harbor are semi-diurnal (two high and two low tides per
25-hour period) with a pronounced diurnal inequality. The total tide range,
or difference between Mean Lower Low Water (average of all lower low
water heights of each tidal day) and Mean Higher High Water (average of
all higher high water heights of each tidal day), is 0.731 meters (2.40 feet)
during the most recent tidal epoch (1983-2001).
1.3.4. Waves
Hilo Bay is directly exposed to waves approaching from the sector north
through east. Figure 1- 4 shows a wave rose from the area offshore of
Hilo Bay between the years 1981 – 2004. Both tradewind waves and
North Pacific swells may approach from this direction. Tradewind waves
may approach from the sector north through southeast, with the
predominant direction from the east-northeast. These waves are present
80 to 90 percent of the time during the summer; the frequency decreases
to 60 to 70 percent during the winter. Tradewind waves have typical
heights of 4 to 12 feet and periods of 7 to 10 seconds. Although Hilo Bay
is exposed to tradewind wave approach, the breakwater shelters Hilo
Harbor from direct approach of all but the most northerly tradewind waves
(Sea Engineering, 1981).
Figure 1-4. Wave rose for 1981 – 2004 (WIS Pacific Station 105)
6
North Pacific swell is generated by winter storms in the North Pacific and
may approach from the sector west through northeast. The most common
approach direction is from the northwest. This wave type is most frequent
from October through April. The average wave period is 14 seconds and
deepwater heights range up to 15 feet. Hilo Harbor is directly exposed to
only the North Pacific swell approaching from the north and northeast.
Total frequency of occurrence of all North Pacific swell is 75 percent;
however, it approaches from the north and northeast only 12 percent of
the time. Because of its large size and long period, though, even swells
approaching from more westerly directions may refract and have some
influence on the wave climate in the harbor (Sea Engineering, 1981).
1.3.5. Navigation and Recreational Use
Hilo Harbor is currently the primary location of commercial waterborne
traffic for the Island of Hawaii. The harbor is used by commercial vessels
(deep and shallow draft) that moor at Piers 1, 2, and 3 along the interior of
the Federal Channel, as well as recreational vessels that occupy the
small-boat harbor at Radio Bay, the mooring area at Reeds Bay and use
the launch ramps at the Wailoa River. Canoe paddling, surfing, fishing
and other water sports are also popular recreational uses of the harbor.
1.4. Previous Circulation Studies
The Public Health Service (1963) conducted a dye tracer study to determine
flushing and mixing patterns in Hilo Bay. The PHS found the following:
“The forces influencing diffusion patterns in Hilo Bay are in some part
attributable to littoral (shallow water) currents, tidal currents, currents due to
fresh water runoff, and locally generated wind-driven currents. Littoral or
alongshore currents resulting from the breaking of the distant generated sea
swell at an angle with the coastline is probably a major force. This, together
with land runoff from both surface streams and subsurface springs and seeps,
is believed to predominate in producing net advective movement and
turbulent diffusion of water masses in the area being considered. Hilo Bay is
at the intersection of a north-south and east-west coastline resulting in a
current somewhat analogous to a rip current moving seaward in a
northeasterly direction in opposition to the incoming swell.
Though tidal currents most certainly add in some way to the dispersion
process, they do not have the pronounced effect found in estuaries. Because
of the high porosity of the breakwater, tidal flows undoubtedly pass through
them quite readily and their effect, therefore, on tidal current patterns within
the harbor is probably very minimal.”
7
Neighbor Island Consultants (1973) collected data in Hilo Harbor between
July 17 and August 21, 1972. The data indicated a two cell circulation pattern
in the surface layers of the harbor. The eastern cell, in Kuhio Bay, circulated
clockwise with the tide while the western cell, centered northwest of Coconut
Island, circulated counterclockwise. Net transport of the entire system was
seaward, due at least in part to fresh water runoff from the Wailoa and
Wailuku Rivers and ground water inflow. Salinity of the deeper harbor waters
indicated replenishment from the ocean.
M & E Pacific, Inc. (1977 and 1980) conducted an evaluation of circulation
characteristics, water quality and geological definition of bottom types. The
circulation measurements were concentrated in the main reaches of the
harbor and the harbor entrance channel. A summary of the M & E findings is
presented below (Sea Engineering, 1981):
There is a two-layer salinity stratified pattern in Hilo Harbor. Vertical
stratification of the water column is caused by the large amounts of
fresh water entering the harbor from both ground water and surface
flow. The salinity gradient is more pronounced during the wet season
(winter).
The net transport of the surface layer is out of the harbor at a rate
dependent upon the quantity of freshwater input and wind speed and
direction.
The subsurface flows at the harbor mouth are influenced by the tide.
During flood tide, subsurface flow was generally into the harbor and
during ebb tide the flow generally reversed. During ebb tide, however,
occasions were noted when an inward flow persisted along the
western half of the harbor mouth. Similarly, there were times when the
subsurface water along the eastern side of the channel moved
continually seaward, even during flood tide. Flood and ebb tide current
speeds in the harbor entrance area averaged approximately
4cm/second during the 1980 phase of the study. Current meter data at
the harbor entrance showed a net transport into the harbor. The
relatively small volume of tidal exchange (the tidal prism) relative to the
large cross-section areas of the harbor entrance results in the very low
tide-related currents.
The two cell circulation in the upper layer described by Neighbor Island
Consultants (1973) was not confirmed by the M & E Pacific findings.
Variations between studies and even between drogues placed on the
same day were thought to be due to eddies with higher speeds than
those associated with the tidal flow.
8
Drogues placed off the mouth of the Wailoa River on April 7, 1977
during an ebbing tide moved north and west, or in a seaward direction.
Drogue depths were surface and 5 feet.
The results indicated weak and variable currents in the study area. An
ebb tide eddy is in the opposite direction of anticipated flow and may
be a countercurrent formed in response to ebb flows in the main harbor
channel. The flow reversals are apparently not wind related, as the
easterly trades would have reinforced movement in the opposite
direction.
M&E Pacific, Inc. (1980) determined that seasonal variations in circulation
and exchange characteristics in Hilo Harbor are primarily a function of the
amount of surface runoff which in turn is a function of rainfall. During the
winter, the larger fresh water inflow results in a thicker surface layer of fresh
water, a more pronounced stratification, and a stronger hydraulic gradient of
seaward flow in the surface layer.
Current studies nearshore in the vicinity of the bayfront beach by Sea
Engineering (1981) generally indicated weak and variable currents, with the
presence of eddies and tidal reversals. The study showed that resultant wave
approach at the bayfront beach is from north or northwest, and the breaking
waves at the shoreline set up an alongshore current moving to the east.
Dudley & Hallacher (1991) found that Hilo Bay is a salt wedge estuary that is
stratified with a freshwater surface layer existing up to a mile offshore. This
stratification is most pronounced during the wet season when surface runoff
to Hilo Bay is high. The dense saline layer moves offshore at depth with the
tide and the upper freshwater layer is pushed shoreward by easterly and
northeasterly trade winds. There is minimal mixing between freshwaters and
saltwater layers inside the breakwater because baywide wind/tidal circulation
and wave energy is low. Low wave energy also allows sediments carried by
the rivers to settle out into the lower salty layer, where they may be
transported back into the bay with the incoming tide. Wind generated tidal
velocities are probably too low to re-suspend bottom sediments, but
suspended sediments will move in and out of Hilo Bay with the wind and tide.
1.5. Study Method
This report was prepared under the authority of Section 22 by the USACE
Honolulu District (POH) with assistance from the U.S. Army Engineer
Research and Development Center’s (ERDC) Coastal and Hydraulics
Laboratory (CHL) and Field Research Facility (FRF), as well as the University
of Hawaii at Hilo (UHH) to collect field data and to implement numerical
modeling of circulation, wave transformation, and potential water quality
improvement in Hilo Bay. The initial focus of the study was to apply
appropriate numerical models to assess various project alternatives to
9
promote greater water circulation in Hilo Bay in order to improve water quality.
Model results and predictions for five alternative plans are documented in this
technical report to facilitate selection of an appropriate course of action. The
criteria for assessing alternative plans in this study were determined by
examining changes in waves, current circulation, water quality, and residence
time, as well as by determining areas subject to stagnant or weak circulation
or focused wave energy resulting from proposed alternatives. The initial
numerical modeling efforts concentrated on quantifying change in circulation
and wave patterns with and without the alternatives in place for a range of
forcing conditions.
The modeling of hydrodynamic and water quality conditions within Hilo Bay
was conducted utilizing a suite of linked models. An example of a previous
application of this linked modeling methodology is found in Bunch et al
(2003). Specific to Hilo Bay, regional and nested grid wave model STeady-
state spectral WAVE model (STWAVE, Smith et al., 1999) simulations were
performed to generate wave climate information and more specifically
radiation stress gradient fields used as wave forcing within water circulation
models ADvanced CIRCulation (ADCIRC, Luettich et al., 1992) and CH3D-
WES (Chapman et al., 1996). Regional scale ADCIRC simulations were
performed to provide water surface elevation boundary conditions for the
near-field CH3D-WES circulation model. Lastly, CH3D-WES simulations
were performed to supply hydrodynamic transport input for the CE-QUAL-
ICM, (Cerco and Cole, 1994) bay flushing and water quality simulations.
Descriptions of the individual models and their respective linkage components
within the overall model linkage and simulation system is provided in Sections
4 and 5 of this report.
2. HARBOR ALTERNATIVES CONSIDERED
2.1. Existing Conditions and Additional Dredging
The Hilo Harbor Federal navigation project consists of an entrance channel
that is 2,200 feet long by 440 feet wide by 39 feet deep, a turning basin 1,800
feet long by 1,400 feet wide by 38 feet deep and a 10,080-foot long
continuous breakwater. Figure 2-1 displays the existing conditions within Hilo
Bay and Hilo Harbor. Dredging of approximately 780,000 cubic yards of silty
material from the interior of the bay is assumed to be required for all
alternatives to provide initial water quality improvements within the bay.
2.2. Alternative 1
This alternative considers deauthorization and removal of the outer 7,500 feet
of the existing Hilo Harbor breakwater, construction of a new 2,000-foot long
breakwater and dredging within Hilo Bay (see Figure 2-2). In order to
increase wave induced circulation within Hilo Bay, the outer 7,500 feet of the
existing breakwater would be completely removed down to pre-construction
10
depths. A new 1,500-foot long breakwater would be constructed along the
seaward extent of the Hilo Harbor turning basin and a portion of the entrance
channel to provide safe navigation and mooring.
Figure 2-1. Existing conditions within Hilo Harbor
11
2.3. Alternative 2
This alternative considers notching six gaps into the existing Hilo Harbor
breakwater and dredging within Hilo Bay (see Figure 2-3). In order to
increase wave induced circulation within Hilo Bay, six 250-foot long gaps
would be notched into the outer portion of the existing breakwater at 750-foot
intervals
Figure 2-2. Alternative 1
12
2.4. Alternative 3
Similar to Alternative 2, this alternative (see Figure 2-4) also includes six
detached breakwaters segments on the harborside of the existing breakwater.
The offset segmented breakwaters would be constructed to reduce direct
wave transmission into the bay.
Figure 2-3. Alternative 2
13
Figure 2-4. Alternative 3
14
2.5. Alternative 4
Alternative 4 (as shown in Figure 2-5) consists of notching the existing
breakwater to provide one gap in the structure’s root. The 500-foot gap in the
breakwater would provide increased water exchange between Hilo Bay and
the ocean.
Figure 2-5. Alternative 4
15
2.6. Alternative 5
Alternative 5 (as shown in Figure 2-6) consists of notching the existing
breakwater to provide one gap in the structure’s root similar to Alternative 4.
A 500-foot offset breakwater would provide increased wave attenuation. The
offset breakwater would be constructed oceanward of the existing structure.
Figure 2-6. Alternative 5
16
2.7. Assumptions and Limitations of Alternatives
For Alternative 1 it was assumed that removal of 7,500 feet of the existing
breakwater would result in significant increase in water circulation within Hilo
Bay and in particular the bayfront beach area and Reeds Bay. Complete
removal of the structure’s cross section (down to pre-construction water
depths) was assumed across Blonde Reef to promote wave energy
transmission into Hilo Bay. Limitations of this assumption are that increased
wave energy transmission could result in increased shoreline erosion and
hazardous navigation conditions.
For Alternative 2 and Alternative 3, it was assumed that notching of the
existing breakwater would significantly improve water circulation within the
bay. A limitation of these alternatives is that notching the breakwater may not
provide enough increase in wave energy into the bay to increase water
circulation and thereby improve water quality.
Assumptions made for Alternative 4 and Alternative 5 are that tidal induced
water circulation would be significantly increased by provision of a means of
exchange between the ocean and the bay at the root of the breakwater. It
was also assumed that navigation would realize minimal negative impacts
due to the implementation of either alternative.
For all alternatives considered, it was assumed that dredging of 780,000
cubic yards of unsuitable material at various locations within the bay to a
vertical extent of 3 feet would significantly enhance the chances of improving
water circulation and water quality. By dredging unsuitable silty material from
the bay, the alternatives may be more successful than otherwise, but large
volumes of similar material will still remain in the bay and additional material
of marginal quality are likely to enter the bay through subsequent storm water
discharges.
3. FIELD DATA COLLECTION
3.1. Waves and Water Circulation Data
Wave and current data collection and processing were completed by the
USACE FRF. The data was collected with three Acoustic Doppler Current
Profilers (ADCPs). Instrument locations are shown in Figure 3-1 and listed in
Table 3-1. The ADCP gages were Teledyne RD Instruments 1200 kHz
Workhorse, bottom mounted facing upward with the sensor head
approximately 0.45 meters off the bottom.
17
The gages were deployed on 21 March 2007, and retrieved on 5 June 2007,
for a total deployment time of approximately 77 days. ADCP 1 (deployed
closest to the entrance channel) was selected for wave and current collection,
the other two gages only collected currents. Visual observations of waves at
ADCP sites 2 and 3, and analysis of data collected by ADCP 1, indicate that
typical wave heights at sites 2 and 3 were too low (nominally below 0.4
meters high) for reliable directional wave estimates to be made with these
instruments. ADCP 3 operated only 20 days before the batteries were
depleted on 7 April 2007.
Figure 3-1. ADCP Instrument Locations
18
Table 3-1. ADCP instrument identifications and locations
Gage
Lat
(deg
min)
Long
(deg min)
Deploy Times
(2007)
Depth
(m)
Wave/
Current
ADCP
1
19
44.3433 155 4.3783 21 Mar – 5 Jun 6
W/C
ADCP
2
19
44.9000 155 4.1350 21 Mar – 5 Jun 6
C
ADCP
3
19
44.3400 155.3.8317 21 Mar - 5 Jun 6
C
3.1.1. Wave Data Collection
ADCP 1 sampled at 2 Hz for directional wave measurements. Each
hourly wave burst was approximately 34 minutes long, starting at the top
of each hour, and consisted of 4,096 points. Wave spectra were
computed using the RDI “WavesMon v2.1” analysis program. This
package computes non-directional spectra from three different
parameters; the subsurface orbital velocity, the surface detection signal,
and the pressure sensor. The velocity data are used to compute
directional spectra. Figure 3-2 displays a typical monthly wave rose for
ADCP 1. Measurements for each month are fairly consistent, with waves
from the northwest with height never exceeding 1 meter.
3.1.2. Current Profile Data Collection
Current profiles were collected at ADCPs 1, 2, and 3 every 10 minutes
from a 200 point average. The gages have four acoustic transducers for
measuring currents and a pressure sensor, from which horizontal and
vertical current profiles were computed at 0.2 m vertical spacing. The RDI
“WavesMon v2.1” program also extracts the current profile data and
computes various parameters like vertically integrated currents and quality
control (QC) information. Current analysis for the ADCP 1 indicates the
surface flow is predominately westward, out of the harbor, as shown in
Figure 3-3. Figure 3-4 shows a typical current profile at Hilo Bay, and
illustrates the variation in current direction within the vertical water column.
19
Figure 3-2. Typical Monthly Wave Rose at ADCP 1 (March 2007)
20
Figure 3-3. Typical Current Measurement Time-Series at ADCP1
21
Figure 3-4. Typical Current Profile Snapshot at ADCP1
22
3.1.3. Current Drogue Data Collection
Current drogues (drifters) were deployed in Hilo Bay to investigate surface
and sub-surface water circulation. Four current drogues were designed
and built at the CHL Field Research Facility (FRF) that used Global
Positioning System (GPS) tracking and radio telemetry for positioning.
They were constructed with off-the-shelf plumbing supplies (PVC pipe,
vertical risers, rubber unions, hose clamps), a Garmin Geko GPS
receivers, and MaxStream (model XStream-PKG-R) radio modems
(Figure 3-5). The sails were approximately one meter in cross-section.
Surface drogues extended from the air/water interface to a depth of
approximately 1 meter. Sub-surface drogues consisted of a floating
section containing all of the requisite hardware and a tethered sail section
extending through the second meter of the water column (see Figure 3-6).
Each Garmin GCP unit internally recorded positions every 5 seconds.
Drogue tracks recorded from 21 to 22 March 2007 are shown in Figure 3-
7a. The drogues were deployed in the vicinity of ADCP 1 and ADCP 3 for
correlation with the gage data and at or near the mouth of Wailoa Stream.
The drogue tracks were generally east to west at both the surface and at
depth. Direction of travel for the surface and sub-surface drogues were
similar. Surface drogue speeds ranged from 5 centimeters per second
(cm/s) to 13 cm/s while the sub-surface drogue speeds ranged from 3
cm/s to 7 cm/s. Overall, the surface drogue speeds were approximately
twice that of the sub-surface drogues. Drogue tracks collected during the
three other deployments (22-23 March, 5-6 June, and 6-7 June) are
shown in Figures 3-7b, 3-7c and 3-7d. Tide stage during drogue
deployment is also shown in the figures for reference.
3.2. Water Quality Data
The University of Hawaii at Hilo (UHH) collected the baseline data on
sediment and nutrient inputs to the bay, and assessed the response of the
bay to these inputs under base and storm flow conditions. This baseline data
was used to calibrate and verify the water quality model.
The water quality monitoring specifically examined how storms affect water
quality (sediments, nutrients, “Chl a”) in Hilo Bay by comparing conditions in
the bay before and following a storm event over a two-year period. A similar
design has been successfully used by Ringuet & Mackenzie (2005) to
evaluate the effects of storms on water quality and algae in southern
Kaneohe Bay, Oahu.
23
Figure 3-5. Photo of Floating Surface Drogue with GPS Antenna
Figure 3-6. Photo of Floating Sub-surface Drogue with GPS Antenna
24
Figure 3-7a. Drogue Tracks between 21 -22 March 2007
Figure 3-7b. Drogue Tracks between 22 - 23 March 2007
25
Figure 3-7c. Drogue Tracks between 5 – 6 June 2007
Figure 3-7d. Drogue Tracks between 6 – 7 June 2007
26
For this monitoring study, eight stations were sampled for suspended
sediments, nutrients, and chlorophyll a (“Chl a”) (Figure 3-8). One station
each was located in the freshwater portion of the Wailuku and Wailoa Rivers
(S1 and S4). These stations were used to determine the amount of
suspended sediments and nutrients entering the bay from surface waters.
Four stations were located inside of Hilo Bay. Two Hilo Bay stations were
located along a transect following the Wailoa River plume (S5 and S6). The
other two Hilo Bay stations were located along a transect following the
Wailuku River plume (S2 and S3). This transect was on a slight angle to the
northwest of the river’s mouth because previous studies have shown that the
Wailuku River plume is deflected northwest in Hilo Bay (Dudley & Hallacher
1991). Two control sites were located outside of the Hilo Bay breakwater (C1
and C2), outside the direct influence of the two rivers.
Figure 3-8. UHH Water Quality Monitoring Stations (Wiegner, T. and Mead, L., 2007)
Wailuku River
Wailoa River
S1
&(
&(
&(
&(
&(
&(
S2
S3
S4
S5
S6
&(
&(
Control 1
Control 2
Proposed Sites
Control Sites
Estimated
Trajectory
&(
&(
27
From January 2007 through February of 2008, eight baseflow and four storm
events were sampled (Table 3-2). Each station was sampled for suspended
sediments, nutrients, and “Chl a”. Because the focus of this study was to
evaluate water quality in Hilo Bay before and after a storm, water samples
were collected from surface waters where river sediments and algae are most
likely concentrated due to the bay’s stratification. To characterize the
conditions at each station when sampling, physiochemical parameters
(salinity, specific conductivity, temperature, dissolved oxygen concentration,
dissolved oxygen percent saturation, light penetration) were measured using
a YSI multi-parameter meter and a Li-Cor light meter, respectively. Depth
profiles for these physiochemical parameters were measured at the six Hilo
Bay stations. Meteorological data (rainfall, winds, waves, and tides) were
also obtained for the sampling dates. For further detail on the methods and
results of the water quality monitoring, refer to Wiegner, T. and Mead, L.,
(2009).
Table 3-2. University of Hawaii at Hilo Sampling Events through
September 2007
(Wiegner, T. and Mead, L., 2009)
Event Date(s)
Storm 1 1/10/2007 – 1/15/2007
Storm 2 3/1/2007 – 3/6/2007
Storm 3 12/12/2007 – 12/17/2007
Storm 4 1/27/2008 – 2/1/2008
Base 1 3/14/2007
Base 2 5/3/2007 – 5/4/2007
Base 3 6/18/2007 – 6/19/2007
Base 4 7/8/2007 – 7/9/2007
Base 5 7/30/2007 – 7/31/2007
Base 6 9/5/2007 – 9/6/2007
Base 7 10/10/2007 – 10/11/2007
Base 8 11/7/2007 – 11/8/2007
28
4. HYDRODYNAMIC MODELING
4.1. Circulation Modeling
4.1.1. Model Descriptions
The ADCIRC numerical model was chosen for simulating the long-wave
hydrodynamic processes in the study area. Utilizing tidal constituent, wind
and atmospheric pressure data, the ADCIRC model can accurately
replicate tide induced and storm-surge water levels and currents. The
ADCIRC model was developed in the USACE Dredging Research
Program (DRP) as a family of two- and three-dimensional finite element-
based models (Luettich, Westerink, and Scheffner 1992). ADCIRC can
simulate tidal circulation and storm-surge propagation over very large
computational domains while simultaneously providing high resolution in
areas of complex shoreline configuration and bathymetry. In two
dimensions, the model is formulated using the depth-averaged shallow
water equations for conservation of mass and momentum. ADCIRC
utilizes a standard quadratic parameterization for bottom and wind stress.
Furthermore, radiation stress gradient forcing fields, supplied by STWAVE
in this application, are applied as a surrogate to wind stress. As such, the
radiation stress gradients represent a stress per unit mass of water having
units of m2/s2 (meters squared per second squared).
The three dimensional numerical hydrodynamic model CH3D-WES
(Curvilinear Hydrodynamics in Three Dimensions–Waterways Experiment
Station) can be applied in two vertical resolution modes, Z-grid and
Sigma–grid. The Z-grid version is documented in Johnson, et al. (1991).
The Sigma-grid version, used in this study, is documented in Chapman et
al. (1996). The basic Sigma-grid model (CH3D) was developed by Sheng
(1986) for WES but has been extensively modified, including the
development of the Z-grid version. These modifications have consisted of
implementing different basic numerical formulations of the governing
equations as well as substantial recoding of the model to provide
additional computational efficiency. CH3D-WES performs hydrodynamic
computations on a non-orthogonal curvilinear or boundary-fitted planform
grid. Physical processes impacting circulation and vertical mixing that are
modeled include tides, wind, density effects (salinity and temperature),
freshwater inflows, turbulence, and the effect of the earth's rotation.
The boundary-fitted coordinate feature of the model provides grid
resolution enhancement necessary to adequately represent deep
navigation channels and irregular shoreline configurations of the flow
system. The curvilinear grid also permits adoption of accurate and
economical grid schematization software. The solution algorithm employs
29
an external mode, consisting of vertically averaged equations, which
provides a solution for the free surface displacement for input to the
internal mode, which contains the full 3D equations. The 2D vertically, or
depth averaged option is used in the present study.
4.1.2. Model Setup and Forcing Data
The development of the ADCIRC grid for determining regional circulation
was initiated by using a previously developed finite element mesh that
encompasses the entire Hawaiian Island chain in an oval grid boundary
(Figure 4-1). This existing mesh used National Geophysical Data Center
(NGDC) ETOPO2 bathymetric data to generate deep water bathymetry at
a resolution of approximately 2 degrees. The grid was previously used for
model studies focusing on Southeast Oahu, and therefore, modifications
to the grid were needed to reduce nearshore resolution in that area (from
~ 50m to ~150m) as well as to increase resolution in the present areas of
interest, Hilo Bay (Figure 4-2) and Hilo Harbor (Figure 4-3) from ~350m to
~50m.
Project
Area
Figure 4-1. ADCIRC Grid for the Hawaiian Islands
30
Figure 4-2. ADCIRC Grid for Hilo Bay
Figure 4-3. ADCIRC Grid for Hilo Harbor
31
The precision of the Hilo Bay coastline in the grid was improved using
several tools including the National Oceanographic and Atmospheric
Administration’s (NOAA) Coastline Extractor, a georectified 2003 aerial
photo of the bay, and a NOAA digital nautical chart. The ADCIRC grid
was also improved with several sets of bathymetric data in the Hilo Bay
and Hilo Harbor areas. NOAA’s National Geophysical Data Center
(NGDC) Geophysical Data Management System (GEODAS) database of
digital historical surveys was used to update the higher-resolution areas of
the grid in intermediate waters both inside and outside the bay. A USACE
channel survey dated August 28, 2005 was incorporated into the grid area
inside Hilo Harbor, and a 2005 multibeam survey of the subaerial surface
of the Hilo Breakwater was used to represent the details of the breakwater
foundation and adjacent portions of Blonde Reef. USACE Scanning
Hydrographic Operational Airborne LiDAR Survey (SHOALS) bathymetry
data was not available for the Hilo Bay area at the time of model setup.
All newly incorporated bathymetry sets were converted to the existing
mesh coordinate system (Geographic, NAD 83, meters) and vertical
datum (Mean Tide Level, meters) for incorporation into the grid. In
addition to the existing condition (Figure 4-4), five additional model grids,
each based on an alternative breakwater configuration (Figure 4-5), were
generated with identical bathymetry and modifications made only to alter
the breakwater for alternatives as described in Section 2.
Figure 4-4. ADCIRC Bathymetry for Existing Breakwater Condition
32
The ADCIRC grid mesh is forced with the free surface position along the
open-water boundary that surrounds the Hawaiian Islands. Tidal forcing
conditions were developed for the ocean boundary condition with the
LeProvost tidal constituent database (LeProvost et al., 1994). The
LeProvost database was applied because it provided a stable solution for
the linked model validation time period. Offshore wind fields developed by
the National Center for Atmospheric Research (NCAR) and the National
Centers for Environmental Predication (NCEP) Reanalysis Project are
implemented as the wind forcing condition. The NCEP/NCAR Reanalysis
Project is a joint project whose goal is to produce new atmospheric
analyses using historical data (1948 onwards) and to produce analyses of
the current atmospheric state (Climate Data Assimilation System, CDAS).
The quality and utility of the re-analyses are superior to NCEP's original
analyses because a state-of-the-art data assimilation is used, more
observations are used, and quality control has been improved.
Atmospheric pressure fields were not included as forcing data.
Radiation stress gradients were applied from results of the STWAVE
model, in order to account for currents generated by wave breaking in the
vicinity of the harbor. The STWAVE wave model is discussed later in this
section. In addition, approximated stream flows into Hilo Bay from the
Figure 4-5. ADCIRC Bathymetry for Typical Alternative (Alternative 3)
33
Wailuku River were included as input to the ADCIRC runs. These values
were determined using daily data from US Geological Survey (USGS)
stream gages 1671300 (Wailuku River at Hilo Bay) and 16704000
(Wailuku River at Pi’ihonua), and determining a correlation factor between
the two to fill in data gaps. Wailoa River flow was not included as input to
the ADCIRC model runs, due to the minimal relative effect that this inflow
would have on the large area covered by the grid domain.
A nested CH3D-WES base grid and five breakwater alternative grids were
developed using shoreline and bathymetric data from the ADCIRC grid.
The entire base Hilo Bay grid with the existing breakwater structure and
bathymetry in meters is shown in Figure 4-6. A typical alternative
configuration CH3D grid (Alternative 3) is shown in Figure 4-7. Forcing
conditions for the CH3D model, including boundary water surface
elevations, winds, and Wailuku river flow, were derived from ADCIRC
simulations using “output stations” at locations within the ADCIRC grid that
correspond to the offshore boundary of the CH3D grid. Radiation stress
gradients were applied using values determined from the wave model, in a
similar manner to the method used for application to the ADCIRC model.
Figure 4-6. CH3D Grid Coverage (Existing Condition)
34
4.1.3. Validation and Calibration
A calibration run of ADCIRC results for water level was completed for the
period of April 10-24, 2001, and a comparison of model results was made
with measured water levels from NOAA tide station 1617760, located near
the Hilo Harbor Pier #3 (Figure 4-8). This time period was selected
because improved NCEP/NCAR wind fields from Oceanweather, Inc. were
available. Oceanweather, Inc utilized Interactive Optimum Kinematic
Analysis System (IOKA, Cox et al. 1995) for the generation of these 2001
regional wind fields. In this method, point source measurements and wind
estimates derived from satellite scatterometers are used improve the
accuracy of the background NCAR/NCEP reanalysis wind fields.
The run included tidal constituents and winds as forcing conditions. The
run showed approximate agreement with the tide gage, but some small
phase differences were noted, as well as a difference in water levels on
the order of approximately 0.2 meters.
Figure 4-7. CH3D Grid for Typical Alternative (Alternative 3)
35
A harmonic analysis of tidal constituents was conducted, and based on
this; adjustments were made to the tidal boundary conditions. A new
calibration run with adjusted tidal constituents resulted in an improved
comparison to the predicted tide gage (Figure 4-9). Additional forcing
conditions were added, including stream inputs from the Wailuku River
and radiation stresses determined from wave transformation model
STWAVE, but the resulting changes to water level were negligible.
Figure 4-8. NOAA Tide Station 1617760 at Hilo Harbor
36
Depth-averaged tidal calibration of the CH3D-WES model was performed
utilizing boundary water surface elevations, wind forcing, and Wailuku
river flow derived from the April 2001 ADCIRC simulation. The tide and
wind forcing data were updated every 0.5 hours during the simulation. The
groundwater inflow from the Wailoa River and Icy Bay was specified as
the approximate average low flow of the Wailuku River, which was
updated every 24 hours during the simulation. The results of the
calibration simulations are presented in Figure 4-10, which show the
predicted and measured tides at the Hilo Harbor gage with ADCIRC and
CH3D-WES model simulation results. The accurate representation of
volume change within the bay is vital to the accurate evaluation of flushing
for various alternatives. It is seen in this figure that after sufficient spin-up
time, both the ADCIRC and CH3D-WES model simulation results
accurately represent the time varying tidal prism within Hilo Harbor. The
departure of the gage prediction and model simulations from the gage
measurements is a result of atmospheric pressure variation during the
model simulation period.
-1
-0.75
-0.5
-0.25
0
0.25
0.5
0.75
1
11 13 15 17 19 21 23 25
April 2001Water Surface Elev, m (MSL)NOAA Gage 1617760 (Pred)
ADCIRC
Figure 4-9. ADCIRC Modeled Water Level Compared to Tide Prediction
37
Subsequent to satisfactory hydrodynamic model calibration, April 2001
simulations using CH3D-WES were performed for the existing condition
and five alternative grids to generate hydrodynamic transport files for CE-
QUAL-ICM water quality calibration. The hydrodynamic transport data
output interval for the water quality model was provided hourly.
4.2. Wave Modeling
4.2.1. Model Description
STWAVE is a spectral wave transformation model, which is capable of
representing depth-induced wave refraction and shoaling, current-induced
refraction and shoaling, depth- and steepness-induced wave breaking,
diffraction, wind-wave growth, wave-wave interaction and whitecapping
(Resio 1988, Smith et al. 2001). The purpose of applying nearshore wave
transformation models such as STWAVE is to describe quantitatively the
change in wave parameters between the offshore and the nearshore
because offshore time-series wave data is usually more commonly
Figure 4-10. ADCIRC & CH3D Comparison to Tide Measurement and Prediction
-1
-0.75
-0.5
-0.25
0
0.25
0.5
0.75
1
11 13 15 17 19 21 23 25
April 2001Water Surface Elev, m (MTL)NOAA Gage 1617760 (Pred)
NOAA Gage 1617760 (Meas)
ADCIRC
CH3D
38
available. STWAVE has previously been applied to numerous sites with a
gently sloping seafloor or small areas of hardbottom. Due to the wide and
relatively shallow reef fronting the Hilo Harbor breakwater, this application
of STWAVE required the added feature of simulating wave transformation
over reefs. Development of a bottom friction capability in STWAVE was
completed to address this unique bathymetry specific to the island
environment.
4.2.2. Model Setup and Forcing Data
A nested STWAVE grid was developed with a deep water “coarse” grid
located so that the offshore grid boundary coincided with the nearest
Pacific Wave Information Study (WIS) Hindcast Station (Station 105), See
Figure 4-11a. The coarse grid has a resolution of 250 meters. A
nearshore “fine” grid was created to continue the transformation into Hilo
Bay and Hilo Harbor at a resolution of 25 meters, see Figure 4-11b.
Output wave spectra are generated from the coarse grid along a row that
overlaps with the fine grid. This spectra is then linearly interpolated along
the offshore boundary of the fine grid and used as a forcing condition.
Figure 4-11a. STWAVE Coarse Grid
39
Bathymetry from each of the ADCIRC grids was interpolated onto the
STWAVE Cartesian grid for the existing condition (Figure 4-12) as well as
each alternative breakwater configuration (Figure 4-13), respectively, with
smoothing adjustments to the STWAVE grid cells for land boundaries and
the breakwater structure made as needed.
Incident wave spectra were generated at the coarse grid offshore
boundary using a parametric spectral shape together with a directional
spreading function and based on wave parameters (wave height, period,
and direction), as determined from various wave data sources including
WIS and National Data Buoy Center (NDBC) buoys, corresponding to the
different run periods. A variable Manning friction coefficient (n= 0.020 for
non-reef, n = 0.20 for reefs) was used on the fine grid to incorporate the
friction effects over Blonde Reef. These friction values were based on
previous applications of STWAVE in a reef environment along the coast of
Southeast Oahu (Cialone, et al., 2008).
Figure 4-11b. STWAVE Fine Grid
40
Figure 4-12. Detailed Bathymetry of STWAVE Grid for Existing Condition
Blonde Reef
HILO BAY
Navigation Channel
Reeds
Pier
Figure 4-13. Detailed Bathymetry of STWAVE Grid for Typical
Alternative (Alt. 3)
41
4.2.3. Validation and Calibration
STWAVE was run for the April 10 – 24, 2001 time period using WIS
hindcast data from Station 105 as offshore input, in order to contribute
radiation stress gradients to the ADCIRC and CH3D calibration runs for
that time period. STWAVE was also run for the period of March – June
2007 and results were compared to wave data collected during the March
– June 2007 instrument deployment period, for the validation and
calibration of the wave model only. For the calibration period during
March – June 2007, the WIS station data at Station 105 was not available
to use as the incident offshore wave condition. The nearest directional
wave data available for this time was at NDBC Station 51001, northwest of
the island of Kauai. A transformation from the NDBC Station to the
STWAVE boundary was completed using WAVTRAN (Gravens, Kraus,
and Hanson, 1991).
As discussed in Thompson and Scheffner (2004), the WAVTRAN model
calculates spectral transformation of waves during propagation from one
depth to another shallower depth, taking into account shoreline orientation
and wave sheltering. The model assumes that sea and swell waves have
an energy spectrum that follows the Texel, MARSEN, ARSLOE (TMA)
spectral form (Bouws et al.1985). Directional spread is calculated by 4th
and 8th power cosine functions. Wave transformation calculation is
dependent on the shoreline orientation because bottom contours are
assumed parallel to the shoreline. If wave sheltering is included, wave
energy coming from directions specified by a sheltered angle band is
deleted from the spectrum. Typically, sheltering is applied as needed to
remove wave energy from any direction which is blocked from a straight-
line approach to the site by protruding land forms. Wave transformation
calculation is dependent on the shoreline orientation, in order to capture
the sheltering effects of the island chain.
In this case, an angle of 10 degrees was implemented in WAVTRAN to
incorporate the sheltering effect of the Hawaiian Island chain from
northwest waves at the offshore boundary of the STWAVE grid. Figure 4-
14 shows the general bathymetry of the island chain, the location of NDBC
Station 51001, the STWAVE coverage area, and a 10 degree sheltering
angle, in comparison with a 20 and 30 degree sheltering angle. Based on
this bathymetry, a 10 degree sheltering angle was considered appropriate.
This process effectively deleted any wave energy in the NDBC 51001
buoy spectra from west of 318 degrees True North. This “reduced” set of
refracted wave parameters in hourly increments was then further reduced
42
and processed into 6-hour incremental wave spectra at the STWAVE
coarse grid boundary as a deep water input condition to the model. A
constant water level equivalent to Mean High Water (0.254 meters above
Mean Tide Level) was used within the entire STWAVE domain. A
compiled model run of the coarse and fine grids was completed for 22
March 2007 through 9 June 2007 at 6-hour increments to develop a
nearshore time series at the instrument where waves were measured
(ADCP 1).
The nearshore time series extracted from the model run at ADCP 1 was
advanced 36 hours in time based on calculations of average wave celerity
(determined by average wave period) and distance from NDBC buoy
51001 to the offshore boundary of the STWAVE grid. This adjustment
was made to accommodate for the lag time between the real-time data at
the NDBC buoy and the time that these waves would reach the boundary
offshore of Hilo Bay.
Figure 4-14. Sheltering Angle used for Wave Transformation with WAVTRAN
STWAVE Grid
Boundary
43
The comparison of the model time series wave height to the measured
gage data at ADCP 1 was relatively good, capturing the trends in wave
height fluctuation, but with a slight bias toward lower wave heights in the
model results (Figure 4-15). Wave height at NDBC buoy 51001 and wave
height used at the STWAVE offshore boundary after transformation using
WAVTRAN is also shown in this time series. Some adjustments were
made to the Manning friction factor in an attempt to better calibrate the
model to the gage data; however, due to the relatively small and
concentrated location of the reef area and the distance from the reef to the
instrument location, these adjustments did not improve the agreement
significantly.
The agreement of the model with measured wave periods was also
generally good; however, the model appears to overpredict wave period at
several times, while missing intermittent higher wave periods recorded by
the instrument (Figure 4-16). Wave direction was also compared for
model results vs. gage data. Due to the strong diffractive effects on
waves moving past the breakwater to the instrument location, it is not
entirely surprising that agreement between the model wave direction and
measured wave direction is not in substantial agreement.
0.0
1.0
2.0
3.0
4.0
5.0
14-Mar-07 28-Mar-07 11-Apr-07 25-Apr-07 9-May-07 23-May-07 6-Jun-07Wave Ht (m)NDBC 51001
WAVTRAN Sheltering
ADCP 1 (Measured)
STWAVE (Modeled)
Figure 4-15. Wave Height Comparison (STWAVE vs. ADCP1)
44
The model-predicted wave direction spans a very small window, while the
measured wave direction varies widely, though both are primarily from the
northwest (coming from the direction of the harbor entrance), as shown in
Figure 4-17. It is also possible that the wave gage captured the direction
of some reflected or wind-generated wave energy that is not accounted for
in STWAVE. Overall, it was felt that since the wave energy was well-
represented by STWAVE in terms of wave height and period, that the
model was validated sufficiently to determine relative changes in wave
energy due to the alternative breakwater modifications.
0
5
10
15
20
14-Mar-07 28-Mar-07 11-Apr-07 25-Apr-07 9-May-07 23-May-07 6-Jun-07Wave Period (s)ADCP 1 (Measured)
STWAVE (Modeled)
Figure 4-16. Wave Period Comparison (STWAVE vs. ADCP1)
45
4.3. Combined Wave and Circulation Modeling
4.3.1. Radiation Stress Gradients in STWAVE
Radiation stress is the flux of momentum which is carried by ocean waves.
When these waves break, that momentum is transferred to the water
column, forcing nearshore currents. As such, the radiation stress
gradients represent a stress per unit mass of water having units of m2/s2.
This effect on currents (and in turn circulation) is important in this model
application due to the wave breaking that occurs on Blonde Reef, in the
vicinity of the breakwater and Hilo Harbor entrance. Radiation stress is
calculated in STWAVE based on linear wave theory. Gradients in
radiation stress are calculated in STWAVE to provide wave forcing to
external circulation models to drive nearshore currents and water level
changes (i.e., wave setup and setdown) (Smith et al. 2001). Fields of
radiation stress gradients, both x and y components, are provided for each
wave condition run in STWAVE, over the entire model domain.
0
90
180
270
360
14-Mar-07 28-Mar-07 11-Apr-07 25-Apr-07 9-May-07 23-May-07 6-Jun-07Wave Direction (TN)ADCP 1 (Measured)
STWAVE (Modeled)
Figure 4-17. Wave Direction Comparison (STWAVE vs. ADCP1)
46
4.3.2. Application of Radiation Stress Gradients in ADCIRC
Outputs of the radiation stress field over the fine STWAVE grid were
reformatted for use in the ADCIRC and CH3D models in order to account
for wave generated currents. A separate computer program was used to
spatially interpolate the radiation stress gradient solution from STWAVE to
the ADCIRC grid at 6 hour intervals (the same interval that offshore wave
conditions were updated in STWAVE). The application of radiation stress
gradients in ADCIRC resulted in increased current velocities where wave
breaking occurs, along the shoreline and on the outside of the breakwater
across Blonde Reef. Current velocities were increased due to the effect of
waves breaking on Blonde Reef by an order of magnitude of
approximately 0.5 meters/second (m/s).
4.3.3. Radiation Stress Gradients and Water Levels in CH3D
The CH3D-WES wind subroutine was modified to accept radiation stress
gradient forcing from the spatially interpolated solution file used for the
ADCIRC application. Non-trivial values (values exceeding 0.0001 m2/s2)
were extracted from the radiation stress gradient file and applied along the
seaward side of the breakwater and exposed shoreline. As in ADCIRC,
the radiation stress gradient forcing is applied as a supplement to wind
stress. The update interval of the radiation stress gradient forcing during
the CH3D-WES simulations was also 6 hours.
4.4. Modeling Simulation of February – March 2007
The primary modeling simulation during this project was completed for the
dates of 19 February 2007 through 19 March 2007. This time range was
selected because it encompassed the following wave, wind, and flow events:
tradewind/wave conditions, northwest swell waves, Kona wind/wave
conditions, light and variable wind conditions, a rain event (above average
river flow), and a dry period (below average river flow). A time series of
various meteorological conditions is shown in Figure 4-18, with the various
events labeled. In addition, UHH water quality data had been collected at two
instances during this period, one baseflow event (‘Base 1’ on 14 March 2007)
and one storm event (‘Storm 2’ during 1 - 6 March 2007). Use of this time
period for the primary modeling simulation would therefore enable
examination of the effects of various breakwater alternatives under several
weather conditions, in addition to verifying the CEQUAL-ICM model to the
UHH water quality data.
47
Figure 4-18. Meteorological Conditions during 19 February – 19 March 2007
48
4.4.1. Forcing Data
The forcing data used during this modeling simulation is similar to the data
used as input for the validation and calibration runs of STWAVE, ADCIRC,
and CH3D-WES. Offshore wave data was not available from Pacific WIS
Hindcast Data for the February – March 2007 time period, so directional
wave data from NDBC buoy 51001 (parameters of wave height, period
and direction shown in Figure 4-18) was again transformed to the
STWAVE boundary using WAVETRAN, and used to generate wave
spectra using a parametric spectral shape at 6 hour intervals. Water level
in STWAVE was again held constant at Mean High Water for wave runs
through both the coarse and fine wave grids. Spatially varying Manning’s
friction coefficients of n= 0.020 for non-reef and n = 0.20 for reefs were
also applied in the manner as the calibration run. Radiation stress
gradients from STWAVE output were interpolated for application in
ADCIRC and CH3D grids.
The harmonic-adjusted LeProvost tidal constituents used in the calibration
run of the ADCIRC model were applied again, with modifications made in
phase start time to correspond to the March 2007 run start time.
NCEP/NCAR Reanalysis wind fields were obtained for February and
March 2007 and applied in the ADCIRC model. As mentioned, radiation
stress gradient fields from the STWAVE model runs were interpolated to
the ADCIRC mesh and applied to the circulation model run for this time
period in areas where wave breaking occurs. Stream flow input at the
Wailuku River was incorporated using the daily USGS stream gage data
and applying the correlation factor previously determined to simulate flow
levels at the entrance to Hilo Bay.
Forcing conditions for the CH3D model (boundary water surface
elevations, winds, and Wailuku river flow) were again derived from
ADCIRC simulations using “output stations” at locations within the
ADCIRC grid that correspond to the offshore boundary of the CH3D grid.
Radiation stress gradients were applied using values determined from the
wave model at 6 hour intervals, in the same method used during
calibration runs. Wailoa River and Icy Bay groundwater flow was again
specified as the approximate average low flow of the Wailuku River, and
updated every 24 hours during the simulation.
4.4.2. Harbor Alternative Runs
STWAVE runs for each alternative were completed for the 19 February –
19 March 2007 time period at an interval of six hours. As discussed,
radiation stress gradients were passed forward to the ADCIRC and CH3D
models throughout the 30 day simulation. ADCIRC simulations were
completed for the same time period using the existing and five alternative
meshes, all had identical input forcing. Water surface elevation, wind, and
49
Wailuku river flow were extracted for the entire time series for each
alternative and used to force CH3D-WES. CH3D-WES simulations using
tide, wind and radiation stress gradient forcing were performed to
generate ICM hydrodynamic transport files for the base grid and five
alternative grid configurations.
The results of the February – March 2007 STWAVE runs were evaluated
in order to determine the changes to wave energy that would occur with
implementation of the various breakwater alternatives. Wave height
everywhere within the harbor was used as the primary indicator of an
increase in wave energy; therefore, a relative difference between the wave
heights for the existing breakwater configuration and each alternative was
used to evaluate the various proposed configurations. These wave height
differences were calculated for each alternative over the entire run period.
It is not feasible to show the wave difference plots throughout the entire
time series, so snapshots during a predominant tradewind condition only
are shown for each alternative (Figures 4-19a through 4-19e). A wave
height difference of 1.0 meter or greater within the navigation channel was
chosen as the evaluation criterion for bringing forward an alternative
breakwater configuration for water quality modeling. As shown in the
following figures, this criterion removed Alternative 1 (removal of the outer
7,500 feet of the existing breakwater and construction of a new 2,000-foot
long interior breakwater) from further consideration and water quality
modeling, after flushing simulation was completed.
Alternative 1 resulted in a marked increase in wave energy within the
harbor as would be expected due to the drastic reduction in breakwater
length (Figure 4-19a). An increase in wave height of 1.0 meter or greater
is shown within a large portion of Hilo Harbor, including the outer portion
of the navigation channel. The maximum increase within the harbor is
1.65 meters at several locations inside the previous breakwater alignment.
This alternative breakwater configuration continues to provide sheltering at
the interior of the harbor and navigation channel, with wave height
differences in the range of 0.0 to 0.2 meters. A significant increase in
wave energy approaching the Hilo Bay shoreline is evident by the green
shading in the figure, indicating a wave height increase of between 0.4
and 0.6 meters along the shoreline. As mentioned, this significant
increase in wave energy in the harbor and channel removed this
alternative from further consideration following flushing simulations.
50
Figure 4-19a. Typical Wave Height Increase for Alternative 1
Existing
Breakwater
Alignment
Navigation
Channel
Limits
Navigation
Channel
Limits
Figure 4-19b. Typical Wave Height Increase for Alternative 2
51
Navigation
Channel
Limits
Figure 4-19c. Typical Wave Height Increase for Alternative 3
Figure 4-19d. Typical Wave Height Increase for Alternative 4
Navigation
Channel
Limits
52
Alternative 2 (placing six gaps in the outer reach of the breakwater) also
shows an increase in wave height within the harbor (Figure 4-19b). Wave
height increase exceeds 1.0 meter within the gaps, which is not
unexpected since there is no wave occurring here under existing
conditions. The maximum increase in wave height in the harbor (not
including area within the gaps) is 0.85 meters just inside the gaps, and the
maximum increase in wave height within the navigation channel is 0.34
meters. Waves at the Hilo Bay shoreline are increased between 0 and
0.15 meters for this alternative. This alternative breakwater configuration
also continues to provide sheltering at the interior of the harbor and
navigation channel, with wave height differences in the range of 0.0 to 0.2
meters. This alternative was brought forward for further consideration of
flushing characteristics and water quality improvement.
Alternative 3 shows a smaller increase in wave energy than Alternative 2
due to the additional interior detached breakwaters present (Figure 4-19c).
Again, the largest overall increase in wave height is in the breakwater
gaps, where no waves are transmitted under existing conditions. The
greatest wave height increase in the harbor aside from this occurs
between the gaps and the detached breakwaters at 0.6 to 0.85 meters.
The maximum wave height increase within the navigation channel is 0.1
Navigation
Channel
Limits
Figure 4-19e. Typical Wave Height Increase for Alternative 5
53
meters. Wave energy at the shoreline is not increased measurably for this
alternative due to the added protection provided by the detached
breakwaters. Wave energy at the interior of the channel and at the vessel
piers is also virtually unchanged for this alternative. This alternative was
brought forward for further consideration of flushing characteristics and
water quality improvement.
Alternative 4 (single gap at breakwater root) shows a somewhat focused
area with an increase in wave height in comparison with the existing
breakwater configuration (Figure 4-19d). Wave height increases are
limited to the interior of the breakwater, near the entrance to the small
boat harbor. Again, the maximum wave height increase occurs in the gap;
however, the area just interior of the gap shows the largest increase
otherwise at 0.84 meters. The largest increase in wave height within the
channel for this alternative is 0.3 meters at the location closest to the
breakwater gap. Wave energy at the shoreline is not increased
measurably for this alternative and increases in wave height at the vessel
piers are less than 0.1 meter for this case. This alternative was brought
forward for further consideration of flushing characteristics and water
quality improvement.
Alternative 5 shows only a minimal amount of increase in wave energy
within the harbor, due to the addition of an exterior detached breakwater
(Figure 4-19e). The maximum increase in wave height within the harbor is
directly interior of the gap and is approximately 0.26 meters. The largest
increase in wave height within the channel is 0.11 meters at the location
closest to the gap. Similar to Alternative 4, wave energy at the shoreline
is not increased measurably for this alternative and increases in wave
height at the vessel piers are less than 0.1 meter for this case. This
alternative was brought forward for further consideration of flushing
characteristics and water quality improvement.
Evaluation of the ADCIRC and CH3D-WES simulations for February –
March 2007 was limited to a verification that water levels for both models
compared well with tide gage data from NOAA tide station 1617760 at Hilo
Harbor, as shown in Figure 4-20. Since these models were previously
shown (during the validation stage) to provide a satisfactory
representation of circulation within the harbor, the hydrodynamic data from
CH3D-WES between February – March 2007 was passed to the water
quality model as forcing data. The flushing studies conducted with CE-
QUAL-ICM are a better indicator of the changes in circulation and flushing
patterns due to each alternative breakwater configuration, and are detailed
in Section 5.2.2 of this report.
54
5. WATER QUALITY MODELING
5.1. Water Quality Model
5.1.1. Model Description
CE-QUAL-ICM (ICM) was designed to be a flexible, widely applicable,
state-of-the-art eutrophication model. Initial application was to
Chesapeake Bay (Cerco and Cole, 1994). Since the initial Chesapeake
Bay study, the ICM model code has been generalized with minor
corrections and model improvements. Subsequent additional applications
of ICM included the Delaware Inland Bays (Cerco et al. 1994), Newark
Bay (Cerco and Bunch, 1997), the San Juan Estuary (Bunch et al. 2000),
Florida Bay (Cerco et al. 2000), St. Johns River (in preparation) and Port
of Los Angeles (in preparation). Each model application employed a
different combination of model features and required addition of system-
specific capabilities.
-0.8
-0.6
-0.4
-0.2
0.0
0.2
0.4
0.6
0.8
19-Feb-07 26-Feb-07 5-Mar-07 12-Mar-07 19-Mar-07Water Level, m (MTL)ADCIRC
NOAA 1617760
CH3D
Figure 4-20. Water Level Validation for 19 February – 19 March 2007 Run Period
55
General features of the model include:
a. Operational in one-, two-, or three-dimensional configurations
b. Twenty-two +state variables including physical properties.
c. Sediment-water oxygen and nutrient fluxes may be computed in a
predictive sub-model or specified with observed sediment-oxygen
demand rates (SOD)
d. State variables may be individually activated or deactivated.
e. Internal averaging of model output over arbitrary intervals.
f. Computation and reporting of concentrations, mass transport,
kinetics transformations, and mass balances.
g. Debugging aids include ability to activate and deactivate model
features, diagnostic output, volumetric and mass balances.
h. Operates on a variety of computer platforms. Coded in ANSI
Standard FORTRAN F77.
ICM is limited by not computing the hydrodynamics of the modeled
system. Hydrodynamic variables (i.e., flows, diffusion coefficients, and
volumes) must be specified externally and read into the model.
Hydrodynamics may be specified in binary or ASCII format and are usually
obtained from a hydrodynamic model such as the CH3D_WES model
(Johnson et al. 1991).
The foundation of CE-QUAL-ICM is the solution to the three-dimensional
mass-conservation equation for a control volume. Control volumes
correspond to cells on the model grid. CE-QUAL-ICM solves, for each
volume and for each state variable, the equation:
in which:
Vj = volume of jth control volume (m3)
Cj = concentration in jth control volume (g m-3)
t, x = temporal and spatial coordinates
n = number of flow faces attached to jth control volume
Qk = volumetric flow across flow face k of jth control volume (m3 s-1)
Ck = concentration in flow across face k (g m-3)
Ak = area of flow face k (m2)
Dk = diffusion coefficient at flow face k (m2 s-1)
Sj = external loads and kinetic sources/sinks in jth control volume (g s-1)
Equation 5-1:
S
l
DACQ = t
C V
jkk
n
1 = k
kk
n
1 = k
jj + x
C + Σ∑∑δ
δ
δ
δ
56
Solution of Equation 5-1 on a digital computer requires discretization of
the continuous derivatives and specification of parameter values. The
equation is solved explicitly using upwind differencing or the QUICKEST
algorithm (Leonard 1979) to represent Ck. The time step, determined by
stability requirements, is usually five to fifteen minutes. For notational
simplicity, the transport terms are dropped in the reporting of kinetics
formulations.
CE-QUAL-ICM incorporates 22 state variables in the water column
including physical variables, multiple algal groups, and multiple forms of
carbon, nitrogen, phosphorus and silica (Table 5-1). Two zooplankton
groups, microzooplankton and mesozooplankton, are available and can be
activated when desired.
Table 5-1.Water Quality Model State Variables
Temperature
Salinity
Fixed Solids
Cyanobacteria
Diatoms
Other Phytoplankton
Dissolved Organic Carbon
Refractory Particulate
Organic Carbon
Labile Particulate Organic Carbon
Nitrate + Nitrite Nitrogen
Ammonium
Dissolved Organic Nitrogen
Refractory Particulate Organic
Nitrogen
Labile Particulate Organic
Nitrogen
Total Phosphate
Dissolved Organic
Phosphorus
Refractory Particulate Organic
Phosphorus
Labile Particulate Organic
Phosphorus
Chemical Oxygen Demand
Dissolved Oxygen
Dissolved Silica Particulate Biogenic Silica
Formatted: Font: 12 pt
Formatted: Font: 12 pt, Do not
check spelling or grammar
Formatted: Font: 12 pt
Formatted: Font: 12 pt
Formatted: Font: 12 pt, Do not
check spelling or grammar
Formatted: Font: 12 pt
Deleted: Equation 5-1
Deleted: Equation 5-1
57
5.1.2. Model Setup
Modifications to the circulation pattern in Hilo Harbor have the potential to
impact water quality conditions in the harbor. The various proposed
breakwater alternatives will each result in a unique circulation pattern
which may redistribute materials in the harbor. Only by using a water
quality model can the impacts from the different configurations be
compared with each other on an equal basis.
The first step in applying a water quality model is to define the problem
and the level of modeling required to address that problem. A model such
as ICM can be applied with varying levels of sophistication. The
complexity of the water quality model application depends upon a number
of features including the nature of the water quality problem to be
investigated, data availability, and funding. Below are some possible types
of water quality modeling efforts that can be applied for a study.
1. Eutrophication – Involves the modeling of dissolved oxygen, algae,
nutrients, and carbon. Realistic loads (observed or estimated) are
required for all major discharges in the system. In addition, information on
constituent concentrations is required for development of boundary
conditions and for calibration. Sediment processes could either be
specified or simulated with a sediment diagenesis model. This is the most
involved approach in time and money and would provide the most
defensible results provided there is an adequate database for model
development.
2. DO/BOD/SOD – This approach is similar to number 1 except that all
oxygen demand is specified as a Biochemical Oxygen Demand (BOD).
Sediment Oxygen Demand (SOD) is specified as a constant rate and
together with BOD are the only sinks for DO. Information is required on
DO and BOD levels throughout the system for cursory model calibration.
Information (observed or estimated) is required for all significant
discharges. While less involved than level 1, this approach still requires
some calibration
3. DO/OD/SOD No calibration – This approach is similar to the second
approach but with more simplified processes. Rather than BOD, a zero-
order, background oxygen demand (OD) in units of mg/L/day is used. No
loads are input. Boundary concentrations are held constant. OD and
SOD are sinks for DO. No reaeration is allowed. The value for
background OD is assumed. No model calibration is required. Dissolved
oxygen is essentially modeled as a non-conservative tracer. Relative
changes in DO can be determined by comparing results from a base
condition simulation (present conditions) to a simulation made with a
proposed island. The driving mechanism in this approach is that localized
58
circulation changes result in differences in residence time which impact
dissolved oxygen.
4. Residence Time - This approach does not provide a measure of
dissolved oxygen but a measure of the impact that circulation changes
have on the time that water stays in a certain area. Assuming that oxygen
demands are the same throughout the system, an increase in residence
time would indicate a decrease in flushing and a decrease in DO.
There are advantages and disadvantages to the different approaches
listed above. The more comprehensive efforts require the most data and
time. Less rigorous approaches require less time and a more modest
amount of data. They rely more on inferences and assumptions and are
better suited for screening studies.
Since the focus of the current study is an investigation of possible
changes to circulation for the purpose of improving in-harbor water quality
a limited suite of constituents was applied. The constituents modeled are
listed in Table 5-2.
Table 5-2. Hilo State variables
Temperature Particulate Organic Carbon
Salinity Ammonia
Dissolved Oxygen Nitrate
Dissolved Organic Carbon Phosphate
These constituents are adequate to capture changes in the water quality
conditions in the Hilo harbor system as a result of any breakwater
modification.
The computational grids used for the Hilo Harbor water quality modeling
effort are the same as those used in the CH3D-WES modeling effort,
Figures 4-6 and 4-7. The ICM and CH3D-WES grids are identical except
that one row of cells is deleted in the water quality grid along the outer
(ocean) boundary of the hydrodynamic model grid. These cells are
removed from the water quality grid due to differences between the way
ICM and CH3D-WES handle flows at ocean boundaries. CH3D-WES
specifies a water surface elevation or head condition at the ocean
boundary while CEQUAL-ICM requires a flow for the face along the
boundary. Removing cells along the ocean boundary has no impact upon
water quality computations on the interior of the grid. Grid information is
contained in Table 5-3 for base and the five alternative cases.
59
Table 5-3. CEQUAL-ICM Grid Characteristics
Alternative Description # Cells #Flow Faces
0 Existing (Base) 16110 31870
1 Shortened Breakwater 16042 31786
2 Multiple Gaps on
Breakwater
16146 31948
3 Multiple Gaps on
Breakwater with
Harborside Detached
Breakwaters
16112 31851
4 Single Nearshore Gap on
Breakwater near shore
16124 31903
5 Single Nearshore Gap on
Breakwater with
Oceanward Detached
Breakwaters
16106 31856
5.2. Modeling Simulation of February – March 2007
5.2.1. Forcing Data
A certain amount of information is required to “drive” CEQUAL-ICM. This
information defines the conditions that exist initially throughout the system,
the conditions at the boundaries from which inflow/outflow rates and
conditions are obtained, and meteorological data which impacts the heat
exchange and temperature in the system. This in turn impacts the rates at
which chemical and biological processes occur. The following information
is required for an application of CEQUAL-ICM:
Meteorological data for Lyman Field Hilo, HI (Meteorological station
912850) was obtained for the period of 01 Jan 2007 through 15 Oct 2007.
This data was processed and daily equilibrium temperatures and heat
exchange coefficients generated for the desired periods (Eiker 1977).
Bathymetry and flow information are all obtained directly from CH3D. The
ICM grid captures its physical properties from the CH3D-WES grid. Cell
volumes, surface areas, and depths in ICM are the same as used in
CH3D. Flow information is averaged in CH3D-WES and output for every
flow face. Procedures and techniques are used so that the flow fields
generated by CH3D-WES are the same flow fields used in ICM to
transport water quality constituents.
60
A clear distinction needs to be made between initial and boundary
conditions and comparison data. Comparison data have no effect on
model performance - they are used only to assess model performance.
Initial and boundary conditions are of greater importance because they
directly affect model performance.
Constituent boundary condition information was obtained from data
collected in support of this study by the University of Hawaii at Hilo,
(Wiegner and Mead, 2009). This information consisted of temperatures
and concentrations of the constituents modeled. The Hilo water quality
model had four boundaries; Wailuku River, Wailoa River, Reed’s Bay, and
open ocean boundary. Data from Wiegner and Mead (2009) station S1
was used for the boundaries of the Wailuku River and S4 for the Wailoa
River. Information from S4 was also used for the boundary conditions in
the Reed’s Bay inflow. Data from stations C1 and C2, located outside of
the breakwater were used for conditions at the ocean boundary.
5.2.2. Flushing Studies
The purpose of this study was to determine to what degree flushing inside
Hilo harbor would be impacted by changes in the breakwater
configuration. Therefore, a primary focus of the water quality modeling
was to simulate conditions that would exist for these configurations and
compare them against conditions for the existing configuration. A
powerful way to do this is via the use of tracer flushing simulations. In
these cases, a conservative, i.e., non-reacting substance is introduced
into the system and its concentration is monitored with time. Since the
substance is conservative, any change in concentration is the result of the
movement and dilution of that substance. There is no decay, uptake, or
creation. The only manner via which the tracer can leave the model
domain is to be transported out a boundary.
A benefit of using a tracer to investigate flushing versus traditional water
quality constituents is that the modeler defines the location and magnitude
of the tracer. Tracers can be applied initially over select portions of a
system or discharged continuously with a tributary or outfall. When
conventional water quality constituents are used, there may or may not be
enough gradient in the values to definitively determine the change in
flushing.
Several different types of tracer flushing tests were investigated. In
different tests, tracers were loaded at the headwaters of the Wailuku and
Wailoa Rivers, the ocean boundary, and selected locations in the interior
of the system. The tracer test that was most illuminating from a flushing
point of view was the one where all cells inside the breakwater had initial
tracer concentrations of 10 mg/l. All cells outside the breakwater had
61
initial concentrations of 0 mg/l and as did all boundary flows. This test
was repeated for all scenarios under consideration.
Shown in Figure 5-1 are eighteen locations inside and outside of Hilo
Harbor where tracer concentrations were monitored. Stations S2, S3, S5,
S6, C1, and C2 correspond to stations sampled by Wiegner and Mead.
Stations A1 through A12 were selected in order to get insight on
conditions at locations in the system that had not been sampled.
Flushing tests were run using hydrodynamics information from February
and March 2007. This is the same period used for the water quality
calibration. As shown in Figures 5-2 through 5-19, all breakwater
configurations tested resulted in improved flushing, i.e., decreased tracer
concentrations, at all locations monitored. This behavior was definitively
observed in plan view images of concentration contours, Figure 5-20
through 5-24.
Figure 5-1. Hilo Bay time series comparison
stations
62
A1
0
1
2
3
4
5
6
7
8
9
10
0 5 10 15
DayPPT Alt 0
Alt 1
Alt 2
Alt 3
Alt 4
Alt 5
Figure 5-2. Tracer concentrations harbor flushing test at station A1
A2
0
1
2
3
4
5
6
7
8
9
10
0 5 10 15
DayPPT Alt 0
Alt 1
Alt 2
Alt 3
Alt 4
Alt 5
Figure 5-3. Tracer concentrations harbor flushing test at station A2
63
A3
0
1
2
3
4
5
6
7
8
9
10
051015
DayPPT Alt 0
Alt 1
Alt 2
Alt 3
Alt 4
Alt 5
Figure 5-4. Tracer concentrations harbor flushing test at station A3
A4
0
1
2
3
4
5
6
7
8
9
10
0 5 10 15
DayPPT Alt 0
Alt 1
Alt 2
Alt 3
Alt 4
Alt 5
Figure 5-5. Tracer concentrations harbor flushing test at station A4
64
A5
0
1
2
3
4
5
6
7
8
9
10
0 5 10 15
DayPPT Alt 0
Alt 1
Alt 2
Alt 3
Alt 4
Alt 5
Figure 5-6. Tracer concentrations harbor flushing test at station A5
A6
0
1
2
3
4
5
6
7
8
9
10
0 5 10 15
DayPPT Alt 0
Alt 1
Alt 2
Alt 3
Alt 4
Alt 5
Figure 5-7. Tracer concentrations harbor flushing test at station A6
65
A7
0
1
2
3
4
5
6
7
8
9
10
051015
DayPPT Alt 0
Alt 1
Alt 2
Alt 3
Alt 4
Alt 5
Figure 5-8. Tracer concentrations harbor flushing test at station A7
A8
0
1
2
3
4
5
6
7
8
9
10
0 5 10 15
DayPPT Alt 0
Alt 1
Alt 2
Alt 3
Alt 4
Alt 5
Figure 5-9. Tracer concentrations harbor flushing test at station A8
66
A9
0
1
2
3
4
5
6
7
8
9
10
051015
DayPPT Alt 0
Alt 1
Alt 2
Alt 3
Alt 4
Alt 5
Figure 5-10. Tracer concentrations harbor flushing test at station A9
A10
0
1
2
3
4
5
6
7
8
9
10
0 5 10 15
DayPPT Alt 0
Alt 1
Alt 2
Alt 3
Alt 4
Alt 5
Figure 5-11. Tracer concentrations harbor flushing test at station A10
67
A11
0
1
2
3
4
5
6
7
8
9
10
0 5 10 15
DayPPT Alt 0
Alt 1
Alt 2
Alt 3
Alt 4
Alt 5
Figure 5-12. Tracer concentrations harbor flushing test at station A11
A12
0
1
2
3
4
5
6
7
8
9
10
0 5 10 15
DayPPT Alt 0
Alt 2
Alt 3
Alt 4
Alt 5
Alt 1
Figure 5-13. Tracer concentrations harbor flushing test at station A12
68
S2
0
1
2
3
4
5
6
7
8
9
10
0 5 10 15
DayPPT Alt 0
Alt 1
Alt 2
Alt 3
Alt 4
Alt 5
Figure 5-14. Tracer concentrations harbor flushing test at station S2
S3
0
1
2
3
4
5
6
7
8
9
10
051015
DayPPT Alt 0
Alt 1
Alt 2
Alt 3
Alt 4
Alt 5
Figure 5-15. Tracer concentrations harbor flushing test at station S3
69
S5
0
1
2
3
4
5
6
7
8
9
10
051015
DayPPT Alt 0
Alt 1
Alt 2
Alt 3
Alt 4
Alt 5
Figure 5-16. Tracer concentrations harbor flushing test at station S5
S6
0
1
2
3
4
5
6
7
8
9
10
051015
DayPPT Alt 0
Alt 1
Alt 2
Alt 3
Alt 4
Alt 5
Figure 5-17. Tracer concentrations harbor flushing test at station S6
70
C1
0
1
2
3
4
5
6
7
8
9
10
0 5 10 15
DayPPT Alt 0
Alt 1
Alt 2
Alt 3
Alt 4
Alt 5
Figure 5-18. Tracer concentrations harbor flushing test at station C1
C2
0
1
2
3
4
5
6
7
8
9
10
0 5 10 15
DayPPT Alt 0
Alt 1
Alt 2
Alt 3
Alt 4
Alt 5
Figure 5-19. Tracer concentrations harbor flushing test at station C2
71
Figure 5-20. Tracer concentrations in flushing test Alt. 0 (base case) for (left to right
from upper left) days 0, 1, 2, and 5.
72
Figure 5-21. Tracer concentrations in flushing test Alt. 2 (multiple gaps in outer
breakwater) for (left to right from upper left) days 0, 1, 2, and 5
73
Figure 5-22. Tracer concentrations in flushing test Alt. 3 (multiple gaps in outer
breakwater with interior groins) for (left to right from upper left) days 0, 1, 2, and 5
74
Figure 5-23. Tracer concentrations in flushing test Alt. 4 (single gap in
breakwater) for (left to right from upper left) days 0, 1, 2, and 5
75
Figure 5-24. Tracer concentrations in flushing test Alt. 5 (single gap in
breakwater with external groin) for (left to right from upper left) days 0, 1, 2, and 5
Further indication of the impact that the different alternatives have on
circulation is shown in Table 5-4. Figures 5-25 through 5-29 illustrate the
flows at the harbor entrance. By definition, positive flows are out of the
harbor, negative are into the harbor. As indicated for the exiting case,
Alternative 0, there is little net flow out of the harbor mouth.
76
Figure 5-25. Harbor mouth flow rate during water quality modeling period for Alt. 0.
Alt 2 Harbor Entrance
-500
0
500
1000
1500
2000
2500
3000
3500
0 100 200 300 400 500 600 700
Hoursm3/s
Figure 5-26. Harbor mouth flow rate during water quality modeling period for Alt 2
77
Alt 3 Harbor Entrance
-500
0
500
1000
1500
2000
2500
3000
3500
0 100 200 300 400 500 600 700
Hoursm3/s
Figure 5-27. Harbor mouth flow rate during water quality modeling period for Alt. 3
Alt 4 Harbor Entrance
-500
0
500
1000
1500
2000
2500
3000
3500
0 100 200 300 400 500 600 700
Hoursm3/s
Figure 5-28. Harbor mouth flow rate during water quality modeling period for Alt. 4
78
Alt 5 Harbor Entrance
-500
0
500
1000
1500
2000
2500
3000
3500
0 100 200 300 400 500 600 700
Hoursm3/s Figure 5-29. Harbor mouth flow rate during water quality modeling period for Alt. 5.
This is understandable as the only difference between the inflow and
outflow at the mouth are the freshwater flows from the streams emptying
in the harbor. The negative value for Alternative 0 is attributed to the
beginning and ending of the simulation not corresponding to the same
time in the tidal cycle. When these results are compared against the
alternatives with breakwater openings, there is a net outflow at the harbor
mouth. This outflow is generated by water entering the harbor through the
causeway gaps. As a result there is a rapid net transport of material out of
the harbor in comparison to the existing conditions where material is led
much longer.
Table 5-4. Net flows at Hilo Harbor Mouth
Alternative Description Net
Flow
M 3/s
0 Existing (Base) -6.9
2 Multiple Gaps on Breakwater 2380.3
3 Multiple Gaps on Breakwater with interior groins 2291.8
4 Single Nearshore Gap on Breakwater 606.0
5 Single Nearshore Gap on Breakwater with
exterior groin
332.3
79
5.2.3. Validation to UH Water Quality Data
Prior to application to Hilo Harbor for water quality simulations, CEQUAL-
ICM was calibrated using observed data collected by the University of
Hawaii at Hilo (Wiegner and Mead 2007). The same period was used for
calibration as was used for the subsequent alternative evaluation
scenarios. The University of Hawaii at Hilo data was used to generate
boundary conditions for the Wailuku and Wailoa Rivers and the ocean
Boundary. Boundary condition information is shown in Table 5-5.
Table 5-5.Water Quality Boundary Conditions
Constituent Ocean Wailuku Wailoa
Temperature (C) 23 23 23
Salinity (ppt) 35 0 0
Dissolved Oxygen (mg/l) 6.0 5.0 5.0
Suspended Solids (mg/l) 14.3 20 21
DOC (mg/l) 1.0 1.0 0.3
POC (mg/l) 0.1 0.2 0.2
Ammonia (mg/l) 0.002 0.01 0.003
Nitrate (mg/l) 0.02 0.05 0.3
Dissolved Inorganic Phosphorus (mg/l) 0.001 0.001 0.001
Model output was compared against observed data collected at stations in
the harbor and outside of the breakwater, Figures 5-30 to 5-38. In some
cases these data were profile samples. In these cases, all data for that
day regardless of depth were plotted on the graph.
80
Overall the model appears to do a good job of representing general trend
in dissolved oxygen, temperature, and salinity during the period it was
applied, Figures 5-30 through 5-32. Model structural ability, i.e., two-
dimension, depth average, limits the model’s ability to capture three
dimensional effects evident in some of the data. For example, ICM in 2-D
mode is unable to capture the variation of dissolved oxygen and salinity
with depth in Hilo Harbor. In the case of dissolved oxygen, the observed
variation with depth is likely the combined result of the processes of
reaeration and sediment oxygen demand and physical stratification.
When the system is modeled as depth average, the effects of stratification
are not included. With that said, the model does do a good job of
capturing the trends of the dissolved oxygen observations in the system.
C1
0
5
10
15
20
25
30
35
40
0 5 10 15 20 25 30
DayPPT C2
0
5
10
15
20
25
30
35
40
0 5 10 15 20 25 30
DayPPT
S2
0
5
10
15
20
25
30
35
40
0 5 10 15 20 25 30
DayPPT S3
0
5
10
15
20
25
30
35
40
0 5 10 15 20 25 30
DayPPT
S5
0
5
10
15
20
25
30
35
40
0 5 10 15 20 25 30
DayPPT
Figure 5-30. Salinity
S6
0
5
10
15
20
25
30
35
40
0 5 10 15 20 25 30
DayPPT
81
C1
0
1
2
3
4
5
6
7
0 5 10 15 20 25 30
Dayg/m3C2
0
1
2
3
4
5
6
7
0 5 10 15 20 25 30
Dayg/m3
S2
0
1
2
3
4
5
6
7
8
0 5 10 15 20 25 30
Dayg/m3S3
0
1
2
3
4
5
6
7
8
0 5 10 15 20 25 30
Dayg/m3
S5
0
1
2
3
4
5
6
7
0 5 10 15 20 25 30
Dayg/m3S6
0
1
2
3
4
5
6
7
0 5 10 15 20 25 30
Dayg/m3Figure 5-31. Dissolved Oxygen
82
C1
0
5
10
15
20
25
30
0 5 10 15 20 25 30
DayTemp C2
0
5
10
15
20
25
30
0 5 10 15 20 25 30
DayTemp
S2
0
5
10
15
20
25
30
0 5 10 15 20 25 30
DayTemp S3
0
5
10
15
20
25
30
0 5 10 15 20 25 30
DayTemp
S5
0
5
10
15
20
25
30
0 5 10 15 20 25 30
DayTemp S6
0
5
10
15
20
25
30
0 5 10 15 20 25 30
DayTemp
Figure 5-32. Temperature
83
C1
0
5
10
15
20
25
30
0 5 10 15 20 25 30
DaySuspended Solids mg/lC2
0
5
10
15
20
25
30
0 5 10 15 20 25 30
DaySuspended Solids mg/l
S2
0
5
10
15
20
25
30
0 5 10 15 20 25 30
DaySuspended Solids mg/lS3
0
5
10
15
20
25
0 5 10 15 20 25 30
DaySuspended Solids mg/l
S5
0
5
10
15
20
25
0 5 10 15 20 25 30
DaySuspended Solids mg/lS6
0
5
10
15
20
25
0 5 10 15 20 25 30
DaySuspended Solids mg/l
Table 5-33. Suspended Solids
84
C1
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
0 5 10 15 20 25 30
DayDOC g/m3
C2
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
0 5 10 15 20 25 30
DayDOC g/m3
S2
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
0 5 10 15 20 25 30
DayDOC g/m3
S3
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
0 5 10 15 20 25 30
DayDOC g/m3
S5
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
0 5 10 15 20 25 30
DayDOC g/m3
S6
0
0.2
0.4
0.6
0.8
1
1.2
1.4
0 5 10 15 20 25 30
DayDOC g/m3
Figure 5-34. Dissolved Organic Carbon
85
C1
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 5 10 15 20 25 30
DayPOC mg/lC2
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 5 10 15 20 25 30
DayPOC mg/l
S2
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 5 10 15 20 25 30
DayPOC mg/lS3
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 5 10 15 20 25 30
DayPOC mg/l
S5
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 5 10 15 20 25 30
DayPOC mg/lS6
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 5 10 15 20 25 30
DayPOC mg/l
Table 5-35. Particulate Organic Carbon
86
C1
0
0.002
0.004
0.006
0.008
0.01
0.012
0.014
0.016
0.018
0.02
0 5 10 15 20 25 30
DayNH3 mg/lC2
0
0.002
0.004
0.006
0.008
0.01
0.012
0.014
0.016
0.018
0.02
0 5 10 15 20 25 30
DayNH3 mg/l
S2
0
0.002
0.004
0.006
0.008
0.01
0.012
0.014
0.016
0.018
0.02
0 5 10 15 20 25 30
DayNH3 mg/lS3
0
0.002
0.004
0.006
0.008
0.01
0.012
0.014
0.016
0.018
0.02
0 5 10 15 20 25 30
DayNH3 mg/lS5
0
0.002
0.004
0.006
0.008
0.01
0.012
0.014
0.016
0.018
0.02
0 5 10 15 20 25 30
DayNH3 mg/lS6
0
0.002
0.004
0.006
0.008
0.01
0.012
0.014
0.016
0.018
0.02
0 5 10 15 20 25 30
DayNH3 mg/l
Table 5-36. Ammonia
87
C1
0
0.05
0.1
0.15
0.2
0.25
0 5 10 15 20 25 30
DayNO3 mg/lC2
0
0.05
0.1
0.15
0.2
0.25
0 5 10 15 20 25 30
DayNO3 mg/l
S2
0
0.05
0.1
0.15
0.2
0.25
0 5 10 15 20 25 30
DayNO3 mg/lS3
0
0.05
0.1
0.15
0.2
0.25
0 5 10 15 20 25 30
DayNO3 mg/l
S5
0
0.05
0.1
0.15
0.2
0.25
0 5 10 15 20 25 30
DayNO3 mg/lS6
0
0.05
0.1
0.15
0.2
0.25
0 5 10 15 20 25 30
DayNO3 mg/l
Figure 5-37. Nitrate
88
C1
0
0.01
0.02
0.03
0.04
0.05
0 5 10 15 20 25 30
DayDIP mg/lC2
0
0.01
0.02
0.03
0.04
0.05
0 5 10 15 20 25 30
DayDIP mg/lS2
0
0.01
0.02
0.03
0.04
0.05
0 5 10 15 20 25 30
DayDIP mg/lS3
0
0.01
0.02
0.03
0.04
0.05
0 5 10 15 20 25 30
DayDIP mg/l
S5
0
0.01
0.02
0.03
0.04
0.05
0 5 10 15 20 25 30
DayDIP mg/lS6
0
0.01
0.02
0.03
0.04
0.05
0 5 10 15 20 25 30
DayDIP mg/l
Table 5-38. Dissolved Inorganic Phosphorus
Similar results are evident on the inner portions of the system, stations S2,
S3, and S5, where the observed salinity profiles indicated significant
freshening near the surface. A two dimensional model is unable to
capture this. When freshening in the model does occur, it is represented
as a decrease in the overall salinity of the water column as was evident at
station S2.
Suspended solids model results, Figure 5-33, indicated fairly constant
values after a few days of simulation. Values at open water stations
showed spikes between model days 10 and 15 (March 2-7). in the meta-
data for the observed data, this period is listed as a being a storm. Of
interest is the fact that the tributary suspended solids concentrations
during this period were much less that the values observed in open water.
89
As the model is driven by values based on the tributary observations, the
model will not capture this behavior. Possible reasons for the open water
observations exceeding the tributary values are that there were other un-
measured sources of suspended solids besides the Wailuku and Wailoa
rivers. As the observed values during this period at stations C1 and C2
were also elevated, it is thought that some sediment resuspension was
ongoing. This would increase solids levels in open waters irregardless of
tributary loadings.
Model results for Dissolved Organic Carbon, DOC, are good, Figure 5-34.
There is a little fluctuation in observed values associated with a storm
event around day 9. The model does a good job of capturing the
conditions in the system both inside and outside of the breakwater.
Model results for Particulate Organic Carbon, POC, were low, Figure 5-35.
Results for stations C1 and C2 were representative. Predictions inside of
the harbor were lower than observed. However, considering the low
magnitude of the observed values, the model performed adequately.
Model results for ammonia indicated that the model consistently over-
predicted observed levels of 0.000 at many stations, Figure 5-36. In
instances where ammonia levels were measured above 0.000, they were
associated with run-off events and were still small.
Model results for nitrate indicate that there is little variation throughout the
period simulated, Figure 5-37. Observed data demonstrate more volatility
associated with runoff events. Some of the observed values such as at
Station S3 exceed values which are in the observed data This is felt to
be an indication of there being loadings to the system that are not
currently captured in the model. Model prediction for stations C1 and C2
are slightly high but still representative.
Observed data for total dissolved phosphorus indicated that values
throughout the system were consistently near 0, Figure 5-38. The model
was able to capture this behavior. However on model day 22 (March 14,
2007) observed total dissolved phosphorus values ranging from 0.022 –
0.042 mg/l were observed at all stations. No other values within an order
of magnitude of these values were observed throughout the nine month
sampling effort. Consequently, the water quality model did not capture
this behavior as there was nothing in the model to generate these
conditions. It is believed that these data either represent a loading,
process not in the model, or were bad data.
90
5.2.4. Harbor Alternative Runs
Water quality simulations were made for alternatives 2, 3, 4, and 5 using
the same model set up and time period as was used for calibration.
Alternative 1 was dropped from further modeling after flushing tests and
wave model results. Total simulation time was for 30 days. Model results
for the different harbor configurations were extracted for the same 18
stations as were used for model calibration. The first 15 days of the
simulation were plotted against Alternative 0 (Base) results. These are
shown in Figures 5-39 to 5-47.
Results indicate that the conditions at the stations plotted for all
alternatives were similar. The base condition results for the constituents
simulated were also similar to those observed for the four harbor
configurations evaluated. The overall similarity of the harbor alternative
configurations is attributed to two things. First, all alternative
configurations evaluated exhibited greatly increased circulation in
comparison to the base case. As such, the waters inside of the harbor
breakwater were mixed with and rapidly replaced by water from outside of
the breakwater. This is supported by the flushing results presented
earlier. Secondly, the depth average approach used in the model
prevented the creation of vertical structure in the water column which
could have resulted in more differences as it was displaced by the added
mixing of the harbor alternative simulations. For example, in a three
dimensional representation, low dissolved oxygen levels at certain
portions of the system might be impacted differently by the various flow
fields resulting from the different harbor alternatives considered.
There were some differences in the base results and the four harbor
alternative results. These differences appear to be the result of delays in
flushing in the base case as opposed to the alternatives. The fact that
differences were not numerically great at all times should not be ignored.
Conditions inside the harbor and outside were very similar in the model.
Therefore mixing of waters from outside of the harbor with the waters
inside did not necessarily result in large changes in concentrations or
temperature. This is somewhat an artifact of the two dimensional nature
of the model.
91
Figure 5-39. Alternative Salinities
A1
0
5
10
15
20
25
30
35
051015
DaySalinity PPTAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A2
0
5
10
15
20
25
30
35
051015
DayPPT Alt 0
Alt 2
Alt 3
Alt 4
Alt 5
A3
0
5
10
15
20
25
30
35
051015
DayPPT Alt 0
Alt 2
Alt 3
Alt 4
Alt 5
A4
0
5
10
15
20
25
30
35
051015
DayPPT Alt 0
Alt 2
Alt 3
Alt 4
Alt 5
A5
0
5
10
15
20
25
30
35
051015
DayPPT Alt 0
Alt 2
Alt 3
Alt 4
Alt 5
A6
0
5
10
15
20
25
30
35
051015
DayPPT Alt 0
Alt 2
Alt 3
Alt 4
Alt 5
A7
0
5
10
15
20
25
30
35
051015
DayPPT Alt 0
Alt 2
Alt 3
Alt 4
Alt 5
A8
0
5
10
15
20
25
30
35
051015
DayPPT Alt 0
Alt 2
Alt 3
Alt 4
Alt 5
A9
0
5
10
15
20
25
30
35
051015
DayPPT Alt 0
Alt 2
Alt 3
Alt 4
Alt 5
A10
0
5
10
15
20
25
30
35
051015
DayPPT Alt 0
Alt 2
Alt 3
Alt 4
Alt 5
A11
0
5
10
15
20
25
30
35
051015
DayPPT Alt 0
Alt 2
Alt 3
Alt 4
Alt 5
A12
0
5
10
15
20
25
30
35
051015
DayPPT Alt 0
Alt 2
Alt 3
Alt 4
Alt 5
C1
0
5
10
15
20
25
30
35
051015
DayPPT Alt 0
Alt 2
Alt 3
Alt 4
Alt 5
C2
0
5
10
15
20
25
30
35
051015
DayPPT Alt 0
Alt 2
Alt 3
Alt 4
Alt 5
S2
0
5
10
15
20
25
30
35
051015
DayPPT Alt 0
Alt 2
Alt 3
Alt 4
Alt 5
S3
0
5
10
15
20
25
30
35
0 5 10 15
DayPPT Alt 0
Alt 2
Alt 3
Alt 4
Alt 5
S5
0
5
10
15
20
25
30
35
0 5 10 15
DayPPT Alt 0
Alt 2
Alt 3
Alt 4
Alt 5
S6
0
5
10
15
20
25
30
35
0 5 10 15
DayPPT Alt 0
Alt 2
Alt 3
Alt 4
Alt 5
92
Figure 5-40. Alternative Dissolved Oxygen Concentrations
A1
0
1
2
3
4
5
6
7
8
9
10
0 5 10 15
DayDO mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A2
0
1
2
3
4
5
6
7
8
9
10
0 5 10 15
DayDO mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A3
0
1
2
3
4
5
6
7
8
9
10
0 5 10 15
DayDO mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A4
0
1
2
3
4
5
6
7
8
9
10
0 5 10 15
DayDO mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A5
0
1
2
3
4
5
6
7
8
9
10
0 5 10 15
DayDO mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A6
0
1
2
3
4
5
6
7
8
9
10
0 5 10 15
DayDO mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A7
0
1
2
3
4
5
6
7
8
9
10
0 5 10 15
DayDO mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A8
0
1
2
3
4
5
6
7
8
9
10
0 5 10 15
DayDO mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A9
0
1
2
3
4
5
6
7
8
9
10
0 5 10 15
DayDO mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A10
0
1
2
3
4
5
6
7
8
9
10
0 5 10 15
DayDO mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A11
0
1
2
3
4
5
6
7
8
9
10
0 5 10 15
DayDO mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A12
0
1
2
3
4
5
6
7
8
9
10
0 5 10 15
DayDO mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
C1
0
1
2
3
4
5
6
7
8
9
10
0 5 10 15
DayDO mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
C2
0
1
2
3
4
5
6
7
8
9
10
0 5 10 15
DayDO mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
S2
0
1
2
3
4
5
6
7
8
9
10
0 5 10 15
DayDO mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
S3
0
1
2
3
4
5
6
7
8
9
10
0 5 10 15
DayDO mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
S5
0
1
2
3
4
5
6
7
8
9
10
0 5 10 15
DayDO mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
S6
0
1
2
3
4
5
6
7
8
9
10
051015
DayDO mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
93
A1
0
5
10
15
20
25
30
0 5 10 15
DayTemp Alt 0
Alt 2
Alt 3
Alt 4
Alt 5
A2
0
5
10
15
20
25
30
0 5 10 15
DayTemp Alt 0
Alt 2
Alt 3
Alt 4
Alt 5
A3
0
5
10
15
20
25
30
0 5 10 15
DayTemp Alt 0
Alt 2
Alt 3
Alt 4
Alt 5
A4
0
5
10
15
20
25
30
0 5 10 15
DayTemp Alt 0
Alt 2
Alt 3
Alt 4
Alt 5
A5
0
5
10
15
20
25
30
0 5 10 15
DayTemp Alt 0
Alt 2
Alt 3
Alt 4
Alt 5
A6
0
5
10
15
20
25
30
0 5 10 15
DayTemp Alt 0
Alt 2
Alt 3
Alt 4
Alt 5
A7
0
5
10
15
20
25
30
051015
DayTemp Alt 0
Alt 2
Alt 3
Alt 4
Alt 5
A8
0
5
10
15
20
25
30
051015
DayTemp Alt 0
Alt 2
Alt 3
Alt 4
Alt 5
A9
0
5
10
15
20
25
30
051015
DayTemp Alt 0
Alt 2
Alt 3
Alt 4
Alt 5
A10
0
5
10
15
20
25
30
051015
DayTemp Alt 0
Alt 2
Alt 3
Alt 4
Alt 5
A11
0
5
10
15
20
25
30
051015
DayTemp Alt 0
Alt 2
Alt 3
Alt 4
Alt 5
A12
0
5
10
15
20
25
30
0 5 10 15
DayTemp Alt 0
Alt 2
Alt 3
Alt 4
Alt 5
C1
0
5
10
15
20
25
30
0 5 10 15
DayTemp Alt 0
Alt 2
Alt 3
Alt 4
Alt 5
C2
0
5
10
15
20
25
30
051015
DayTemp Alt 0
Alt 2
Alt 3
Alt 4
Alt 5
S2
0
5
10
15
20
25
30
0 5 10 15
DayTemp Alt 0
Alt 2
Alt 3
Alt 4
Alt 5
S3
0
5
10
15
20
25
30
0 5 10 15
DayTemp Alt 0
Alt 2
Alt 3
Alt 4
Alt 5
S5
0
5
10
15
20
25
30
0 5 10 15
DayTemp Alt 0
Alt 2
Alt 3
Alt 4
Alt 5
S6
0
5
10
15
20
25
30
051015
DayTemp Alt 0
Alt 2
Alt 3
Alt 4
Alt 5
Figure 5-41. Alternative Temperatures
94
A1
0
5
10
15
20
25
0 5 10 15
DaySuspended Solids mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A2
0
5
10
15
20
25
051015
DaySuspended Solids mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A3
0
5
10
15
20
25
051015
DaySuspended Solids mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A4
0
5
10
15
20
25
0 5 10 15
DaySuspended Solids mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A5
0
5
10
15
20
25
0 5 10 15
DaySuspended Solids mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A6
0
5
10
15
20
25
051015
DaySuspended Solids mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A7
0
1
2
3
4
5
6
7
8
9
10
051015
DayPPT Alt 0
Alt 2
Alt 3
Alt 4
Alt 5
A8
0
5
10
15
20
25
051015
DaySuspended Solids mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A9
0
5
10
15
20
25
051015
DaySuspended Solids mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A10
0
5
10
15
20
25
051015
DaySuspended Solids mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A11
0
5
10
15
20
25
051015
DaySuspended Solids mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A12
0
5
10
15
20
25
051015
DaySuspended Solids mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
C1
0
5
10
15
20
25
0 5 10 15
DaySuspended Solids mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
C2
0
1
2
3
4
5
6
7
8
9
10
051015
DayPPT Alt 0
Alt 2
Alt 3
Alt 4
Alt 5
S2
0
5
10
15
20
25
0 5 10 15
DaySuspended Solids mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
S3
0
5
10
15
20
25
0 5 10 15
DaySuspended Solids mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
S5
0
5
10
15
20
25
0 5 10 15
DaySuspended Solids mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
S6
0
5
10
15
20
25
0 5 10 15
DaySuspended Solids mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A7
0
5
10
15
20
25
051015
DaySuspended Solids mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
C2
0
5
10
15
20
25
051015
DaySuspended Solids mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A11
0
5
10
15
20
25
0 5 10 15
DaySuspended Solids mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
S5
0
5
10
15
20
25
0 5 10 15
DaySuspended Solids mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
Figure 5-42. Alternative Suspended Solids
95
A1
0
0.2
0.4
0.6
0.8
1
1.2
0 5 10 15
DayDOC mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A2
0
0.2
0.4
0.6
0.8
1
1.2
0 5 10 15
DayDOC mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A3
0
0.2
0.4
0.6
0.8
1
1.2
0 5 10 15
DayDOC mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A4
0
0.2
0.4
0.6
0.8
1
1.2
0 5 10 15
DayDOC mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A5
0
0.2
0.4
0.6
0.8
1
1.2
0 5 10 15
DayDOC mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A6
0
0.2
0.4
0.6
0.8
1
1.2
0 5 10 15
DayDOC mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A7
0
0.2
0.4
0.6
0.8
1
1.2
0 5 10 15
DayDOC mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A8
0
0.2
0.4
0.6
0.8
1
1.2
051015
DayDOC mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A9
0
0.2
0.4
0.6
0.8
1
1.2
0 5 10 15
DayDOC mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A10
0
0.2
0.4
0.6
0.8
1
1.2
0 5 10 15
DayDOC mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A11
0
0.2
0.4
0.6
0.8
1
1.2
0 5 10 15
DayDOC mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A12
0
0.2
0.4
0.6
0.8
1
1.2
0 5 10 15
DayDOC mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
C1
0
0.2
0.4
0.6
0.8
1
1.2
0 5 10 15
DayDOC mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
C2
0
0.2
0.4
0.6
0.8
1
1.2
051015
DayDOC mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
S2
0
0.2
0.4
0.6
0.8
1
1.2
051015
DayDOC mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
S3
0
0.2
0.4
0.6
0.8
1
1.2
051015
DayDOC mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
S5
0
0.2
0.4
0.6
0.8
1
1.2
051015
DayDOC mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
S6
0
0.2
0.4
0.6
0.8
1
1.2
051015
DayDOC mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
Figure 5-43. Alternative Dissolved Organic Carbon Concentrations
96
A1
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0 5 10 15
DayPOC mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A2
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0 5 10 15
DayPOC mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A3
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0 5 10 15
DayPOC mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A4
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
051015
DayPOC mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A5
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0 5 10 15
DayPOC mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A6
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0 5 10 15
DayPOC mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A7
0
1
2
3
4
5
6
7
8
9
10
0 5 10 15
DayPPT Alt 0
Alt 2
Alt 3
Alt 4
Alt 5
A8
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0 5 10 15
DayPOC mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A9
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0 5 10 15
DayPOC mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A10
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0 5 10 15
DayPOC mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A11
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0 5 10 15
DayPOC mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A12
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0 5 10 15
DayPOC mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
C1
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0 5 10 15
DayPOC mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
C2
0
1
2
3
4
5
6
7
8
9
10
051015
DayPPT Alt 0
Alt 2
Alt 3
Alt 4
Alt 5
S2
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0 5 10 15
DayPOC mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
S3
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0 5 10 15
DayPOC mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
S5
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0 5 10 15
DayPOC mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
S6
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0 5 10 15
DayPOC mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A7
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0 5 10 15
DayPOC mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
C2
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0 5 10 15
DayPOC mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
Figure 5-44. Alternative Particulate Organic Carbon Concentrations
97
A1
0
0.002
0.004
0.006
0.008
0.01
0.012
0.014
0.016
0.018
0.02
0 5 10 15
DayNH3 mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A2
0
0.002
0.004
0.006
0.008
0.01
0.012
0.014
0.016
0.018
0.02
051015
DayNH3 mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A3
0
0.002
0.004
0.006
0.008
0.01
0.012
0.014
0.016
0.018
0.02
051015
DayNH3 mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A4
0
0.002
0.004
0.006
0.008
0.01
0.012
0.014
0.016
0.018
0.02
0 5 10 15
DayNH3 mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A5
0
0.002
0.004
0.006
0.008
0.01
0.012
0.014
0.016
0.018
0.02
051015
DayNH3 mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A6
0
0.002
0.004
0.006
0.008
0.01
0.012
0.014
0.016
0.018
0.02
051015
DayNH3 mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A7
0
1
2
3
4
5
6
7
8
9
10
0 5 10 15
DayPPT Alt 0
Alt 2
Alt 3
Alt 4
Alt 5
A8
0
0.002
0.004
0.006
0.008
0.01
0.012
0.014
0.016
0.018
0.02
0 5 10 15
DayNH3 mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A9
0
0.002
0.004
0.006
0.008
0.01
0.012
0.014
0.016
0.018
0.02
0 5 10 15
DayNH3 mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A10
0
0.002
0.004
0.006
0.008
0.01
0.012
0.014
0.016
0.018
0.02
0 5 10 15
DayNH3 mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A11
0
0.002
0.004
0.006
0.008
0.01
0.012
0.014
0.016
0.018
0.02
0 5 10 15
DayNH3 mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A12
0
0.002
0.004
0.006
0.008
0.01
0.012
0.014
0.016
0.018
0.02
0 5 10 15
DayNH3 mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
C1
0
0.002
0.004
0.006
0.008
0.01
0.012
0.014
0.016
0.018
0.02
0 5 10 15
DayNH3 mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
C2
0
1
2
3
4
5
6
7
8
9
10
0 5 10 15
DayPPT Alt 0
Alt 2
Alt 3
Alt 4
Alt 5
S2
0
0.002
0.004
0.006
0.008
0.01
0.012
0.014
0.016
0.018
0.02
0 5 10 15
DayNH3 mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
S3
0
0.002
0.004
0.006
0.008
0.01
0.012
0.014
0.016
0.018
0.02
0 5 10 15
DayNH3 mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
S5
0
0.002
0.004
0.006
0.008
0.01
0.012
0.014
0.016
0.018
0.02
0 5 10 15
DayNH3 mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
S6
0
0.002
0.004
0.006
0.008
0.01
0.012
0.014
0.016
0.018
0.02
0 5 10 15
DayNH3 mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A7
0
0.002
0.004
0.006
0.008
0.01
0.012
0.014
0.016
0.018
0.02
0 5 10 15
DayNH3 mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
C2
0
0.002
0.004
0.006
0.008
0.01
0.012
0.014
0.016
0.018
0.02
0 5 10 15
DayNH3 mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
Figure 5-45. Alternative Ammonia Concentrations
98
A1
0
0.005
0.01
0.015
0.02
0.025
0.03
0.035
0.04
0.045
0.05
051015
DayNO3 mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A2
0
0.005
0.01
0.015
0.02
0.025
0.03
0.035
0 5 10 15
DayNO3 mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A3
0
0.01
0.02
0.03
0.04
0.05
0.06
0 5 10 15
DayNO3 mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A4
0
0.05
0.1
0.15
0.2
0.25
0 5 10 15
DayNO3 mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A5
0
0.005
0.01
0.015
0.02
0.025
051015
DayNO3 mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A6
0
0.005
0.01
0.015
0.02
0.025
0 5 10 15
DayNO3 mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A7
0
1
2
3
4
5
6
7
8
9
10
051015
DayPPT Alt 0
Alt 2
Alt 3
Alt 4
Alt 5
A8
0
0.005
0.01
0.015
0.02
0.025
051015
DayNO3 mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A9
0
0.005
0.01
0.015
0.02
0.025
051015
DayNO3 mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A10
0
0.005
0.01
0.015
0.02
0.025
051015
DayNO3 mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A11
0
0.005
0.01
0.015
0.02
0.025
051015
DayNO3 mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A12
0
0.005
0.01
0.015
0.02
0.025
051015
DayNO3 mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
C1
0
0.005
0.01
0.015
0.02
0.025
051015
DayNO3 mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
C2
0
1
2
3
4
5
6
7
8
9
10
0 5 10 15
DayPPT Alt 0
Alt 2
Alt 3
Alt 4
Alt 5
S2
0
0.005
0.01
0.015
0.02
0.025
0.03
0.035
0 5 10 15
DayNO3 mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
S3
0
0.005
0.01
0.015
0.02
0.025
0 5 10 15
DayNO3 mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
S5
0
0.005
0.01
0.015
0.02
0.025
0.03
0.035
0.04
0.045
0 5 10 15
DayNO3 mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
S6
0
0.005
0.01
0.015
0.02
0.025
0.03
0 5 10 15
DayNO3 mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A7
0
0.005
0.01
0.015
0.02
0.025
051015
DayNO3 mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
C2
0
0.005
0.01
0.015
0.02
0.025
0 5 10 15
DayNO3 mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
Figure 5-46. Alternative Nitrate Concentrations
99
A1
0
0.0002
0.0004
0.0006
0.0008
0.001
0.0012
0 5 10 15
DayDIP mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A2
0
0.0002
0.0004
0.0006
0.0008
0.001
0.0012
0 5 10 15
DayDIP mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A3
0
0.0002
0.0004
0.0006
0.0008
0.001
0.0012
0 5 10 15
DayDIP mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A4
0
0.0002
0.0004
0.0006
0.0008
0.001
0.0012
0 5 10 15
DayDIP mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A5
0
0.0002
0.0004
0.0006
0.0008
0.001
0.0012
0 5 10 15
DayDIP mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A6
0
0.0002
0.0004
0.0006
0.0008
0.001
0.0012
0 5 10 15
DayDIP mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A7
0
1
2
3
4
5
6
7
8
9
10
0 5 10 15
DayPPT Alt 0
Alt 2
Alt 3
Alt 4
Alt 5
A8
0
0.0002
0.0004
0.0006
0.0008
0.001
0.0012
0 5 10 15
DayDIP mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A9
0
0.0002
0.0004
0.0006
0.0008
0.001
0.0012
0 5 10 15
DayDIP mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A10
0
0.0002
0.0004
0.0006
0.0008
0.001
0.0012
0 5 10 15
DayDIP mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A11
0
0.0002
0.0004
0.0006
0.0008
0.001
0.0012
0 5 10 15
DayDIP mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A12
0
0.0002
0.0004
0.0006
0.0008
0.001
0.0012
0 5 10 15
DayDIP mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
C1
0
0.0002
0.0004
0.0006
0.0008
0.001
0.0012
0 5 10 15
DayDIP mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
C2
0
1
2
3
4
5
6
7
8
9
10
0 5 10 15
DayPPT Alt 0
Alt 2
Alt 3
Alt 4
Alt 5
S2
0
0.0002
0.0004
0.0006
0.0008
0.001
0.0012
0 5 10 15
DayDIP mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
S3
0
0.0002
0.0004
0.0006
0.0008
0.001
0.0012
0 5 10 15
DayDIP mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
S5
0
0.0002
0.0004
0.0006
0.0008
0.001
0.0012
0 5 10 15
DayDIP mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
S6
0
0.0002
0.0004
0.0006
0.0008
0.001
0.0012
0 5 10 15
DayDIP mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
A7
0
0.0002
0.0004
0.0006
0.0008
0.001
0.0012
0 5 10 15
DayDIP mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
C2
0
0.0002
0.0004
0.0006
0.0008
0.001
0.0012
051015
DayDIP mg/lAlt 0
Alt 2
Alt 3
Alt 4
Alt 5
Figure 5-47. Alternative Dissolved Inorganic Phosphorus Concentrations
Salinity results, Figure 5-39, indicated that the conditions inside of the
harbor are the same at all stations regardless of the harbor alternative
chosen. There were some subtle differences at station A1, which is at the
mouth of the Wailuku River. This is the most active station monitored in
the model output as a result of its location in the Wailuku River. At this
location, the results for the base case exhibited a drop in salinity around
day 2.0. Similar decreases were seen in the salinity for the other
100
alternatives but not to the degree observed in the base case. Salinity
results for Station S2 indicate a similar dip in salinity around day 2.0.
Station S2 is the nearest station to station A1. No dips are seen in the
salinity concentrations at stations S2 for the proposed harbor alternative
configurations. This should be taken as an indication of the added
flushing in this portion of the system as a result of the openings in the
breakwater. This flushing would aid in the dispersal of any materials in
originating from the Wailuku River.
Results for dissolved oxygen were similar for the base and four harbor
alternatives evaluated, Figure 5-40. This is somewhat expected. The
dissolved oxygen conditions in the harbor are good. The two dimensional
approach used in the model prevents creation of stratification and lower
dissolved oxygen levels below the surface. Reaeration acts to counter
any deficient between computed dissolved oxygen levels and saturation
and will in the absence of significant oxygen demand result in a dissolved
oxygen concentration near saturation.
Temperature results for the 18 stations were similar for all harbor
alternatives and the base case, Figure 5-41. The only difference occurs in
the first few days of the simulation when there is a slight decrease in
temperature as the result of the waters from outside of the harbor
replacing waters inside of the harbor. This is caused by the temperature
boundary condition used on the outer boundary being slightly less than the
initial temperature specified throughout the model. Of interest is that the
temperature decrease at stations inside of the harbor is lagged in the base
case when compared to all of the harbor alternatives. This is again an
indication of the degree of flushing occurring as a result of the harbor
breakwater breaches.
Suspended solids results indicated similar behavior for the four proposed
harbor alternatives evaluated, Figure 5-42. There were some deviations
during the initial flush period around day 2 between the multi-gap and
single gap alternatives. After that they were near identical. The base
case though exhibited higher concentrations as there was limited flushing.
By day 10 suspended solids levels were similar for the four proposed
alternatives and the base case.
Dissolved organic carbon results for the four harbor alternative
configurations were similar to base results, Figure 5-43. There were
subtle variations between the alternatives but nothing of significance.
Ammonia exhibited similar behavior as did nitrate.
Particulate organic carbon results indicated more variation among the four
alternatives (2, 3, 4, and 5), Figure 5-44. However, all were different than
the base case. The greatest differences occurred at locations where the
flushing in the base would be less than in the alternatives, stations S2 and
101
S5, and A8, A9, A10, A11. The reason for these differences is thought to
be the role of settling. In more quiescent waters, settling will remove
material more so than in waters where there is flushing and mixing. The
fact that model predictions for particulate organic carbon are higher for the
proposed harbor configurations than the base is indication of the relatively
limited exchange in the current configuration. The difference is due to the
POC settling out of the water column more so in the base than in the
proposed harbor configurations.
Ammonia concentrations in the alternative evaluated were very similar.
There was some variation at stations closest to the tributary sources, but
elsewhere, differences were indistinguishable, Figure 5-45. This is mainly
due to the very low levels of ammonia in the system.
Nitrate concentrations in the different alternatives showed slightly more
variation than other constituents modeled, Figure 5-46. This is due to the
differing boundary conditions on the two major tributaries. However,
differences were minimal in the model and possibly not detectible in the
actual system.
Dissolved inorganic phosphorus results for the proposed harbor
alternatives were similar, Figure 5-47. There were some lags and
deviations at some stations (A5, A6, and S5) early in the simulations
during the period of the initial flush. However, after that point all proposed
harbor configuration had near identical values. The base case results
were different though. Much longer periods were required for
concentrations to decrease at all stations. This can be taken as an
indication of the limited level of flushing and mixing in the current system
in comparison to the proposed alternatives evaluated.
6. ESTIMATED COSTS AND OTHER CONSIDERATIONS
6.1. Construction Costs
Cost estimates for the conceptual alternatives were prepared by the Honolulu
District to enable comparison between the various conceptual features. A
contingency of 20% was assumed for all cost estimate line items based on
the uncertainty of the input data. Assumptions used in the preparation of the
cost estimates included that the rock for new breakwater construction would
come from the existing breakwater and that all excess rock and dredged
material would be disposed of in an approved offshore disposal site.
102
6.1.1. Alternative 1
For Alternative 1, the approximate amount of material to be removed from
the existing breakwater through demolition of 7,500 feet of the structure
was assumed to be 620,000 tons. At a unit price for $42/ton, breakwater
removal would cost about $31,248,000. Construction of approximately
2,000 feet of new breakwater would require placement of approximately
210,000 tons of material at $72/ton for a cost of $18,144,000. Mobilization
and demobilization costs (including preparation of the work road and
staging area) would be $2,400,000. Dredging requirements would be
approximately 780,000 cubic yards at a unit price of $14 for a cost of
$13,104,000. The total cost of Alternative 1 would therefore be
$66,096,000.
6.1.2. Alternative 2
For Alternative 2, the approximate amount of material to be removed from
the existing breakwater to create the six notches was assumed to be
125,000 tons. At a unit price for $42/ton, breakwater notching would cost
about $6,300,000. Once the notches are made in the breakwater, each of
the 12 ends would need to be reshaped at a unit price of $180,000 and a
cost of $2,592,000. Mobilization and demobilization costs would be
$2,400,000. Dredging requirements would be approximately 780,000
cubic yards at a unit price of $14 for a cost of $13,104,000. The total cost
of Alternative 2 would therefore be $24,396,000.
6.1.3. Alternative 3
For Alternative 3, the approximate amount of material to be removed from
the existing breakwater to create the six notches was assumed to be
125,000 tons. At a unit price for $42/ton, breakwater notching would cost
about $6,300,000. Once the notches are made in the breakwater, each of
the 12 ends would need to be reshaped at a unit price of $180,000 and a
cost of $2,592,000. The six offset detached breakwaters would require
placement of approximately 180,000 tons of material at $156/ton for a cost
of $33,696,000. Mobilization and demobilization costs would be
$2,400,000. Dredging requirements would be approximately 780,000
cubic yards at a unit price of $14 for a cost of $13,104,000. The total cost
of Alternative 3 would therefore be $58,092,000.
6.1.4. Alternative 4
For Alternative 4, the approximate amount of material to be removed from
the existing breakwater to create the 500-foot gap along the root of the
structure was assumed to be 42,000 tons. At a unit price for $42/ton,
breakwater notching would cost about $2,117,000. Once the gap is made
103
in the breakwater, each of ends would need to be reshaped at a unit price
of $180,000 and a cost with contingency of $432,000. Mobilization and
demobilization costs would be $2,400,000. Dredging requirements would
be approximately 780,000 cubic yards at a unit price of $14 for a cost of
$13,104,000. The total cost of Alternative 4 would therefore be
$18,053,000.
6.1.5. Alternative 5
For Alternative 5, the approximate amount of material to be removed from
the existing breakwater to create the 500-foot gap along the root of the
structure was assumed to be 42,000 tons. At a unit price for $42/ton,
breakwater notching would cost about $2,117,000. Once the gap is made
in the breakwater, each of newly exposed ends of the structure would
need to be reshaped at a unit price of $180,000 and a cost with
contingency of $432,000. The 500-foot offset detached breakwaters
would require placement of approximately 60,000 tons of material at
$156/ton for a cost of $11,232,000. Mobilization and demobilization costs
would be $2,400,000. Dredging requirements would be approximately
780,000 cubic yards at a unit price of $14 for a cost of $13,104,000. The
total cost of Alternative 5 would therefore be $29,285,000.
6.1.6. Cost Summary
Tables 6-1, 6-2 and 6-3 provide summaries of the estimated costs for
each alternative investigated in this study. Table 6-1 displays costs for
dredging approximately 750,000 cubic yards of material as described
above for each of the alternatives. Mobilization and demobilization of
dredging equipment is estimated at $1,000,000 with actual dredging and
dredged material disposal costs of $10,920,000. With a contingency of
20%, the total cost of dredging is estimated at $14,304,000 for each
alternative. Table 6-2 shows the cost of mobilization, demobilization,
demolition of specific portions of the existing breakwater, reshaping of
exposed breakwater ends and construction of detached breakwaters
where applicable for each alternative. Mobilization and demobilization of
breakwater construction equipment is estimated at $1,000,000. A
contingency of 20% is assumed throughout. Breakwater costs for the five
alternatives range from $3,749,000 for Alternative 4 to $51,792,000 for
Alternative 1. Total estimated cost of dredging and breakwater work for
each alternative is provided in Table 6-3.
104
Table 6-1. Estimated dredging cost for each alternative
Alternative Mob/Demob Dredging Breakwater Contingency
(20%) Total
1 $1,000,000 $10,920,000 $0 $2,384,000 $14,304,000
2 $1,000,000 $10,920,000 $0 $2,384,000 $14,304,000
3 $1,000,000 $10,920,000 $0 $2,384,000 $14,304,000
4 $1,000,000 $10,920,000 $0 $2,384,000 $14,304,000
5 $1,000,000 $10,920,000 $0 $2,384,000 $14,304,000
Table 6-2. Estimated breakwater cost for each alternative
Alternative Mob/Demob Dredging Breakwater Contingency
(20%) Total
1 $1,000,000 $0 $42,160,000 $8,632,000 $51,792,000
2 $1,000,000 $0 $7,410,000 $1,682,000 $10,092,000
3 $1,000,000 $0 $35,490,000 $7,298,000 $43,788,000
4 $1,000,000 $0 $2,124,000 $625,000 $3,749,000
5 $1,000,000 $0 $11,484,000 $2,497,000 $14,981,000
Table 6-3. Estimated total cost for each alternative
Alternative Mob/Demob Dredging Breakwater Contingency
(20%) Total
1 $2,000,000 $10,920,000 $42,160,000 $11,016,000 $66,096,000
2 $2,000,000 $10,920,000 $7,410,000 $4,066,000 $24,396,000
3 $2,000,000 $10,920,000 $35,490,000 $9,682,000 $58,092,000
4 $2,000,000 $10,920,000 $2,124,000 $3,009,000 $18,053,000
5 $2,000,000 $10,920,000 $11,484,000 $4,881,000 $29,285,000
6.2. Other Considerations
Other considerations that should be evaluated prior to detailed planning of
any breakwater modifications include the effect on the Hilo Bay shoreline, the
changes to breakwater access, and the impact to Blonde Reef. The
evaluation of these additional impacts was not included in the scope of this
report.
105
The increase in wave energy inside Hilo Harbor that will occur with any of the
proposed alternatives may result in increased wave height and/or current
speed at the shoreline, depending on a simultaneous increase in water level.
This could potentially increase existing rates of sediment transport and
exacerbate any current erosion problems along the sandy shoreline and in
the littoral zone. The magnitude of these effects will vary by alternative. A
thorough review of existing sediment studies of the area as well as an
evaluation of changes to sediment transport rates and patterns for proposed
breakwater modification is suggested prior to implementation of any structural
changes.
In addition, access to the breakwater will be affected by any alternative that
places gaps in the structure (Alternatives 2, 3, 4, and 5). Access to the trunk
and head of the breakwater for inspection, repair, or for recreational uses will
not be feasible on foot or by vehicle following implementation of these
alternatives. Access to these areas will be limited to an approach by boat.
The impacts of this result on both the maintenance of the structure (USACE
responsibility) as well as the ability of the public to use the structure for
recreational purposes such as fishing should be considered and coordinated
with the appropriate Federal and state agencies including USACE, and/or
community groups.
Several of the proposed modifications also involve construction of new
rubble-mound structures adjacent to the existing breakwater (Alternatives 1,
3, and 5). This new construction would result in adding to the structure
“footprint”, which may have impacts to Blonde Reef as well as generating
other environmental concerns to plant/marine species, fish habitat, etc.
Consultation with Federal, State and County environmental agencies as well
as community groups should be undertaken early in the planning stages of
any of these alternatives.
7. SUMMARY OF RESULTS AND RECOMMENDATIONS FOR FUTURE
WORK
7.1. Effects on Waves in Hilo Harbor
The wave transformation modeling completed for all five proposed breakwater
alternatives under various wave conditions indicates increases in wave
energy within the harbor will occur to varying degrees following their
implementation. Alternative 1 appears to have the most drastic effect
including significant increases in wave energy within the navigation channel
(from 1.0 to 1.65 meters) and at the shoreline, making it the least desirable
option, regardless of the improvements it may make to water quality within the
harbor.
106
Alternatives 2 and 4 involve placing unprotected gaps in the breakwater.
These alternatives allow waves to propagate through the gaps into the
harbor, with some energy dissipation due to depth-limited wave breaking
across Blonde Reef before entering the gaps, and diffraction due to the
sheltering effect of the adjacent breakwater structure. These alternatives
result in a maximum wave increase of 0.34 and 0.30 meters in the navigation
channel for the typical wave condition, respectively, which may be considered
suitable for maintaining safe navigation. Due to the greater number of gaps in
Alternative 2 and their location along the outer portion of the breakwater, this
alternative allows more wave energy to advance to the bay shoreline than in
Alternative 4, which could have long-term effects on this area. Based on the
observed relative changes to waves within the harbor under typical
conditions, these alternatives warrant further consideration in combination
with their cost and minor increase in wave energy within the harbor.
Alternatives 3 and 5 include the excavation of gaps within the existing
breakwater, in addition to supplemental wave protection with the construction
of interior or exterior detached breakwaters. These additional structures
provide added wave energy reduction due to their effect on direct wave
breaking as well as diffraction of waves around the detached breakwaters.
The result is a maximum wave height increase in the navigation channel of
approximately 0.1 meters for both alternatives under typical wave conditions.
This is likely an acceptable increase for maintaining safe navigation. Both
alternatives appear to cause minimal increases to wave height at the
shoreline. Based on the observed relative changes to waves within the
harbor under typical conditions, these alternatives warrant further
consideration in combination with their cost and minor increase in wave
energy within the harbor.
7.2. Effects on Hilo Bay Flushing and Water Quality
Water quality data recently collected by Wiegner and Mead (2009) indicate
that the water within the bay is not critically impaired. Nutrients tended to be
low during the period sampled and modeled. At the same time dissolved
oxygen levels were high, near saturation in many instances. There were
occurrences of stratification in the observed data. It appears that this
stratification is the result of the creation of a freshwater lens via tributary flow
into the harbor. This is a three-dimensional process which was not captured
with the two dimensional depth average water quality model used in the
present study. All alternative configurations under consideration caused
increased flushing which resulted in ocean waters entering the harbor at a
much greater rate than is currently possible. Since the ocean waters are of a
higher quality than the waters in the harbor, this results in improved water
quality. Results for the four harbor configurations modeled (which excluded
Alternative 1) indicate that they would result in significant positive impacts to
the water quality of Hilo Harbor.
107
7.3. Summary Matrix
The following table summarizes the principal results of the wave, circulation,
and water quality modeling completed, as well as the estimated costs for
each conceptual alternative. This matrix is intended as a brief synopsis of the
primary decision criteria that have been evaluated in this report, to facilitate
selection of alternatives for further study. The areas shaded in red represent
the least desirable results for each evaluation criterion, the yellow areas
represent marginal results, and green areas show the most desirable results
for that decision criterion. Navigation impacts are represented as the
maximum wave height increase (in meters) observed in the navigation
channel for the typical tradewind wave condition. Water quality improvements
are shown as a qualitative evaluation of whether improvement would occur.
Finally, breakwater costs are presented for each alternative as the estimated
cost for the structural changes to the breakwater only (no dredging).
Table 7-1. Summary of Alternatives
Alternative Navigation
Impacts (meters)Water Quality Breakwater Costs
1 > 1.00 Improved $51,792,000
2 0.34 Improved $10,092,000
3 0.10 Improved $43,788,000
4 0.30 Improved $3,749,000
5 0.10 Improved $14,981,000
Overall, water quality model predictions indicated little difference in the results
for any of the proposed harbor alternatives. At some locations there were
differences in some constituent values such as particulate organic carbon.
However, these differences appear to be due to phasing in the model
response to the circulation and were relatively small and short lived. As
discussed in the following section, three-dimensional water quality modeling
will be required to quantify proposed alternative performance variability.
7.4. Recommendations for Future Work
The next phase of study would include detailed three dimensional water
quality modeling. The existing system exhibits three-dimensional behavior
which is not captured in this modeling effort. A review of field measures of
the temperature and salinity within Hilo Bay (M & E Pacific (1980), Dudley
(1991), Wiegner and Mead (2009)) reveals that strong and persistent vertical
and horizontal gradients exit. Although the use of a depth-averaged two
dimensional hydrodynamic model is justified for comparing bay flushing
characteristics for the five alternative breakwater configurations, the
108
prediction of bay-wide circulation in support of a comprehensive water quality
study will require three dimensional baroclinic hydrodynamic modeling.
The effects of changing the circulation in this system should be evaluated
using a three-dimensional model so that the impacts upon the water column
structure can be assessed. Water quality modeling results indicate that there
are currently locations of poorer circulation, which may be depositional zones
inside the harbor. If these truly are depositional zones, which accumulate
organic matter over time, it is likely that they exert a localized impact upon
water quality via increased sediment oxygen demand. These issues could be
better addressed with a three dimensional water quality model utilizing
sediment diagenesis.
An issue not addressed in this study is the issue of microbiological
contamination. From this study it is evident that opening the breakwater will
have a positive impact upon conditions in the harbor. Flushing should help
address issues related to microbial contamination. To fully ascertain the
degree to which this will occur, the source or sources of suspected
contamination need to be identified. With this information it would be possible
to discern the relative impacts of one harbor configuration over another in
terms of microorganisms in the water.
Numerical modeling to evaluate the effects of breakwater modifications on
sediment transport along the Hilo Bay shoreline is also recommended.
Sediment transport modeling linked with the three-dimensional circulation
model would allow a quantification of shoreline response from the proposed
alternatives being considered. The modeling would provide predictions of
erosion or accretion in addition to changes in sediment transport pathways
and rates.
Water quality in Hilo Harbor and Hilo Bay is dependent on several interrelated
environmental processes, which include the effects of the breakwater, as
detailed in this report. Another major contributor to the water quality in Hilo
Bay is the input of pollutants and organic materials from the Hilo Bay
watershed via surface water, ground water, and storm water runoff. In order
to comprehensively evaluate the bay’s water quality and possible methods for
improvement, these sources of contaminants must also be included in an
overall watershed study that encompasses the ancient Hawaiian ahupua’a
concept of “mountain to the sea” stewardship. This approach has been
initiated and led by the Hilo Bay Watershed Advisory Group and Dr. Tracy
Wiegner at UHH, and should be continued with a more detailed evaluation of
breakwater modifications and their effect on water quality included as an
integral component of the study.
109
8. REFERENCES
Bouws, E., Gunther, H., Rosenthal, W., and Vincent, C.L. (1985). “Similarity of
the Wind Wave Spectrum in Finite Depth Water 1 Spectral Form,” Journal of
Geophysical Research 1 (C1), 975 - 986.
Bunch, B. W., Channel, M., Corson, W. D., Ebersole, B. A., Lin, L., Mark, D. J.,
McKinney, J. P., Pranger, S. A., Schroeder, P. R., Smith, S. J., Tillman, D. H.,
Tracy, B. A., Tubman, M. W., and Welp, T. L. (2003). “Evaluation of Island
and Nearshore Confined Disposal Facility Alternatives, Pascagoula River
Harbor Dredged Material Management Plan,” Technical Report ERDC-03-3,
U.S. Army Engineer Research and Development Center, Vicksburg, MS.
Bunch, B. W., Tillman D. H., and Mark, D. J. (in preparation). “Port of Los
Angeles (POLA) Main Channel Deepening Water Quality and Hydrodynamic
Modeling Study,” Technical Report EL-00-XXX, U.S. Army Engineer Research
and Development Center, Vicksburg, MS.
Bunch, B., Cerco, C., Dortch, M., Johnson, B., and Kim, K. (2000). “Hydrodynamic
and Water Quality Model Study of San Juan Bay and Estuary,” ERDC TR-00-
1, U.S. Army Engineer Research and Development Center, Vicksburg MS.
Cerco, C. and Bunch, B, (1997). “Passaic River Tunnel Diversion Model Study,
Report 5, Water Quality Modeling,” Technical Report HL-96-2, U.S. Army
Engineer Waterways Experiment Station, Vicksburg, MS.
Cerco, C. and Cole, T. (1994). “Three-Dimensional Eutrophication Model of
Chesapeake Bay,” Technical Report EL-94-4, U.S. Army Engineer Waterways
Experiment Station, Vicksburg, MS.
Cerco, C., Bunch, B., Cialone, M., and Wang, H. (1994). “Hydrodynamic and
Eutrophication Model,” Technical Report EL-94-4, U.S. Army Engineer
Waterways Experiment Station, Vicksburg, MS.
Chapman, R. S., Johnson, B. H., and Vemulakonda, S.R. (1996). “Users Guide
for the Sigma Stretched Version of CH3D-WES; A Three-Dimensional
Numerical Hydrodynamic, Salinity and Temperature Model,” Technical Report
HL-96-21, U.S. Army Engineer Waterways Experiment Station, Vicksburg, MS.
Cialone, M. A., Brown, M. E., Smith, J. M., and Hathaway, K.K. (2008).
“Southeast Oahu Coastal Hydrodynamic Modeling with ADCIRC and
STWAVE,” Technical Report TR-08-9, U.S. Army Engineer Research and
Development Center, Vicksburg MS.
Cox, A.T., Greenwood, J.A., Cardone, V.J., and Swail, V.R. (1995). “An
Interactive Objective Kinematic Analysis System,” 4th International Workshop
110
on Wave Hindcasting and Forecasting. October 16-20, 1995, Bannf, Alberta,
Canada.
Dudley, W.C. Jr and L. E. Hallacher. (1991). “Distribution and Dispersion of
Sewerage Pollution in Hilo Bay and Contiguous Waters, Final report,” County
of Hawai`i Department of Public Works.
Eiker, E. E. (1977). “Heat Exchange Programs, Thermal Simulation of Lakes,
User’s Manual,” U. S. Army Engineer District, Baltimore, Baltimore, MD.
Gravens, M. B., Kraus, N. C., and Hanson, H. (1991). “GENESIS: Generalized
Model for Simulating Shoreline Change. Report 2, Workbook and System
User’s Manual,” Technical Report CERC-89-18, U.S. Army Engineer
Waterways Experiment Station, Vicksburg, MS.
Johnson, B. H., Heath, R. E., Hsieh, B. B., Kim, K. W., and H. L. Butler. (1991).
“Users Guide for a Three-dimensional Numerical Hydrodynamic, Salinity and
Temperature Model of Chesapeake Bay,” Technical Report HL-91-20, U.S.
Army Engineer Waterways Experiment Station, Vicksburg, MS.
LeProvost, C. M., L. Genco, F. Lyard, P. Vincent, and P. Canceil. (1994).
Spectroscopy of the World Ocean Tides from a Finite Element
Hydrodynamical Model. Journal of Geophysical Research 99(C12):24777–
24797.
Luettich, R. A., Jr., Westerink, J. J., and N. W. Scheffner. (1992). “ADCIRC: An
Advanced Three-Dimensional Circulation Model for Shelves, Coasts, and
Estuaries,” Technical Report DRP-92-6, U.S. Army Engineer Waterways
Experiment Station, Vicksburg, MS.
M&E Pacific, Inc. (1977). “Hilo Harbor First Spring Season Environmental Study.”
Prepared for the USACE, Honolulu District.
M&E Pacific, Inc. (1980). “Geological, Biological, and Water Quality
Investigations of Hilo Bay.” Prepared for the USACE, Honolulu District.
Neighbor Island Consultants (1973). “Baseline Environmental Investigations of
Hilo Harbor.” Prepared for the USACE, Honolulu District.
Public Health Service (1963). “A Study of Dispersion in Hilo Bay, Hawaii.”
Sea Engineering, Inc. (1981). “Circulation and Sediment Transport Study, Hilo
Bayfront Beach, Hilo, Hawaii.” Prepared for the USACE, Honolulu District.
111
Sheng, P. Y. (1986). “A Three-Dimensional Mathematical Model of Coastal,
Estuarine and Lake Currents Using Boundary Fitted Grid,” Report No. 585,
A.R.A.P. Group of Titan Systems, New Jersey, Princeton, NJ, 22.
Smith, J. M., Resio, D. T., and Zundel, A. K. (1999). “STWAVE: Steady-State
Spectral Wave Model. Report 1, User’s Manual for STWAVE Version 2.0,”
Instructional Report CHL-99-1, U.S. Army Engineer Waterways Experiment
Station, Vicksburg, MS.
Thompson, E. F. and Scheffner, N. W. (2004). “Typhoon-Induced Stage-
Frequency Relationships for the Island of Rota, Commonwealth of the
Northern Mariana Islands,” Technical Report TR-04-1, U.S. Army Engineer
Research and Development Center, Vicksburg MS.
Wiegner, T. and Mead, L., (2009). “Water Quality in Hilo Bay Hawai’I under
Baseflow and Storm Conditions,” Project Report Preliminary Analysis, Marine
Science Department, University of Hawai’i at Hilo.