APPLICATION OF A HYDRODYNAMIC WATER QUALITY MODEL
IN SUNGAI JOHOR ESTUARY
MAZNAH BINTI ISMAIL
A thesis submitted in fulfillment of the
requirements for the award of the degree of
Master of Engineering (Hydraulics and Hydrology)
Faculty of Civil Engineering
Universiti Teknologi Malaysia
JULY 2009
iii
ALHAMDULILLAH
All Praise To Allah, Creator Of This Universe
Thanks For The Precious Iman & Islam You Blessed On Me
Thanks For All The Strength And Knowledge You Granted On Me
And Also Peace Be Upon The Holy Prophet Muhammad
All This Hardship I Dedicate To Some Special People In My Life
My Caring & Loving Parents, Ismail Ali & Jamilah Ibrahim
My Siblings Robiah, Mazlan, Mohd. Yazid and Radzuran
Endless Appreciation On All Sacrifices You Did For Me
My Caring Lecturer & Supervisor, Dr. Noor Baharim
Ever Loving and Caring Friends
I love all of you dearly
Thanks
iv
ACKNOWLEDGEMENTS
First and foremost I would like to express my thanks to Almighty ALLAH on
the successful completion of this research work and thesis. I would like to thank my
supervisor Dr. Noor Baharim bin Hashim for his sincere advice and guidance
provided throughout my studies. His trust, patience, knowledge and friendly
personality has always been an inspiration for me and will deeply influence my
career and future life. The technical support and help of Paul M. Craig of the
Dynamic Solutions, LLC and Professor James L. Martin of Mississippi State
University for their guidance in numerical model is deeply appreciated. Special
thanks and appreciation is extended to Noor Hasbullah Hashim, Anie Raflikha Abdul
Malek, Mohd Zaharifudin Muhamad Ali, Juwita Asfar, Norazlina Bateni, and
Nuryazmeen Farhan Haron and other undergraduate students for their invaluable
assistance on data gathering and field data collection during the intensive survey.
Appreciation is extended to the staff of the following agencies; Department of
Environmental (DOE) Malaysia for the marine and river water quality data of Sungai
Johor, Department of Irrigation and Drainage (DID) for the hydrology data,
Malaysian Meteorological Service (MMS) for the meteorological data, Department
of Chemistry, Faculty of Science, UTM for the laboratory data analysis, and
Department of Hydraulics and Hydrology, Faculty of Civil Engineering, UTM for
the hydrological field data collection. Fund for this study was provided through a
grant from the Ministry of Science, Technology and Innovation (MOSTI) Malaysia
under VOT 79003 and supported by Research Management Center (RMC) UTM.
A very special gratitude is reserved for my family; my parents Ismail Ali and
Jamilah Ibrahim for their kindness and support especially motivate me during
completion of my thesis. My appreciation also goes to my siblings Robiah, Mazlan,
Mohd Yazid and Radzuran for their continuous warm encouragement and support
given. Special thank to my friends Siti Nurhazwani Abdul Kahar and Marina Md
Arshad for their understanding and support.
v
ABSTRACT
Sungai Johor estuary is a vital water body in the south of Johor and greatly
affects the water quality in the Johor Straits. In the development of the hydrodynamic
and water quality models for Sungai Johor estuary, the Environmental Fluid
Dynamics Code (EFDC) model was selected. In this application, the EFDC
hydrodynamic model was configured to simulate time varying surface elevation,
velocity, salinity, and water temperature. The EFDC water quality model was
configured to simulate dissolved oxygen (DO), dissolved organic carbon (DOC),
chemical oxygen demand (COD), ammoniacal nitrogen (NH3-N), nitrate nitrogen
(NO3-N), phosphate (PO4), and Chlorophyll a. The hydrodynamic and water quality
model calibration was performed utilizing a set of site specific data acquired in
January 2008. The simulated water temperature, salinity and DO showed good and
fairly good agreement with observations. The calculated correlation coefficients
between computed and observed temperature and salinity were lower compared with
the water level. Sensitivity analysis was performed on hydrodynamic and water
quality models input parameters to quantify their impact on modeling results such as
water surface elevation, salinity and dissolved oxygen concentration. It is anticipated
and recommended that the development of this model be continued to synthesize
additional field data into the modeling process.
vi
ABSTRAK
Muara Sungai Johor merupakan salah satu sungai yang penting di selatan
Johor dan sangat mempengaruhi kualiti air di Selat Johor. Dalam pembangunan
model hidrodinamik dan kualiti air bagi sistem muara Sungai Johor, model
Environmental Fluid Dynamic Code (EFDC) telah dipilih. Dalam aplikasi model
hidrodinamik EFDC ini, simulasi ketinggian permukaan, halaju, dan kemasinan air
dapat diukur. Model kualiti air EFDC pula dapat mengukur beberapa parameter
seperti oksigen terlarut (DO), karbon organik terlarut (DOC), permintaan oksigen
kimia (COD), nitrogen ammonia (NH3-N), nitrogen nitrat (NO3-N), fosfat (PO4), dan
Chlorophyll a. Kalibrasi model hidrodinamik dan kualiti air dilakukan dengan
menggunakan set data di tapak dalam bulan Januari 2008. Simulasi suhu air,
kemasinan air dan oksigen terlarut menunjukkan hasil yang baik dengan hasil
cerapan. Pengiraan korelasi di antara model dan cerapan suhu dan kemasinan adalah
lebih rendah berbanding dengan paras air. Analisis kepekaan dijalankan terhadap
parameter masukan model hidrodinamik dan kualiti air untuk mengukur kesan
terhadap hasil permodelan seperti paras air permukaan, kemasinan dan konsentrasi
oksigen terlarut. Adalah diharapkan agar pembangunan model ini akan diteruskan
dengan mengumpulkan lebih banyak data lapangan dalam proses permodelan.
vii
TABLE OF CONTENTS
CHAPTER
1
2
TITLE
PAGE
AUTHOR’S DECLARATION
ii
DEDICATION
iii
ACKNOWLEDGEMENTS
iv
ABSTRACT
v
ABSTRAK
vi
TABLE OF CONTENTS
vii
LIST OF TABLES
xi
LIST OF FIGURES
xiii
LIST OF ABBREVIATIONS
xvi
LIST OF SYMBOLS
xviii
LIST OF APPENDIXES
xx
INTRODUCTION
1
1.1
Introduction
1
1.2
Problem Statement
3
1.3
Objectives
6
1.4
Scope of Study
7
1.5
Significance of Research
7
LITERATURE REVIEW
9
2.1
Introduction
9
2.2
Overview of Water Quality Modeling in Malaysia
10
2.3
Estuarine Model
12
2.4
Review of Available Hydrodynamic Models
15
2.5
Review of Available Water Quality Models
19
viii
2.6
2.7
3
4
Environmental Fluid Dynamic Code (EFDC)
22
2.6.1 Summary of EFDC Theory
23
2.6.2 Modeling Assumption
26
EFDC Water Quality Model
27
2.7.1 Water Quality Model Formulation
28
RESEARCH METHODOLOGY
30
3.1
Introduction
30
3.2
Database
32
3.2.1 Geographical Information System (GIS)
33
3.3
Equipments
34
3.4
Model Set Up
36
3.4.1 Model Domain
37
3.4.2 Bathymetry
37
3.4.3 Model Boundary Conditions
39
3.4.4 Model Initial Conditions
39
3.4.5 Model Kinetics
41
3.4.6 Model Input Parameters
42
3.5
Model Calibration
43
3.6
Model Sensitivity Analysis
44
3.7
Statistical Comparison Techniques
45
DATA COLLECTION & ANALYSIS
4.1
Stream Data Assessment
48
48
4.2
Meteorological Data
49
4.3
Freshwater Flow
49
4.4
DOE Water Quality Monitoring Data
50
4.5
Intensive Survey
57
4.6
Discrete Water Chemistry Samples
65
4.7
Continuous Monitoring Data
67
4.8
Vertical Profiles
68
ix
5
RESULTS & DISCUSSIONS
5.0
Introduction
72
72
5.1
Hydrodynamic Model Calibration
73
5.1.1 Water Surface Elevation Calibration
74
5.1.2 Temperature Calibration
76
5.1.3 Salinity Calibration
79
Water Quality Model Calibration
81
5.2.1 Dissolved Oxygen
82
5.2.2 Ammonia Nitrogen
84
5.2.3 Nitrate Nitrogen
85
5.2.6 Total Phosphate
85
5.2.7 Chlorophyll a
86
Hydrodynamic Model Sensitivity Analysis
87
5.2
5.3
5.3.1 Sensitivity of Water Surface Elevation to
Bottom Roughness
87
5.3.2 Sensitivity of Salinity to Bottom Roughness 89
5.3.3 Sensitivity of Salinity to Horizontal Eddy
Viscosity
89
5.3.4 Sensitivity of Salinity to Upstream
Boundary Condition
91
5.3.5 Sensitivity of Salinity to Downstream
Salinity Boundary
5.4
91
Water Quality Model Sensitivity Analysis
92
5.4.1 Sensitivity of DO to COD Decay Rate
93
5.4.2 Sensitivity of DO to Nitrification Rate
95
5.4.3 Sensitivity of DO to SOD
96
5.4.4 Sensitivity of DO to Maximum Algal
Growth Rate
99
5.4.5 Sensitivity of DO to Loads from
Point Sources
5.4.6 Sensitivity of DO to reaeration rate
99
100
x
6
CONCLUSIONS & RECOMMENDATIONS
6.1
Conclusions
101
101
6.2
102
Recommendations
REFERENCES
104
APPENDICES
114
xi
LIST OF TABLES
TABLE NO.
TITLE
PAGE
2.1
Available estuary model
14
2.2
Evaluation of capability for hydrodynamic models
18
2.3
Evaluation of capability for water quality models
21
3.1
Data Sources of Sungai Johor Estuary Model
32
3.2
Historical Water Quality and Quantity Data for Sungai
Johor Estuary
33
3.3
Initial Conditions for Sungai Johor Estuary Model
40
3.4
EFDC kinetic coefficients in the present model application
41
3.5
Files required for hydrodynamic model simulation
42
4.1
Hydraulic Data for Sungai Johor Basin
49
4.2
Meteorological Data for Sungai Johor Basin
49
4.3
Averaged flow discharges for Sungai Johor Estuary Model
50
4.4
Water Quality Index (WQI)
52
4.5
DOE Water Quality Index Classification
53
4.6
Water Quality Parameters and Value Ranges of Sungai Johor 56
4.7
Summary of Intensive Surveys Data
58
4.8
Water Quality Parameters Data from Kota Tinggi to
Tanjung Buai
60
4.9
Summary of Continuous Monitoring Stations of Sungai Johor 67
5.1
Error analysis for observed and simulated water surface
Elevations
75
xii
5.2
Error analysis for observed and simulated temperatures
77
5.3
Error analysis for observed and simulated salinity
80
5.4
Error analysis for observed and simulated DO
82
5.5
Sensitivity analysis of water surface elevation with
bottom roughness at Tanjung Surat
88
Sensitivity analysis of water surface elevation with
bottom roughness at Teluk Sengat
88
Sensitivity analysis of water surface elevation with
bottom roughness at Kota Tinggi Bridge
88
5.6
5.7
5.8
5.9
Sensitivity analysis of salinity with bottom roughness
At Tanjung Surat
Sensitivity analysis of salinity with horizontal eddy viscosity
At Tanjung Surat
89
90
5.10
Sensitivity analysis of DO to COD decay rate at Tg Surat
94
5.11
Sensitivity analysis of DO to COD decay rate at Teluk Sengat 94
5.12
Sensitivity analysis of DO to nitrification rate at Tg Surat
5.13
Sensitivity analysis of DO to nitrification rate at Teluk Sengat 96
5.14
Sediment oxygen demand values
97
5.15
Sediment oxygen demand for Sungai Tebrau
97
5.16
Sensitivity analysis of DO to SOD at Tanjung Surat
98
95
xiii
LIST OF FIGURES
FIGURE NO.
TITLE
PAGE
1.1
Map of Sungai Johor study area
6
2.1
Primary models of the EFDC model
23
2.2
Structure of the EFDC hydrodynamic model
23
2.3
Structure of the EFDC water quality model
27
2.4
Schematic diagram for the EFDC water column water
Quality model
29
3.1
Research Methodology Flow Chart
31
3.2
Garmin GPS 72
34
3.3
YSI Water Quality Monitoring System
35
3.4
Hondex Digital Depth Sounder
35
3.5
Solinst Lovelogger
36
3.6
Curvilinear-orthogornal horizontal model domain for
the Sungai Johor
38
3.7
The model bathymetry
38
4.1
Stream flow at Rantau Panjang in year 2006
48
4.2
Location of DOE river and marine water quality stations
51
4.3
Range of WQI for selected rivers along Sungai Johor basin
54
4.4
Water Quality trends at DOE river monitoring stations
55
4.5
Trend of temperature from downstream boundary at Tanjung
Pengelih towards upstream boundary at Rantau Panjang
56
4.6
Trend of DO from downstream boundary at Tanjung
Pengelih towards upstream boundary at Rantau Panjang
57
xiv
4.7
Trend of salinity from downstream boundary at Tanjung
Pengelih towards upstream boundary at Rantau Panjang
57
4.8
Location of intensive survey stations along Sungai Johor
59
4.9
Trend of salinity from Kota Tinggi towards downstream
at Tanjung Buai
60
Trend of temperature from Kota Tinggi towards
downstream at Tanjung Buai
61
Trend of DO from Kota Tinggi towards downstream
at Tanjung Buai
61
4.12
Nitrogen and Phosphorus Loads at Rantau Panjang
63
4.13
Water quality concentration in Sungai Johor tributaries
64
4.14
Water quality concentration along Sungai Johor
66
4.15
Location of vertical profile stations
68
4.16
Vertical profiles of observed salinity, temperature and DO
69
5.1
Observed and simulated water surface elevation
at Tanjung Surat
75
Observed and simulated water surface elevation
at Teluk Sengat
75
Observed and simulated water surface elevation
at Kota Tinggi Bridge
76
5.4
Observed and simulated temperature at Tanjung Surat
77
5.5
Observed and simulated temperature at Teluk Sengat
78
5.6
Temperature profile along Sungai Johor
78
5.7
Temperature profile along Johor Straits
79
5.8
Observed and simulated salinity at Tanjung Surat
80
5.9
Salinity profile along Sungai Johor
81
5.10
Salinity profile along Johor Straits
81
5.11
Observed and simulated DO at Tanjung Surat
83
4.10
4.11
5.2
5.3
xv
5.12
DO concentration profile along Sungai Johor
83
5.13
DO concentration profile along Johor Straits
84
5.14
Ammonia nitrogen concentration profile
84
5.15
Nitrate nitrogen concentration profile
85
5.16
Total Phosphate concentration profile
86
5.17
Chlorophyll a concentration profile
87
5.18
Sensitivity of salinity to bottom roughness at Tg Surat
89
5.19
Sensitivity of salinity to horizontal eddy viscosity at
Station Tanjung Surat
90
5.20
Sensitivity of salinity to upstream boundary condition
91
5.21
Sensitivity of salinity to downstream salinity boundary
92
5.22
Sensitivity of DO to COD decay rate at Tanjung Surat
93
5.23
Sensitivity of DO to COD decay rate at Teluk Sengat
94
5.24
Sensitivity of DO to nitrification rate at Tanjung Surat
95
5.25
Sensitivity of DO to nitrification rate at Teluk Sengat
96
5.26
Sensitivity of DO to SOD at Tanjung Surat
98
5.27
Sensitivity of DO to SOD
98
5.28
Sensitivity of DO to maximum algal growth rate
99
5.29
Sensitivity of DO to loads from point sources
100
5.30
Sensitivity of DO to reaeration rate
100
xvi
LIST OF ABBREVIATIONS
AME
Absolute mean error
ASCE
American Society of Civil Engineers
BOD
Biochemical Oxygen Demand
CE-QUALICM
Three-Dimensional Eutrophication Model
CE-QUALW2
Two-Dimensional, Laterally Averaged Hydrodynamic and
Water Quality Model
CH3DWES
Curvilinear Hydrodynamics in 3-Dimensions Waterways Experiment
Station
COD
Chemical oxygen demand
DID
Department of Irrigation and Drainage
DO
Dissolved oxygen
DOC
Dissolved organic carbon
DOE
Department of Environment
DYNHYD5
Link Node Tidal Hydrodynamic Model
EFDC
Environmental Fluid Dynamic Code
EPA
Environmental Protection Agency
HABs
Harmful algal blooms
HEM-3D
Three dimensional hydrodynamic eutrophication model
INWQS
Interim National Water Quality Standards for Malaysia
ME
Mean error
N
Nitrogen
xvii
NBOD
Nitrogenous Biochemical Oxygen Demand
NH+4
Ammonium ions
NH3-N
Ammoniacal Nitrogen
NO-3
Nitrate ions
NO3-N
Nitrate Nitrogen
P
Phosphorus
PAHs
Polycyclic Aromatic Hydrocarbons
PCBs
Polychlorinated Biophenyls
PO4
Inorganic Phosphorus
PO4-3
Inorganic Phosphate ions
POM
Princeton Ocean Model
QUAL-2E
Enhanced Stream Water Quality Model
RIVMOD-H River Hydrodynamic Model
RMSE
Root mean square error
SOD
Sediment oxygen demand
SS
Suspended Solids
TMDLs
Total Maximum Daily Loads
TOC
Total Organic Carbon
USACOE
United State Army Corps of Engineer’s
UTM
Universiti Teknologi Malaysia
WASP5
Water Quality Simulation Program
WQI
Water Quality Index
xviii
LIST OF SYMBOLS
u
=
velocity component in the x-direction
v
=
velocity component in the y-direction
w
=
velocity component in the z-direction
x, y
=
curvilinear horizontal coordinates
Kx
=
turbulent diffusivities in the x-direction
Ky
=
turbulent diffusivities in the y-direction
Kz
=
turbulent diffusivities in the z-direction
SC
=
internal and external sources and sinks per unit volume
κ
=
von Karman constant
∆1
=
dimensionless thickness of the bottom layer
zo
=
dimensionless roughness height
b
=
buoyancy
ρ
=
actual water density
ρo
=
reference water density
ρa
=
air density
ρw
=
water density
cs
=
wind stress coefficient
cb
=
bottom drag coefficient
Uw, Vw =
component of the wind velocity at 10 meter above the water surface
AH
=
horizontal turbulence mass diffusion coefficient
Ab
=
vertical turbulence mass diffusion coefficient
Rc
=
physical and biogeochemical sources and sinks
QH
=
volume sources and sinks including rainfall, evaporation, infiltration,
and lateral inflows and outflows having negligible momentum fluxes
mx , my =
scale factors of the horizontal coordinates
w
vertical velocity in the stretched vertical coordinate
=
xix
z s*
=
physical vertical coordinate of the free surface
z b*
=
physical vertical coordinate of the bottom bed
H
=
total water column depth
φ
=
free surface potential
Fe
=
effective Coriolis acceleration
Q
=
horizontal momentum diffusion terms
Av
=
vertical turbulent viscosity
ρatm
=
atmospheric pressure
u*
=
shear velocity
cb
=
bottom stress coefficient
U
=
flow velocity at the bottom layer
Aν
=
viscosity
Kν
=
turbulent diffusivity
Rq
=
turbulent intensity Richardson Number
xx
LIST OF APPENDICES
APPENDIX.
TITLE
PAGE
A.1
DO concentration profiles along Sungai Johor
114
A.2
Salinity profiles along Sungai Johor
115
A.3
Temperature profiles along Sungai Johor
116
B
Methods of Analysis
117
C
Temporal Profiles of Stream Flow
118
D
Temporal Profiles of DO Concentrations and
Nutrient Loads
119
CHAPTER 1
INTRODUCTION
1.1
Introduction
An estuary is a semi-enclosed coastal body of water with one or more rivers
or streams flowing into it, and with a free connection to the open sea (Pritchard,
1967). They are affected by both marine influences, such as tides, waves, and the
influx of saline water; and riverine influences, such as flows of fresh water and
sediment. As a result they may contain many biological niches within a small area,
and so are associated with high biological diversity.
Estuaries face a host of common challenges. As more people flock to the
shore, we are upsetting the natural balance of estuaries and treating their health. We
endanger our estuaries by polluting the water and building on the lands surrounding
them. These activities can contribute to unsafe drinking water and beach, closing of
shellfish bed, harmful algal blooms, declines in fisheries, loss of habitat, fish kills,
and a host of other human health and natural resource problems.
Although each of the estuaries in our country, Malaysia is unique, all of them
face similar environmental problems and challenges, such as over enrichment of
nutrients, contamination of pathogen, toxic chemicals, alteration of freshwater
inflow, loss of habitat, declines in fish and wildlife, and introduction of invasive
species. While no regional or national conclusions can be drawn about the overall
health of estuaries in Malaysia, these problems tend to cause declines in water
quality, living resources, and overall ecosystem health.
2
With regard to tidal estuaries and given specific input, such as stream flows,
engineers and hydrologists use hydrodynamic models to predict outputs, such as
water surface elevations and velocities.
In addition, given hydrodynamics and
contaminant loading rates, they make use of water quality models to predict
contaminant concentrations. The hydrodynamic and water quality models utilization
consist of a detailed set of equations that serve to represent complex physical
processes. However, as the number of required equations to describe the processes
in question increases, the computational time and model complexity also increases.
As a result, numerical models have been developed to aid in the solution of complex
process equations.
Computer models for simulating estuary hydrodynamics and water quality
have existed for more than 40 years (Chapra, 1997). Significant improvements have
been made to the original computer models, therefore, improving the quality of
model outputs. Today’s computer models allow users to simulate in one, two, and
three dimensions. In addition, they enable users to model water bodies that are either
in steady state or dynamic systems. Model solution techniques have also been
improved and the two most commonly employed are finite differences or finite
elements. As a result of these improvements, users are now applying computer
models to larger and larger estuary systems, in order to estimate hydrodynamics and,
more importantly, water quality.
In order to advance the process of improving water quality assessment
techniques within an estuarine system, a computer model will be necessary to
estimate river and estuarine hydrodynamics and contaminant concentrations (Liu et
al., 2008). However, since the model study area is tidally influenced, the model
utilized will have to take into consideration the impacts of the flood and ebb tides
that occur multiple times a day.
In addition, the model should be fairly
straightforward in terms of setup, and the computational time should not be excessive
while using a current desktop computer. The model should be able to simulate a
dynamic system, and it should employ a robust numerical solution technique, such as
finite elements or finite differences.
3
A
multifunction
surface
water
modeling
system,
which
includes
hydrodynamic, sediment-contaminant and water quality models, was developed and
applied for Sungai Johor estuary. The hydrodynamic model, based on the principles
of conservation of volume, momentum and mass, predicts surface elevation, current
velocity and salinity. The water quality model, based on the conservation of mass
balance, predicts seven parameters, such as dissolved oxygen, COD, DOC, ammonia
nitrogen, nitrate nitrogen, total phosphate and Chlorophyll a. The model equations
were solved using finite difference scheme. Model parameters were estimated from
existing and collected datasets during initial setup. A calibration time period was
used to modify and refine the model parameters.
1.2
Problem Statement
River and marine water quality monitoring in Malaysia has been
conducted by Malaysian Department of Environment (DOE) since 1978,
primarily to establish the status of water quality, detect water quality changes and
identify pollution sources.
In 2007, there were a total of 1063 water quality
monitoring stations located within 143 river basins through out Malaysia. This
involves routine monitoring at predetermined stations, in-situ and laboratory
analysis, and data interpretation in terms of their physic-chemical and biological
characteristics. River water quality appraisal is based on Water Quality Index
(WQI) involving parameters such as Dissolved Oxyen (DO), Biochemical
Oxygen Demand (BOD), Chemical Oxygen Demand (COD), Ammoniacal
Nitrogen (NH3N), Suspended Solids (SS) and pH. The WQI serves as a basis for
assessment of a water course in relation to pollution categorization and designated
classes of beneficial uses in accordance with the Proposed Interim National Water
Quality Standards for Malaysia (INWQS). The DOE maintains 39 monitoring
stations for Sungai Johor river basin. DOE (2007) reported that there were a total
of 51 marine water quality monitoring stations located at Johor estuary and
coastal area.
4
Dr. Noor Baharim, lecturer in the Faculty of Civil Engineering, Universiti
Teknologi Malaysia, Skudai said a significant increase of freshwater inflow into
the estuarine areas during flash flood at Kota Tinggi has affected the aquatic life
for up to three weeks. Many of the fish die or swim to other river resulting in
dwindling catch for the local fishermen. He said the nutrient level of the water
samples collected from the Straits of Johor showed there was a risk of algal
bloom in the region. There is an increasing concern of oversupply of nutrients
from multiple sources in Sungai Johor estuary. This has ecological effects on the
shallow coastal and estuarine areas. These effects include decrease in dissolved
oxygen, impacts on living resources and loss of aquatic habitat. Degraded water
quality has adverse effects on critical habitats in Sungai Johor estuary such as
seagrass, which is an essential food for dugong and many herbivorous fish (Land
Clearing, Sand Mining Affect Rivers, 2007, April 16). Hence, there is the need
for comprehensive and intensive baseline studies of the water quality in the
Sungai Johor estuary in order to assess the impact of all existing and future
developments on the ecology of the Sungai Johor estuarine and coastal waters.
1.2.1
Description of Study Area
Johor is the second largest state in Peninsular Malaysia with an area of 18,941
km2. Sungai Johor is considered the main river in Johor. The river flows in a
roughly north-south direction and empties into the Johor Straits. The water quality of
Sungai Johor has been deteriorating with increasing levels of various pollutants.
Besides, it persists to be silted and chocked by rubbish and wastes as a consequence
of storage of enforcement by local-authorities. These contaminants eventually flow
into Sungai Johor estuary, which are rich in habitats that provide spawning and
feeding areas for fish and poultry.
The Johor estuarine system, an interconnected series of estuarine systems, is a
vital water body in the Southern Johor as the South Gateway to Peninsular Malaysia.
The estuary provides convenient and inexpensive navigation and transportation
support for the economic activities in the area.
5
The Sungai Johor river basin is located at the southeastern tip of Peninsular
Malaysia as shown in Figure 1.1. Sungai Johor has a total length of about 122.7 km
with catchment area of 2636 km2. The river originates from Gunung Belumut and
Bukit Gemuruh in the north and flows to the southeastern part of Johor and finally
into the Straits of Johor. The major tributaries of Sungai Johor are Sungai Sayong,
Sungai Linggiu, Sungai Semanggar, Sungai Lebam, Sungai Seluyut and Sungai
Tiram.
A great amount of pollutants from various sources, such as the sewerage
network of the Johor Bahru, Pasir Gudang, Ulu Tiram, and Kota Tinggi cities, the
industrial wastewaters from many industries in the surroundings, and agricultural
wastewater containing fertilizers and pesticides are discharged into the Sungai Johor.
The catchment is irregular in shape. The maximum length and breadth are 80 km an
45 km respectively. About 60% of the catchment is undulating highland rising to a
height of 366 m while the remainders are lowland and swamps. The highland in the
north is mainly jungle. In the south, a major portion had been cleared and planted
with oil palm and rubber. The highland areas have granite soil cover consisting of
fine to coarse sand and clay. The alluvium consists of fine sand and clay.
The catchment receives an average annual precipitation of 2470 mm while
the mean annual discharge measured at Rantau Panjang (1130 km2) were 37.5 m3/s
during the period 1963-1992. The temperature in the basin ranges from 21 ºC to 32
ºC. Due to the tidal influence, the Sungai Johor is an ideal study area for research on
model implementation and the prediction of hydrodynamics and water quality in
tidally influenced areas.
6
N
Rantau Panjang
W
Sg.Johor
E
Kota Tinggi
S
Sg.Seluyut
Sg.Berangan
Sg.Temon
Sg.Redan
Johor Lama
Sg.Tiram
Sg.Papan
Sg.Layang
Sg.Chemangar
Teluk Sengat
Sg.Layau
Sg.Tebrau
Sg.Serai
Tg.Serindet
Sg.Lebam
Tg.Surat
Pasir Gudang
Sg.Kim Kim
Sg.Johor
Sg.Buloh
P.Ubin
Sg.Belungkor
P.Tekong
Sg.Sebina
Sg.Santi
Tg.Pengelih
Figure 1.1: Map of Sungai Johor study area.
1.3
Objective of the Study
The primary purpose of the study is to apply EFDC hydrodynamic and water
quality model to Sungai Johor estuarine system. The objectives of the study are
listed as follows:
i.
To calibrate the model utilizing historical data and field data collection.
ii.
To do sensitivity analysis of input parameters of hydrodynamic and water
quality model.
7
1.4
Scope of the Study
The scope of this research concentrates on developing hydrodynamic and
water quality modeling tools for Sungai Johor estuarine system. These tools have
been used to minimize the cost of analysis. This study was limited to the following
scope of work to meet the specific objectives:-
i.
To collect existing data and information relevant to the study area, Sungai
Johor. The relevant data and information consist of bathymetry data,
water level data, water quality data, fresh water inflow data, wind data,
and temperature data.
ii.
To analyze the result from the hydrodynamics and water quality model
that will give a reasonable and practical results to real condition in Sungai
Johor estuarine system.
iii.
To develop hydrodynamic and water quality model that can be useful for
the further development at Sungai Johor estuarine system.
1.5
Significance of Research
Estuarine flow and DO distribution are three dimensional in nature. To
simulate these completely, a three dimensional model with time dependent
momentum and continuity equations, mass balance equations with details description
of the biochemical kinetics, and sources and sinks of all dissolved constituents are
necessary. It seems that the state of the art computer technology enables us to do
three dimensional simulations, particularly of hydrodynamics (Blumberg and Mellor,
1987). The current sampling capacity, however, cannot provide us with the quantity
and quality of field data that are indispensable for the calibration and verification of
the model, particularly the water quality model.
This hydrodynamic model was applied in the study to simulate the flow field
and salinity distribution, and the corresponding water quality model was developed
to simulate the distributions of DO and other related water quality parameters.
8
EFDC hydrodynamics and water quality modeling is applied in the Sungai
Johor estuarine system. The recommendation is based on the following arguments:
i.
EFDC model comprises an advanced three dimensional surface water modelling
system for hydrodynamic and reactive transport simulation of rivers, lakes,
reservoirs, wetland systems, estuaries and the coastal ocean.
ii.
EFDC has sufficient hydrodynamic and water quality capability to model the fate
and transport of dissolved oxygen and nutrients within tidal estuary.
iii.
EFDC was originally developed at Virginia Institute of Marine Science as part of
a long term research program to develop operational models for resource
management applications in Virginia’s estuarine and coastal waters (Hamrick,
1992).
iv.
EFDC is a public domain with current users including universities, governmental
agencies and engineering consultants.
v.
EFDC can be used for long term simulation over many tidal cycles.
9
CHAPTER 2
LITERATURE REVIEW
2.1
Introduction
Water quality problems in a tidal system generally result from a combination
of physical and biochemical processes as human activities exert stress on the system.
Therefore, water quality management in estuarine and coastal waters has received
increased attention in recent years as human activities in these areas increase (Park,
1993).
Dissolved oxygen (DO) deficiency, as an index of deteriorated water quality,
has been widely observed in estuarine and coastal waters such as Chesapeake Bay
(Officer et al., 1984) and Rappahannock Estuary in Virginia, United States of
America (Park, 1993). Hypoxia (deficient dissolved oxygen) has been observed
frequently in the deep basin of Rappahannock River in Virginia (Park, 1993).
Biological productivity in most of the world’s oceans is controlled by the
supply of nutrients to surface waters (Beman et al., 2005). They demonstrated that
nitrogen deficient areas of the tropical and subtropical oceans are acutely vulnerable
to nitrogen pollution. They used satellite imagery to demonstrate how agricultural
runoff fuels large algal blooms in vulnerable areas of the ocean in the Gulf of
California. According to Beman et al. (2005), artificially induced algal blooms could
have major impacts on recreational and commercial fishing, a major industries in the
gulf. Another concern is hypoxia, or oxygen depletion, caused by excessive algae
growth.
10
As much as these manifold processes affect the DO distribution in the water
column, it is difficult to assess the relative importance of each process. A numerical
model based on physical and biogeochemical principles is useful both to aid in
understanding the system and to provide consistent, rational predictions of dynamic
responses of the system to changes in specified factors (Park, 2003).
2.2
Overview of Water Quality Modeling in Malaysia
As stated by Ambrose et al. (1993), the water volume and water quality of
constituent masses being studied in the estuarine system are tracked and accounted
for over time and space using a series of mass balancing equations. The developed
three dimensional hydrodynamic model has also been successfully applied to
Singapore’s coastal waters (Zhang and Gin, 1998).
These areas included the
southern part of the state of Johor, the main island of Singapore and the southern
island as well as Batam Island of Indonesia.
The water quality modeling analysis of Sungai Lebam was studied by Koh
and Lee (2001). It’s a tidal estuary that drains into Sungai Johor. They studied the
impact of discharged wastes and residuals arising from the resort development which
is located in Mukim Pantai Timur Kota Tinggi Johor on the water quality of the
receiving water.
Four parameters of concern in the analysis were Biochemical
Oxygen Demand, Dissolved Oxygen, Suspended Solids and Fecal Coliforms (FC).
They demonstrated the distribution of FC along the river stretch, downstream of
project site, indicating that for some 10 km downstream of project site, the estuarine
water was unsafe for swimming and other contact sports.
The study on the behavior of salinity transport and water quality of Sungai
Lebam, which is the largest main stream of Sungai Johor estuary were studied by
Khairul Ihsan (2008) and Muhammad Hanif (2008). It is essential to determine the
origin of the freshwater, where it will be heading and where it will end up. The data
was collected at several selected points along Sungai Lebam, ranging from the most
upstream of the river to the meet point of Sungai Johor.
11
Syamsidik and Koh (2003) studied the impact assessment modeling of coastal
reclamation at Pulau Tekong, a Singaporean island. They studied a water system
consisting of several rivers, namely Sungai Johor, Sungai Lebam, Sungai Santi and
Selat Johor and 16 other tributaries such as Sungai Layau, Sungai Serai, Sungai
Khatib Bonsu, Sungai Punggol, and Sungai Seletar. Two softwares were used to
assess the possible impacts generated by these reclamation projects, namely Water
Quality Analysis Simulation Program (WASP5) and AQUASEA. They concluded
that the coastal reclamations at Pulau Tekong and Changi’s Bay will have the
potential to generate several negative impacts to the tidal hydrodynamic regimes by
significantly increasing tidal velocity around Pulau Tekong and suspended sediment
concentrations in the water column. This may lead to diverse problems such as
eutrophication, which will cause regular occurrence of algae blooms.
Lee (2006) studied the river pollution in the Sungai Johor estuary system
using EFDC hydrodynamic model. The results of velocity, salinity and sediment
profiles had been simulated over a seven days. The study of the results revealed the
great influences of the value of various pollution sources on the distribution of
pollution loads in the Sungai Johor estuary (Hashim et al.,2008).
Application of EFDC model has been used to simulate suspended sediment
dispersion at Sungai Pulai estuary (Hashim et al., 2007). They described an effort to
understand and predict the transport of sediment in Sungai Pulai estuary using EFDC
hydrodynamic model. Oversupply of sediment from multiple sources will cause
unstable conditions to rivers, lakes, and shallow coastal and estuaries areas.
Development in industrial activities along the Sungai Tiram area had an
adverse effect on its water quality. Besides that, rapid development in agricultural
sector caused the pollution and effects to the aquatic life balance at Sungai Tiram.
Application of EFDC Water Quality has successfully been developed for Sungai
Tiram by Mohd Shafiq (2008) and Roslan (2008).
Nurhidayah (2007) studied the status of Sungai Batu Pahat based on the
quantitative and qualititative of water quality and biodiversity analysis. According to
12
DOE-WQI, Sungai Batu Pahat is classified as Class III at upstream and downstream
but dropped to class IV at the middle stream.
The EFDC hydrodynamic and
sediment model was selected to perform the hydrodynamic and sediment modeling
for Sungai Batu Pahat estuary area because of its comprehensive development
(Muhamad, 2008; Mohamad Faazil, 2008).
Said et al.(2009) explored the use of a water quality simulation model via
InfoWorks River Simulation (RS) for the Maong River, which flows into the
Kuching City, Sarawak. They focused on predicting the dissolved oxygen (DO)
parameters in the river basin. The model adopted a hydrodynamic and transport flow
simulation tools to capture the advection and diffusion processes for different flow
conditions. From the results, it is clear that the model generally predicted and
identified the main impacts of flooding on the river water quality.
Ali et al. (2009) study predicted water quality parameters at Johor River basin
utilizing Artificial Neural Network (ANN) modeling. The results showed that the
proposed ANN prediction model had a great potential to simulate and predict the
total dissolved solids, electrical conductivity, and turbidity with absolute mean error
of 10% for the difference water bodies.
2.3
Estuarine Models
Estuarine models are needed to assess the hydrodynamic and water quality
conditions in the bay and estuarine environments. This section presents a brief
survey of available hydrodynamic models that were considered for the rivers and
estuaries water quality-modeling projects.
A number of existing models were
reported in a document entitled “Estuarine and Coastal Modeling” published by the
American Society of Civil Engineers (Spaulding, 2000). Most currently available
three-dimensional models were developed for application to large lakes, estuaries,
and other coastal regions. Mendelsohn et al.(1999) studied the Lower Savannah
River Estuary using a 3-D, coupled, prognostic, general curvilinear coordinate,
boundary fitted, hydrodynamic and water quality model system (WQMAP). They
13
discussed on the factors affecting salinity intrusion in the estuary and model
predictions of that intrusion. A three dimensional primitive equation ocean model
was used to study the dynamics of the Gaspe Current and cyclonic circulation over
the northwestern gulf of St. Lawrence (Sheng, 1999).
Numerical models of estuarine and coastal processes begun to be used in
planned engineering works when Lorentz (1926) simulated the closure of the Zuider
Zee in Netherlands. Harlemann (1971) reviewed the early development of onedimensional, estuarine model to simulate mixing and salinity.
O’Connor (1960) and Thomann (1962) pioneered the finite volume approach
to modeling estuarine water quality. It is similar to the box modeling approach
(Officer, 1978) developed for one dimensional applications.
The use of one-
dimensional models is limited to simulations of averaged cross sections based on the
Chezy or Manning equation to specify losses and dispersion coefficient to quantify
mixing (Martin et al., 1999). The models should be limited to well mixed estuary.
A large number of two dimensional, horizontal plane, finite difference or
finite elements models assume that the estuary is well mixed vertically, and resolve
longitudinal and lateral gradients (Martin et al., 1999). In practice, the longitudinal
and lateral mass dispersion coefficients are taken to be equivalent. There are a
number of quasi two dimensional models available such as link node models. The
link node models apply the one dimensional equations of motion in a two
dimensional framework consisting of a series of computational nodes that are
connected by channels.
The earliest 3-D circulation models were developed in the 1960s for the
applications of oceans and lakes (Smith, 1997). The first three-dimensional (3-D)
models applicable to the modeling of tidal motions in estuaries and coastal seas were
done in the early 1970s (Heaps, 1972; Leendertse et al., 1973; Sundermann, 1975).
14
Table 2.1: Available Estuary Model
Model name
Branch network flow model
CE-QUAL-RIV1
Primary
application
Dimensions
1-D
1-D
Reference
Rivers, estuaries
Streams, rivers,
estuaries
Estuaries
1-D
Schaffranek, 1987
Environmental Lab,
1995
Genet et al., 1974
Dynamic estuary model
(DEM)
MIT Transient Water
Quality Network Model
DYNHYD
EXPLORE-I
Estuaries
1-D
Harleman et al., 1977
Rivers, estuaries
Rivers, estuaries
1-D
1-D
Ambrose et al., 1988
Baca et al., 1973
DYNTOX
Rivers, estuaries
1-D
TABS-MD
2-D (horizontal)
WIFM-SAL
Rivers, estuaries,
bays, marshes
Rivers, estuaries,
bays, marshes
Estuaries
Martin and
McCutcheon, 1999
Thomas and
McAnally, 1985
Martin and
McCutcheon, 1999
Schmalz, 1995
CAFEX
Estuaries
2-D (horizontal)
HSCTM-2D
Rivers, estuaries
2-D (horizontal)
Wang and Conner,
1975
Hayter et al., 1997
FESWMS-2DH
2-D (horizontal)
Froehlich, 1989
2-D (horizontal)
Leendertse, 1970
FETRA
Streams, rivers,
estuaries
Estuaries, bays,
marshes
Rivers, estuaries
2-D (horizontal)
Onishi et al., 1979
H.S. Chen’s Model
TRIM
Rivers, estuaries, seas
Estuaries, bays
2-D (horizontal)
2-D (horizontal)
Chen, 1978
Cheng et al., 1993
CE-QUAL-W2
Lakes, reservoirs,
estuaries
Lakes, estuaries, bays
Lakes, rivers,
estuaries, bays
Lakes, estuaries
2-D (vertical)
3-D
Cole and Buchak,
1995
Blumberg, 1977
Sheng and Butler,
1982
Sheng et al., 1986
3-D
Hamrick, 1996
RMA Models
Rivers, lakes,
estuaries, bays
Rivers, estuaries, bays
3-D
HOTDIM
Estuaries, seas
3-D
WASP5
CE-QUAL-ICM
Rivers, estuaries, bays
Rivers, estuaries, bays
3-D
3-D
HYDRO-3D/ SED3D
Rivers, estuaries
3-D
Martin and
McCutcheon, 1999
Waldrop and Tatom,
1976
Ambrose et al., 1988
Martin and
McCutcheon, 1999
Martin and
McCutcheon, 1999
RMA2-WES
SIMSYS2D
Blumberg’s Model
CH3D/CH3D-WES
EHSM3D
EFDC
2-D (horizontal)
2-D (horizontal)
2-D (vertical)
3-D
15
Estuary models, although often times more complex, solve the same set of
equations as any other hydrodynamic model for lakes, rivers, or oceans. The primary
difference in estuarine models is the processes by which the governing equations are
solved, the scope of the parameters, and the functional structure of the model such as
1-D, 2-D, or 3-D. Various estuary models, listed in Table 2.1, were considered for
this study.
2.4
Review of Available Hydrodynamic Models
Hydrodynamic models can potentially represent the features of water
movement in rivers, streams, lakes, reservoirs, estuaries, near coastal waters, and
wetland systems. Depending on the type of system and model capabilities, spatial
dimensions of the simulation can include 1-D longitudinal, 2-D in the vertical, 2-D in
the horizontal or fully 3-D formulations.
Some 3-D models can effectively
“collapsed” to simulate systems in 1-D or 2-D. Hydrodynamic models employ
numerical solutions of the fundamental governing equations in order to predict water
movement based on the bottom topography and shoreline geometry. Hydrodynamic
models simulate the dynamic or time varying features of water transport which,
include water quantity and velocity of flow.
Hydrodynamic models employ
numerical solutions of the fundamental equations in order to predict water movement
based on the bottom topography, shoreline geometry and external boundary
conditions. Hydrodynamic models are either internally or externally coupled to
water quality models for dynamic simulation of receiving waters.
Some hydrodynamic models such as DYNHYD5 and EFDC are distributed as
stand alone models and can be externally coupled with water quality models such as
WASP5 and CE-QUAL-ICM. DYNHYD5 model is distributed as part of the
comprehensive WASP5 modeling system and is typically applied externally to
provide hydrodynamic flow computations, which are then input as the WASP5 water
quality model. Warwick and Heim (1995) provide comparison of the performance of
DYNHYD and RIVMOD models. DYNHYD has been applied to Sungai Skudai
Estuary and Sungai Langat Estuary as part of the eutrophication studies (Hashim et
16
al., 2004).
Tetra Tech (1995) describes a full 3-D application of EUTRO5 in
conjunction with the EFDC hydrodynamic model to assess the effectiveness of total
nitrogen removal from a wastewater treatment plant. Estuaries are most frequently
simulated using full 3-D hydrodynamic grids to account for the complex mixing and
transport processes.
Link Node Tidal Hydrodynamic Model (DYNHYD5) is a one-dimensional
model that uses the relatively simple link node concept to represent a waterbody
(Ambrose et al., 1987). The link node representation is best applied to branching
systems such as tidal rivers. The model solves the one dimensional equations of
continuity and momentum of a long wave in a shallow water system. The model is
distributed as a companion model to WASP5 and is typically applied externally to
provide hydrodynamic flow computations, which are then input to WASP5. Most
application of DYNHYD5 will use a simulation time step on the order of 30 seconds
to 5 minutes due to stability requirements. However, the hydrodynamic output file
created by DYNHYD5 may be stored at any user specified interval for use by a water
quality simulation program. (Tetra Tech, 1995)
Environmental Fluid Dynamics Computer Code (EFDC) is a general purpose
three dimensional hydrodynamic and salinity numerical model (Hamrick, 1992). The
theoritical and computational basis for the model is documented in Hamrick (1992).
The model may be applied to a wide range of boundary layer type environmental
flows that can be regarded as vertically hydrostatic. The model code uses a finite
difference scheme to solve the equations of motion and transport, simulating density
and topographically induced circulation, as well as tidal and wind driven flows, and
spatial and temporal distributions of salinity, temperature, and sediment
concentration. EFDC has been integrated with a water quality model to develop a
three dimensional hydrodynamic eutrophication model, HEM-3D (Park et al., 1995).
The EFDC model was used to develop a three dimensional hydrodynamic and
salinity numerical model of the Indian River Lagoon (Tetra Tech, 1994). The EFDC
model was linked to WASP5 for application to the Norwalk Harbor estuary in
Norwalk, Connecticut, for the purposes of developing a TMDL (Stoddard et al.,
1995). EFDC is one of the most extensively used hydrodynamic models and has
17
been tested in more than 60 modeling studies (Ji et al., 2001). The EFDC model has
been successfully applied to simulated hydrodynamics in rivers, reservoirs, lakes,
wetlands, harbors, estuaries and coastal seas, and the majority of EFDC applications
have been conducted in estuaries (Moustafa and Hamrick, 2000; Ji et al., 2001; Liu et
al., 2008).
Other hydrodynamic models are internally coupled, or connected, to the water
quality and toxic simulation programs such as CE-QUAL-W2 and CE-QUAL-RIV1.
CE-QUAL-W2 model has been applied to rivers, lakes, and estuaries (Martin, 1988;
Scott, 2000).
Table 2.2 describes the key features of both the stand-alone
hydrodynamic models and the internally coupled models. A review of the table
shows that capabilities vary widely in terms of dimension.
Two dimensional, Laterally Averaged, Hydrodynamic and Water Quality
Model (CE-QUAL-W2) is a two dimensional, laterally averaged hydrodynamic and
water quality model (Cole and Buchak, 1994). CE-QUAL-W2 is best applied to
stratified waterbodies like reservoirs and narrow estuaries where large variations in
lateral velocities and constituents do not occur. The water quality and hydrodynamic
routines are directly coupled; however, the water quality routines can be updated less
frequently than the hydrodynamic time step, which can reduce the computation
burden for complex systems.
The starting point of the present hydrodynamic model is the three
dimensional time dependent Navier Stokes equations for incompressible fluid with
hydrostatic pressure distribution and Boussinesq approximation.
Although three
dimensional models have been successfully developed to describe circulations driven
by wind, tide, and density gradients in estuaries, lakes, and marine waters (Sheng,
1987), it is not necessary to use three dimensional models at all times. For example,
for certain simpler geometry and bathymetry, it is possible to use a two dimensional
laterally averaged model and even a one dimensional laterally and vertically
averaged model.
For these simpler models, equations and numerical solutions
become significantly simpler than those for a three dimensional model.
18
Table 2.2: Evaluation of Capability – Hydrodynamic Models
Externally coupled models
Internally coupled models
CE-
Hydrodynamic
RIVMOD
DYNHYD5
EFDC
Models
CE-
CH3D- QUAL- QUALWES
RIV1
W2
HSPF
Water body type
Rivers/ streams
1
1
1
1
1
1
1
Lakes/
3
3
1
1
3
1
3
3
2
1
1
1-D
1
1
1
1
1
1
1
2-D
-
-
1
1
-
1
-
3-D
-
-
1
1
-
-
-
reservoirs
Dimension
Input data
Requirements
3
3
1
1
3
2
2
Calibration
1
1
1
1
1
1
1
Grid
-
-
1
3
-
-
-
generation/
interface
Output data
Format options
1
1
1
3
1
1
1
Graphics
3
2
3
3
3
3
3
Hydrologic
1
3
1
1
1
1
2
3
3
1
1
3
2
2
1
1
1
1
1
1
1
structure
simulation
Expertise
required for
application
Documentation
Shoemaker et al. (1997).
1
High
2
Medium
3
Low
- Not incorporated
19
2.5
Review of Available Water Quality Models
Water quality models can simulate the chemical and biological processes that
occur within a waterbody system, based on external and internal inputs and reactions.
Eutrophication models include those, which simulate biological inputs, nutrients, and
algal growth in rivers, streams, lakes, reservoirs, and estuaries. Other receiving
water models specialize in the simulation of toxic constituents and their
transformation and degradation in waterbodies.
Water quality models can be grouped by how they address change over time.
Some models employ a steady state formulation for simulation purposes. Typical
steady state applications include use of design flow, or pre-selected critical
conditions, for the assessment of steady state water quality impacts. Steady state
formulations are the most commonly used and the easiest to implement. A steady
state model such as QUAL-2E is appropriate to be used for a system that has steady
state inputs. QUAL-2E has been applied for prediction of water quality in Sungai
Langat system (Unit Perundingan Universiti Malaya, 2002), Sungai Tebrau and
Sungai Segget systems (IPASA, 2002).
For more detailed assessments of time varying conditions in receiving waters,
water quality models can be linked with hydrodynamic models. The use of dynamic
water quality models allows for a more detailed evaluation of time varying inputs,
such as non point sources and the examination of the short and longer term receiving
water response. Hydrodynamic and water quality models for Sungai Skudai Estuary
system has been applied using WASP5 model (Chua, 2002; Hashim et al., 2004).
Lung and Larson (1995) have used WASP5 model to evaluate phosphorus loading
reduction scenarios for the Upper Mississippi River and Lake Pepin.
WASP5 is a modeling system for assessing the fate and transport of
conventional and toxic pollutants in surface water bodies (Ambrose, 1993). It is
designed for linkage with the link node hydrodynamic model, DYNHYD5 and
dynamic simulation purposes that may be driven by either constantly repetitive or
variable tides. WASP5 was successfully linked with other hydrodynamic program
20
such as EFDC (Stoddard et al., 1995). WASP5 includes two submodels for water
quality or eutrophication and toxics, referred to as EUTRO5 and TOXI5,
respectively. Hashim et al. (2005) used the model to simulate the transport and fate
of dissolved oxygen (DO), biochemical oxygen demand (BOD), and nutrient loads
for the development of hydrodynamic and water quality models for Sungai Skudai
Estuary and Sungai Langat Estuary.
EFDC was integrated with a water quality model to develop a threedimensional hydrodynamic-eutrophication model, HEM-3D (Park et al., 1995). The
water quality portion of the model simulates the spatial and temporal distributions of
21 water quality parameters including dissolved oxygen, suspended algae (3 groups),
and various components of carbon, nitrogen, phosphorus and silica cycles, and fecal
coliform bacteria. Salinity, water temperature, and total suspended solids are needed
for computation of the twenty-one state variables within the hydrodynamic model.
Kinetic processes are similar to those in the Chesapeake Bay three-dimensional water
quality model, CE-QUAL-ICM (Cerco and Cole, 1994). The model was used to
develop a three dimensional hydrodynamic and salinity numerical model of the
Indian River Lagoon/Turkey Creek, with calibration and validation for St. Johns
river water management district, Palatka, Florida (Tetra Tech, 1994).
The principal differentiating factor for characterizing water quality models is
how they address the process of advection, dispersion and reaction. Advection is the
primary transport mechanism in a downstream and/or lateral direction. Dispersion
transport represents mixing (lateral and longitudinal) caused by local velocity
gradients. Reactions include the processes and transformation of constituents within
a water body. Table 2.3 presents a summary of the key features of water quality
models.
21
Table 2.3: Evaluation of Capability – Water Quality Models
Dynamic
Water Quality
DYNTOX
WASP5
Models
Steady state
CE-
CE-
QUAL-
EPA
QUAL-
QUAL-
2E
SCREENING
ICM
W2
Water body type
Rivers/ streams
1
1
1
1
1
1
Lakes/
-
3
1
1
3
1
Estuaries
-
1
1
2
2
1
Coastal
-
2
1
3
-
-
Physical processes
Advection
1
1
1
1
1
1
Dispersion
-
1
1
1
1
1
Heat balance
-
-
1
1
-
-
Particle fate
-
2
1
2
-
3
Eutrophication
-
1
1
1
1
1
Chemical fate
3
1
3
3
3
1
Sediment-
-
2
1
2
3
3
3
1
1
1
1
1
-
-
-
-
-
-
User interface
1
3
-
-
3
-
Documentation
2
1
2
2
1
1
reservoirs
water
External
loading
dynamic
Internally
calculated non
point sources
(NPS) loading
Shoemaker et al. (1997).
1
High
2
Medium
3
Low
-
Not
incorporated
22
2.6
Environmental Fluid Dynamic Code (EFDC)
The Environmental Fluid Dynamics Code is a general purpose modeling
package for simulating three-dimensional flow, transport, and biogeochemical
processes in surface water systems including rivers, lakes, estuaries, reservoirs,
wetlands, and coastal regions. The EFDC model was originally developed at the
Virginia Institute of Marine Science for estuarine and coastal applications and is
considered public domain software. In addition to hydrodynamic and salinity and
temperature transport simulation capabilities, EFDC is capable of simulating
cohesive and noncohesive sediment transport, near field and far field discharge
dilution from multiple sources, eutrophication processes, the transport and fate of
toxic contaminants in the water and sediment phases, and the transport and fate of
various life stages of finfish and shellfish.
Special enhancements to the
hydrodynamic portion of the code, including vegetation resistance, drying and
wetting, hydraulic structure representation, wave-current boundary layer interaction,
and wave-induced currents, allow refined modeling of wetland marsh systems,
controlled flow systems, and near shore wave induced currents and sediment
transport. The EFDC model has been extensively tested and documented for more
than 20 modeling studies. The model is presently being used by a number of
organizations including universities, governmental agencies, and environmental
consulting firms.
The structure of the EFDC model as shown Figure 2.1 in includes four major
modules: (1) a hydrodynamic model, (2) a water quality model, (3) a sediment
transport model, and (4) a toxics model. The EFDC hydrodynamic model itself,
which was used for this study, is composed of six transport modules including
dynamics, dye, temperature, salinity, near field plume, and drifter as shown in Figure
2.2.
23
EFDC Model
Hydrodynamics
Water Quality
Sediment
Transport
Toxics
Figure 2.1: Primary modules of the EFDC model.
Hydrodynamics
Dynamics
(E, u, w, mixing)
Dye
Temperature
Salinity
Near Field
Plume
Drifter
Figure 2.2: Structure of the EFDC hydrodynamic model.
2.6.1
Summary of EFDC Model Theory
The EFDC model's hydrodynamic component is based on the threedimensional hydrostatic equations formulated in curvilinear-orthogonal horizontal
coordinates and a sigma or stretched vertical coordinate. The momentum equations
are:
∂ t (m x my Hu)+ ∂ x (m y Huu)+ ∂ y (mx Hvu) + ∂ z (mx my wu)− fe mx my Hv
A
⎛
= −my H∂ x ( p + patm + φ ) + my (∂ x z*b + z∂ x H )∂ z p + ∂ z ⎝ mx my v ∂z u⎞⎠ + Qx
H
(2.1)
24
∂t (mx my Hv)+ ∂ x (my Huv) + ∂ y (mx Hvv ) + ∂z (mx mywv )+ f emx m y Hu
= −mx H∂ y ( p + patm
A
⎛
+ φ ) + m x (∂ z + z∂ y H )∂ z p + ∂z ⎝ mx my v ∂z v⎞⎠ + Qy
H
(2.3)
mx my fe = mx my f − u∂ y mx + v∂ x my
(τ ,τ ) = A H
xz
yz
v
(2.2)
*
y b
∂ z (u,v)
−1
(2.4)
where u and v are the horizontal velocity components in the curvilinear
horizontal coordinates x and y, respectively. The scale factors of the horizontal
coordinates are mx and my. The vertical velocity in the stretched vertical coordinate z
is w. The physical vertical coordinates of the free surface and bottom bed are zs*,
and zb* respectively. The total water column depth is H, and φ is the free surface
potential, which is equal to gzs*. The effective Coriolis acceleration fe incorporates
the curvature acceleration terms according to (2.3). The Q terms in equations 2.1 and
2.2 represent optional horizontal momentum diffusion terms. The vertical turbulent
viscosity Av relates to the shear stresses by the vertical shear of the horizontal
velocity components by equation 2.4.
The kinematics atmospheric pressure,
referenced to water density ρatm; the excess hydrostatic pressure in the water column
and is given by
∂ z p = −gHb = −gH(ρ − ρ o )ρ o−1
(2.5)
where ρ and ρo are the actual and referenced water densities and b is the
buoyancy. The three-dimensional continuity equation in the stretched vertical and
curvilinear horizontal coordinate system is
∂t (mx my H) + ∂ x (my Hu)+ ∂ y (mx Hv) + ∂ z (mx my w) = QH
(2.6)
with QH representing volume sources and sinks including rainfall,
evaporation, infiltration, and lateral inflows and outflows having negligible
momentum fluxes.
The generic three-dimensional transport and transformation
25
equation for a dissolved or suspended material represented by the concentration
variable C is
∂t (m x my HC)+ ∂ x (my HuC)+ ∂y (mx HvC ) + ∂z (mx mywC )
⎛m
⎛m
⎞
⎛m m
⎞
⎞
= ∂ x ⎜ y HAH ∂ x C⎟ + ∂ y ⎜ x HAH∂ y C⎟ + ∂ z ⎜ x y Ab∂ z C + mx my HRC
⎝ H
⎠
⎝ mx
⎠
⎝ my
⎠
(2.7)
where AH and Ab are horizontal and vertical turbulent mass diffusion
coefficients and Rc represents physical and biogeochemical sources and sinks. The
horizontal mass diffusion terms in (2.7) are generally omitted in the numerical
solution when the model is configured for three-dimensional simulation.
Vertical boundary conditions for the solution of the momentum equations are
based on the specification of the kinematic shear stresses
(τ ,τ ) = (τ ,τ )= c
xz
yz
bx
by
b
u12 + v12 (u1 , v1 )
(2.8)
Uw2 + Vw2 (Uw , Vw )
(2.9)
and
(τ ,τ ) = (τ ,τ ) = c
xz
yz
sx
sy
s
At the bottom, z=0, and on the free surface, z=1, with Uw and Vw being the
components of the wind velocity at 10 meters above the water surface. The subscript
1 refers to the velocity and elevation at the mid-point of the bottom layer. The
bottom drag coefficient is given by
⎛
⎞
κ
cb = ⎜
⎟
⎝ ln(∆ 1 / 2zo ) ⎠
2
(2.10)
where κ, is the von Karman constant, ∆1 is the dimensionless thickness of the
bottom layer, and zo=zo*/H is the dimensionless roughness height. The wind stress
coefficient is given by
26
cs = 0.001
(
ρa
0.8 + 0.065 Uw2 + Vw2
ρw
)
(2.11)
for the wind velocity components in meters per second, with ρa and ρw
denoting air and water densities, respectively. A no flux vertical boundary condition
is used for the transport equation (2.7) where C represents salinity.
Turbulent
viscosity and diffusion coefficients in the momentum and transport equations,
respectively, are determined using a turbulence closure model (Galperin et al., 1988;
Mellor and Yamada, 1982). The numerical solution procedures used in the EFDC
model are documented by Hamrick (1992) and summarized by Hamrick and Wu
(1997).
2.6.2
Modeling Assumption
The main objective of applying the EFDC model was to set up hydrodynamic
information throughout Sungai Johor estuary and its tributaries for the application of
the water quality model. It was necessary to accurately understand the variability of
flow throughout the stream network. Major assumptions taken contributed to the
final approach taken includes:
i.
Wind effects on flow and transport were not a critical factor due to the one
dimensional flow pattern.
ii.
The water-body was well mixed laterally and vertically, therefore a
longitudinal one dimensional configuration is appropriate.
iii.
Thermal stratification was not likely due to the shallow and narrow
characteristics of the estuary and river, thus temperature is not an important
driving force for flow and transport.
iv.
The impact of groundwater interaction on flow and transport was minimal
during low flow conditions, thus flow distribution can be obtained through
directly balancing the upstream and downstream flow rates.
27
2.7
EFDC Water Quality Model
Hydrodynamic
Model
Dynamics
Organic
Carbon
Algae
Phosphorus
Water Quality
Nitrogen
Silica
DO
COD
TAM
FCB
Sediment
Diagenesis
Predicted
Flux
Greens
Diatoms
Specified
Flux
Other
Figure 2.3: Structure of the EFDC water quality model (Park et al., 1995).
The EFDC subsequently has been internally integrated with a water column
eutrophication model and sediment-diagenesis model (Park et al., 1995) to develop
the Three Dimensional Hydrodynamic Eutrophication Model (HEM-3D). HEM-3D
consists of a hydrodynamic and water quality model linked internally as shown in
Figure 2.3. The computational scheme used in the internal eutrophication models
employs a fractional step extension of the same advective and diffusive algorithms
used
for salinity
and temperature, which
guarantee positive constituent
concentrations. A novel ordering of the reaction sequence in the reactive source and
sink fractional step allows the linearized reactions to be solved implicitly, further
guaranteeing positive concentrations. The eutrophication models accept an arbitrary
number of point and nonpoint source loadings as well as atmospheric and ground
water loadings.
28
The central issues in the water quality model are primary production of
carbon by algae and concentration of dissolved oxygen. Primary production provides
the energy required by the ecosystem to function. However, excessive primary
production is detrimental since its decomposition in the water and sediments
consume oxygen. Dissolved oxygen is necessary to support the life functions of
higher organisms and is considered an indicator of the health of estuarine systems.
In order to predict primary production and dissolved oxygen, a large suite of model
state variables is necessary. The nitrate state variable in the model represents the
sum of nitrate and nitrite nitrogen. The three variables (salinity, water temperature,
and total suspended solids) needed for computation of the above 21 state variables
are provided by the EFDC hydrodynamic model. The interaction among the state
variables is illustrated in Figure 2.4.
In this study, the water quality model consists of seven interlinked
components including phytoplankton population (Chlorophyll a), dissolved oxygen
(DO), ammonia nitrogen (NH3N), nitrate nitrogen (NO3N), phosphate (PO4),
chemical oxygen demand (COD) and dissolved organic carbon (DOC). Another
nutrient not included in this model is silica. Silica is a limiting nutrient for diatoms
only and thus it is generally modeled only when diatoms are simulated as a separate
phytoplankton group (Bowie et al., 1985). The present model uses the chlorophyll a
concentration to quantify the whole phytoplankton population.
2.7.1
Water Quality Model Formulation
This section presents the basic equations used in the EFDC's water quality
model, and provides an overview of the methods used to obtain a numerical solution.
The information in this section was obtained from a technical report written by Park
et al. (1995).
The water column eutrophication model in HEM-3D solves mass balance
equations for the 21 state variables in the water column, simulating three algal
groups, cycles of organic carbon, phosphorus, nitrogen, silica, dissolved oxygen
29
dynamics, and fecal coliform bacteria. The mass balance equation for state variables
may be expressed as:
∂C ⎞
∂C ∂ (uC ) ∂ (vC ) ∂ (wC ) ∂ ⎛
∂C ⎞ ∂ ⎛
∂C ⎞ ∂ ⎛
⎟⎟ + ⎜ K z
+
+
+
= ⎜Kx
⎟ + S C (2.12)
⎟ + ⎜⎜ K y
∂z ⎠
∂t
∂x
∂y
∂z
∂x ⎝
∂x ⎠ ∂y ⎝
∂y ⎠ ∂z ⎝
where
C = concentration of a water quality state variable
u, v & w = velocity components in the x-, y- and z-directions, respectively
Kx, Ky & Kz = turbulent diffusivities in the x-, y- and z-directions, respectively
SC = internal and external sources and sinks per unit volume.
Figure 2.4: Schematic diagram for the EFDC water column water quality model
(Park et al., 1995).
30
CHAPTER 3
RESEARCH METHODOLOGY
3.1
Introduction
Numerical water quality models have been extensively used to study and
manage water quality conditions in aquatic systems. Deterministic models for the
water column conditions are based on mass balance equations for dissolved and
particulate substances in water column, which consists of physical transport
(advective and turbulent diffusive transport) processes and biogeochemical
processes. Information on physical processes is usually obtained by applying the
hydrodynamic models.
Depending on the characteristics of a system, one may
choose an appropriate hydrodynamic model. For a large coastal system where both
horizontal and vertical gradient are significant, one needs to employ a three
dimensional hydrodynamic model, for example the Environmental Fluid Dynamic
Code, EFDC (Hamrick, 1992).
Generally, there are two main purposes for the nature process to be simulated
into a model. The first one is, to understand the relationship between the cause and
the consequences in where they can be used to make an important decision about the
maintenance of the river, while the second one is to analyze the input data
mathematically to achieve a liable and trusted output.
In order to fulfill a complete water quality model that can be used to assist in
decision making for the whole management activity in the future, a hydrodynamic
model and water quality model need to be developed. For this study, Environmental
31
Fluid Dynamic Code (EFDC) has been selected. A powerful three-dimensional
hydrodynamic model is required to construct an estuarine and coastal hydrodynamic
model because the model cannot be treated as a simple adjective system such as the
many rivers due to the consequence of these complex transport processes. Figure 3.1
shows the methodology approach to this study.
Data Collection & Analysis
(primary and secondary data)
Hydrodynamics and Water Quality
Model development
Hydrodynamic and water quality
model calibration
Hydrodynamics and water quality
sensitivity analysis
Hydrodynamics and water quality
model application
Figure 3.1: Research methodology flow chart
32
3.2
Database
It is important to collect the necessary data for the hydrodynamic and water
quality modeling. An intensive field observation surveys was conducted by the
UTM research group during January- June 2008. Field data are required to specify
the initial conditions, boundary conditions, and forcing function distributions in time
and space for the calibration, verification and application of a three-dimensional
hydrodynamic and water quality model of the study area. The data source of the
study area is summarized in Table 3.1.
Table 3.1: Data Sources of Sungai Johor Estuary Model
Date
Agency
Project
component
Flow measurement
Data category
Freshwater inflows
Tide stage
2008
UTM researchers
Water surface
elevation
Currents
Velocity and
direction
Water quality study Water quality
parameters
In situ WQ water
DO, salinity,
column profiles
conductivity,
temperature
Continuous DO
DO, salinity,
monitoring
conductivity,
temperature
Bacteriological
Chlorophyll a
sampling
Historical water quality and quantity data collection was conducted in the
Sungai Johor basin and its tributaries are summarized in Table 3.2. Data are also
obtained from other resources such as government departments and private agencies.
33
Table 3.2: Historical water quality and quantity data for Sungai Johor estuary
Survey date
1997-2006
Agency
1997-2006
River DOE
1997-2008
DID
Monthy T, salinity,
DO, pH
Monthy T, DO,
salinity, pH, SS, BOD,
COD, NH3-N, NO3,
Cond,
-
1997-2004
MMS
-
1998-2008
NAVY
-
2004
DHI
-
2004
DHI
-
3.2.1
Marine DOE
Water quality data
Hydrological data
-
Daily DID stream flow
measurement at Station
Rantau Panjang
Rainfall data, wind
speed and direction,
temperature
Tide level at Sungai
Belungkor
Tide level at Tanjung
Pengelih
Current meter data at
Tanjung Pengelih
Geographical Information System (GIS)
GIS is a computer-aided resource assessment and planning tool for the input,
storage, retrieval, analysis and display of interpreted geographic data. Data sets, such
as water quality parameters, land use patterns, topographic projections and others, are
used, analyzed and interpreted according to a set of criteria in determining potential
areas for coastal aquaculture development.
Such evaluation process using both
spatial and attribute information (empirical data sets) can be effectively generated
using ArcView GIS. Farah Salwa (2008) applied ArcView GIS to analyze the water
quality for Sungai Johor.
34
3.3
Equipments
This research work has been done using the advanced technology instruments
because the work that this research was involved in is complex in mechanism and
needs an accuracy reading to make sure that the data obtained can be trusted and
reliable.
Data sampling at various locations along the Sungai Johor was conducted
from January to June 2008. Garmin GPS-72 (Geographical Positioning System) as
showed in Figure 3.2 was used to determine each coordinate of water quality
stations.
Figure 3.2: Garmin GPS 72
A commercially available multi parameter water quality probe, Yellow
Springs Industries (YSI) water quality monitoring system model 6600 and 600-R are
shown in Figure 3.3. It was configured with conductivity, dissolved oxygen, salinity,
pH and temperature sensors to measure water quality conditions through water
column. The probes were lowered into the water column and the reading taken after
a steady value was obtained. The YSI meter is pre-calibrated as recommended by the
instrument manual in the laboratory and rechecked in the field prior to each
measurement. The internal data logger that came with the YSI sonde was not
flexible enough to accommodate our sampling data, and therefore a separate, external
data logger (EcoWatch) was used. EcoWatch is the PC software interface of YSI
environmental monitoring system used to upload the collected data on the equipment.
35
Figure 3.3: YSI water quality monitoring system model
While the rest of the water quality parameters such as ammonia nitrogen,
nitrate, BOD, orthophosphate and suspended solids have been analyzed at the
Chemistry Laboratory of Faculty of Science UTM from a water sample that was put
into a 2 liter polyethylene bottle which was cleaned according to the Standard
Method APHA 4500-P. The water sample was then being preserved by putting a few
drops of nitrite acid (H-NH3) and stored at 4°C cold room as soon as the BOD
analysis had carried out in order to minimize the biological activities in the water.
Table B.1 summarized the water quality parameters and their corresponding methods
of analysis in Appendix B.
Hondex digital depth sounder as shown in Figure 3.4 was used with the
portable digital depth sounder to measure the depth of water body.
Figure 3.4: Hondex Digital Depth Sounder
36
Solinst levelogger as shown in Figure 3.5 was used to record the water
surface elevation fluctuations and temperature in Sungai Johor estuary system. Time
series data of water level were used to calibrate the water level.
Figure 3.5: Solinst levelogger
3.4
Model Set Up
The general procedure for the application of the EFDC model to study area
follows a sequence of steps beginning with the model set-up or configuration. Model
configuration involves the construction of a horizontal grid of the water body and
interpolation of bathymetric data to the grid, construction of EFDC input files, and
compilation of the source code with appropriate parameter specification of array
dimensions. In this application, the EFDC model was configured to simulate time
varying surface water elevation, velocity, salinity, water temperature and water
quality parameters.
37
3.4.1 Model domain
The model domain, as shown in Figure 3.6, defines the Sungai Johor
estuarine system and major surrounding tributaries that provide fresh water inflow
into the estuary. The curvilinear-orthogonal horizontal model grid for the Sungai
Johor estuary system was comprised of approximately 6613 active quadrilateral
water cells, with each cell divided into 10 layers to capture the vertical variations of
simulated constituents.
The grid systems were constructed by using GEFDC
program. GEFDC is a preprocessor system, which can generate either Cartesian or
curvilinear orthogonal grids using methods outlined by Mobley and Stewart (1980)
and Ryskin and Leal (1983).
There are 48 downstream (seaward) boundary cells located at Tanjung
Pengelih. The upstream boundary cells include Sungai Seluyut, Sungai Berangan,
Sungai Redan, Sungai Temon, Sungai Tiram, Sungai Layang, Sungai Layau, Sungai
Papan, Sungai Lebam, Sungai Pelepah, Sungai Serai, and several unnamed
tributaries.
3.4.2
Bathymetry
The bathymetric data of Sungai Johor estuarine system and its surrounding
areas were obtained from the Malaysian Navy navigational chart. Overall
bathymetric features are illustrated in Figure 3.7.
38
Figure 3.6: The curvilinear orthogornal horizontal model domain of Sungai Johor
estuary
Figure 3.7: The model bathymetry incorporated data from bathymetric data.
39
3.4.3 Model Boundary Conditions
The EFDC hydrodynamic model was configured to simulate four
characteristics of the Sungai Johor estuarine model: salinity, velocity, temperature,
and water surface elevation. As a condition to the numerical solution of the equation
used to predict the four variables, values for salinity, velocity, temperature and water
elevation must be specified at the model boundaries. Conditions at the Sungai Johor
estuary seaward boundary was defined by water elevation, temperature, and salinity
time series. Conditions at the upstream boundary was defined by averaged daily river
flow and freshwater (zero salinity). In this study, the model is capable of reading
separate input files for time series specifications of tidal height as well as salinity at
the seaward boundary and freshwater discharges at upstream and tributaries
locations.
The downstream boundary point for the model is located at Tanjung Pengelih.
This downstream seaward boundary was forced by a tidal elevation series as
measured at Sungai Belungkor.
The upstream boundary of Sungai Johor is located outside the tidal influence,
allowing the use of a simple flow time series boundary condition at Kampung Rantau
Panjang. A flow condition was defined by the Department of Irrigation and Drainage
(DID) data.
Information on wind speed and direction was obtained from the Senai Airport
meteorological stations to define wind conditions within the model domain. The
boundary conditions of surface water elevation, wind speed and direction, air
temperature, atmospheric pressure, and solar radiation are used for calibration period.
3.4.4
Model Initial Conditions
When the EFDC model is first activated, an initial flow field of tidal height
and salinity values is required.
It is important that the residual effects of any
40
inaccurate initial spatial distribution be eliminated for each state variable (tidal
elevation, salinity, and temperature) prior to comparing model results to field data.
It is advantageous to restart the model with a set of values from a prior run for it to
be as realistic as possible.
In this study, tidal heights were initialized to zero (level free surface)
everywhere. Prescribed boundary conditions were used to drive the system until an
equilibrium condition.
The initial conditions of salinity and temperature were specified in salt.inp
and temp.inp files, respectively. Constant values were prescribed throughout the
domain for each variable.
Initial
conditions
for
EFDC
water
quality
model
include
initial
concentrations, as well as solid transport field for each solid in each segment/ grid. In
this study, the initial concentrations of ammonia nitrogen, nitrate nitrogen and
dissolved oxygen was based on the depth averaged conditions. Linear interpolations
were made on the available water quality sampling stations to initialize the
concentrations of water quality parameters throughout the water quality segments.
Table 3.3 shows the water quality initial conditions for Sungai Johor estuary model.
Table 3.3: Initial conditions for Sungai Johor water quality model
Description
Value
Total phosphate
0.01
Ammonia nitrogen
0.01
Nitrate nitrogen
0.01
Chemical oxygen demand
10.00
Dissolved oxygen
6.00
41
3.4.5 Model Kinetics
Model kinetics and parameters determine the decay of the pollutants and the
oxygen uptake amount in the system.
These kinetic rates and parameters are
determined based on the measured data and standard water quality modeling
assumptions. The parameters are shown in Table 3.4.
Table 3.4: EFDC kinetic coefficients in the water quality model application
Description
Reaeration constant
Temperature rate constant for reaeration
Oxygen half-saturation constant for COD decay (mg/l O2)
COD decay rate (per day)
Reference temperature for COD decay (ºC)
Temperature rate constant for COD decay
Maximum algal growth rate (per day)
Optimum temperature for algal growth (ºC)
Effect of temperature on algal growth below optimum temperature (ºC-2)
Effect of temperature on algal growth below optimum temperature (ºC-2)
Algal basal metabolism rate at 20ºC (d-1)
Half saturation constant for nitrogen uptake of algae (g N m-3)
Half saturation constant for Phosphorus uptake of algae (g P m-3)
Half saturation constant for Silica uptake of diatoms (g Si m-3)
Algal predation rate at 20ºC (d-1)
Algal settling rate (m d-1)
Decay rate of organic carbon at 20ºC (d-1)
Decay rate of organic phosphorus at 20ºC (d-1)
Decay rate of organic nitrogen at 20ºC (d-1)
Settling rate of particulate organic matter (m d-1)
Maximum nitrification rate at 27ºC (g N m-3 d-1)
Dissolution rate of particulate silica at 20ºC (d-1)
Sediment oxygen demand (g O2 m-2 d-1)
Benthic flux of ammonia (g N m-2 d-1)
Benthic flux of nitrate (g N m-2 d-1)
Benthic flux of phosphorus (g P m-2 d-1)
Benthic flux of dissolved silica (g N m-2 d-1)
Value
2.00
1.024
1.5
0.1
20
0.041
2.25
15.0
0.008
0.008
0.03
0.01
0.001
0.05
0.16
0.10
0.005
0.005
0.005
1.0
0.07
0.03
2.00
0.1
0.005
0.005
0.075
42
3.4.6
Model Input Parameters
Application of the EFDC model to Sungai Johor estuarine system is managed
through a series of input files listed in Table 3.5. A complete description of user
designated input can be found in Hamrick (1996).
Two files output by the GEFDC grid processor (dxdy.out and lxly.out) were
renamed to dxdy.in and lxly.inp and used as input to the hydrodynamic portion. File
dxdy.inp includes each cell, the I and J indices, the x and y horizontal dimensions,
depth, and bottom elevations, and bottom roughness and vegetation class. The file
lxly.inp specifies both the horizontal cell center coordinates but also the cell
orientations (cartesian or curvilinear).
Table 3.5: Files required for EFDC model simulation
Filename
Efdc.for
Type
Model
Comments
No changes required
source
Cell.inp
Input
Grid cell types
Celllt.inp
Input
Grid cell types
Dxdy.inp
Input
Cell dimensions, bottom roughness
Lxly.inp
Input
Cell locations in UTM (m)
Salt.inp
Input
Initial salinity concentration
Temp.inp
Input
Initial temperature
Dye.inp
Input
Initial dye concentration
Aser.inp
Input
Atmospheric data: air temperature, solar radiation,
atmospheric pressure, evaporation, rainfall, cloud cover
Wser.inp
Input
Wind speed and direction
Pser.inp
Input
Tidal height time series
Qser.inp
Input
Time series data for freshwater discharge
Tser.inp
Input
Temperature specification at boundaries.
Sser.inp
Input
Salinity concentration specification at boundaries.
Wq3dwc.inp
Input
Water quality concentration specification at boundaries.
43
The cell.inp is the important file for grid generation and EFDC application. It
includes geometry or domain boundary and a designator file. This is accomplished
by assigning a cell type (water or dry land) within the specified grid matrix.
The files salt.inp and temp.inp were used to initialize the model domain with
a pre-determined salinity and temperature, respectively. Salinity and temperature
were input to each layer of the cells.
Time series data for freshwater discharge were input to EFDC via file
qser.inp. The file ser.inp specifies the tidal height time series at the open seaward
boundary. Meteorological data including atmospheric pressure, air temperature, solar
radiation, rainfall and evaporation are important to the model and are input via the
file aser.inp. Wind speed and direction are input to the file wser.inp. Salinity and
temperature specification at the Tanjung Pengelih and upstream boundary at
Kampung Rantau Panjang are also required. These variables were input via files
sser.inp and tser.inp, respectively.
3.5
Model Calibration
Model calibration appears in various forms, dependent on data availability,
characteristics of water body, and most of all, the perceptions and opinions of
modelers. It often happens that calibrated model results are compared with observed
data to demonstrate agreements, with little explanation of which coefficients were
adjusted (and the nature of the adjustment) in arriving at the calibrated model.
Hsu et al. (1999) calibrated their vertical (lateral averaged) 2D model in three
steps; preliminary calibration, fine tune calibration of friction coefficient and
calibration of mixing processes. They recommended using the distribution of the
tidal range as a function of distance from the river mouth to calibrate the friction
coefficient.
It is an accepted requirement that a numerical model of estuarine
hydrodynamics be calibrated and verified before being put into any practical usage.
44
3.6
Model Sensitivity Analysis
Sensitivity analysis is used to determine how sensitive a model is to changes
in the value of the parameters of the model and to changes in the structure of the
model. In this study, we focused on parameter sensitivity. Parameter sensitivity is
usually performed as a series of tests in which the modeler sets different parameter
values to see how a change in the parameter causes a change in the dynamic behavior
of the stocks. By showing how the model behavior responds to changes in parameter
values, sensitivity analysis is a useful tool in model building as well as in model
evaluation.
Sensitivity analysis helps to build confidence in the model by studying the
uncertainties that are often associated with parameters in models. Many parameters
in system dynamics models represent quantities that are very difficult, or even
impossible to measure to a great deal of accuracy in the real world. In addition, some
parameter values change in the real world. Therefore, when building a system
dynamics model, the modeler is usually at least somewhat uncertain about the
parameter values he chooses and therefore, must use estimation. Sensitivity analysis
allows him to determine what level of accuracy is necessary for a parameter in order
to make the model sufficiently useful and valid. If the tests reveal that the model is
insensitive, then it may be possible to use an estimation rather than a value with
greater precision. Sensitivity analysis can also indicate which parameter values are
reasonable to be used in the model. If the model behaves as expected from real
world observations, it gives some indication that the parameter values reflect, at least
in part, the “real world.”
Sensitivity tests help the modeler to understand dynamics of a system.
Experimenting with a wide range of values can offer insights into behavior of a
system in extreme situations. Discovering that the system behavior greatly changes
for a change in a parameter value can identify a leverage point in the model, a
parameter whose specific value can significantly influence the behavioral mode of
the system.
45
3.7
Statistical Comparison Techniques
Basic comparative statistics and their associated tests (Ambrose, 1987) were
compiled into a recent book about hydrodynamic modeling (Martin and
McCutcheon, 1999) and were used to validate the EFDC model.
The simplest statistical test was the mean error (ME) defined as the difference
of the means
ME =
∑ x−c
(3.1)
N
where x is the observed values, c is the model-predicted or calculated values and N is
the number of data pairs compared.
Where data are missing the error is not
computed and therefore not included in the calculation. The mean error is as it
sounds, a comparison of how the simulated averages of a parameter (e.g., salinity)
compares with the measured average over the period chosen for comparison. The
units of the mean error were the same as those for the observations.
A mean error of zero is ideal. Positive mean error indicates that on average,
model predictions are less than observations. A negative mean error indicates model
predictions exceed observations, on average.
The absolute mean error is also used. It is determined to be similar to the
mean error except that the absolute value of the difference between calculated and
observed values is taken. The absolute mean error is defined as:
AME =
∑ x−c
N
(3.2)
where x is the observed values, c is the model-predicted or calculated values and N is
again the number of data pairs. It represents the average difference in the two signals
as compared to the differences in the average. The closer AME is to zero, the better
the model accuracy is.
46
Mean error or absolute mean error is a measure of model accuracy, bias, or
systematic error. If ME > 0, the model had systematically oversimulated and if ME
< 0, the model had systematically undersimulated the response of the natural system.
Neither mean error nor absolute mean error measured imprecision or scatter in
residuals (observed-simulated). Overall, the ME was useful for guiding calibration
and judging validation testing, but other statistics were necessary to determine the
precision and correlation.
Precision of a simulation was determined from the standard error of
estimation or root mean square error,
∑ (S
N
RMSE =
i =1
i
− Oi )
2
N
(3.3)
where N is the total number of measurements in space or time. It is
statistically well behaved and is a direct measure of the model error. RMSE had the
same units and dimensions as the observations and simulations.
A root mean square error of zero is ideal. The root mean square error is an
indicator of the deviation between predictions and observations. The root mean
square is an alternative to (and usually larger than) the absolute mean error.
The dimensionless coefficient of variation or % root mean square error is
% RMSE =
RMSE
× 100
O
(3.4)
where O was the average of observation Oi
Correlation between model and observed data completed using the
regression analysis was more useful in evaluating the calibration or validation of a
model. The regression equation is
Oi = α 1 + β1 S i + ∈
(3.5)
47
where α 1 and β1 were dimensionless regression coefficients representing the true yintercept and slope of a plot of observed versus simulated values, and ∈ was the
error in simulation S i . The coefficient of determination R ( or for correlation
coefficient R2) is
R2 =
n
⎛ n
⎞ ⎛ n
⎞
⎜ n ∑ (S i × Oi )⎟ − ⎜ ∑ Oi × ∑ S i ⎟
i =1
⎝ i =1
⎠ ⎝ i =1
⎠
⎛⎛ n 2 ⎞ ⎛ n
⎞
⎜⎜ ⎜ n∑ S i ⎟⎟ − ⎜ ∑ S i ⎟
⎝ ⎝ i =1 ⎠ ⎝ i =1 ⎠
2
⎞
⎟×
⎟
⎠
2
n
⎛⎛ n
⎞
⎜ ⎜ n O 2 ⎞⎟ − ⎛⎜ O ⎞⎟ ⎟
∑
∑
i
i
⎜ ⎝ i =1
⎠ ⎝ i =1 ⎠ ⎟⎠
⎝
(3.6)
and provided an indication of the degree correlation between the observed (Oi) and
predicted (Si) data for a given number of observations, n. The dimensionless
correlation coefficient, R2 approached one when a high degree of correlation
occurred, and approached zero when data were not correlated.
Whether the
simulation was accurate was determined by testing the significance of the
intercept α 1 and the slope β1 . If the slope was not significantly different from one
and the intercept was less than or equal to some tolerance centered about zero, the
statistics indicated accuracy.
Salinity was chosen as the observed and simulated values to test the accuracy,
precision, and correlation of the model to field data. The ME was used to determine
model accuracy where the more accurate the model, the closer to zero the ME
approached. The RMSE was used as an indicator of model precision with values
close to zero are more precise. Finally, the R2 statistic was used as a measure of the
variability in the observed data that was explained by the model.
48
CHAPTER 4
DATA COLLECTION AND ANALYSIS
4.1
Stream Data Assessment
As indicated in Table 4.1, stream stage and discharge measurements on the
Sungai Johor at Rantau Panjang are available from a Department of Irrigation and
Drainage (DID) gauge station that was maintained from 1965 to 2008. This flow
measurement station provided the upstream freshwater inflow data for the model.
The Rantau Panjang gage is located approximately 80 kilometers upstream at
Tanjung Pengelih. A representative hydrograph for calendar year 2006 is presented
in Figure 4.1. The temporal profiles of stream flow at Station Rantau Panjang from
year 1996 to 2006 are shown in Appendix C.
Station Sungai Johor at Rantau Panjang
350
300
Flow (m3/s)
250
200
150
100
50
0
Jan06
Feb06
Mar06
Apr- May06
06
Jun- Jul-06 Aug- Sep06
06
06
Oct06
Nov- Dec06
06
Jan07
Tim e (m onth, year)
Figure 4.1: Stream Flow Data of Sungai Johor at Rantau Panjang (2006).
49
Table 4.1: Hydraulic Data for Sungai Johor basin
Location
DID station
Available data
Duration
Frequency
Rantau
1737451
Stage,
1965-2007
Daily
Panjang
4.2
Discharge
Meteorological Data
Meteorological data are available from climatological station 48679 Senai
Airport. As indicated in Table 4.2, hourly meteorological data includes rainfall
duration, temperature, solar radiation, and cloud cover is required as input file in
order to run EFDC model.
Table 4.2: Meteorological Data for Sg Johor basin
Location
4.3
Available data
Duration
Frequency
Air temperature
1974-2004
~Monthly
Rainfall
1974-2004
~Monthly
Wind speed and direction
1990-2001
~Monthly
Senai Airport
Rainfall duration and amount
January 2008
Hourly
(48679)
Dry and wet bulb temperature
January 2008
Hourly
Solar radiation
January 2008
Hourly
Mean sea level pressure (MSL)
January 2008
Hourly
Total cloud cover
January 2008
Hourly
Freshwater Flow
The Sungai Johor hydrodynamic model includes major point sources
discharges from tributaries.
The average flow discharge for Sungai Johor
hydrodynamic model input files is presented in Table 4.3.
50
Table 4.3: Averaged flow discharges for Sungai Johor estuary model
Flow ID
Cell I,J
Description
Flow discharge
(m3/s)
Flow01
157, 329
Rantau Panjang
Time variables
Flow02
185, 221
Sungai Seluyut
0.104
Flow03
157, 215
Sungai Berangan
0.012
Flow04
162, 174
Sungai Temon
0.012
Flow05
150, 171
Sungai Redan
0.012
Flow06
141, 159
Sungai Tiram
1.500
Flow07
160, 150
Sungai Layang
0.012
Flow08
176, 141
Sungai Serai
0.012
Flow09
210, 122
Sungai Layau Kiri
0.023
Flow10
247, 148
Sungai Papan
1.240
Flow11
272, 140
Sungai Lebam
1.610
Flow12
211, 90
Sungai Belungkor
1.500
Flow13
255, 75
Sungai Santi
1.240
Flow14
268, 66
Sungai Sebina
0.509
Flow15
160, 71
Sungai Kim Kim
1.500
Flow16
128, 74
Sungai Buloh
0.012
Flow17
45, 118
Sungai Tebrau
1.500
Flow18
174, 525
Sungai Pelepah
2.500
Flow19
169,567
Sungai Sisek
0.509
Flow20
168,671
Sungai Lebak
1.500
Flow21
148,686
Sungai Semangar
0.509
4.4
DOE Water Quality Monitoring Data
According to DOE (2007), a total of 51 marine water quality monitoring
stations were set up throughout the Johor State and 29 river water quality monitoring
stations were set up in the Johor river basin. Under the Malaysia Singapore Joint
Committee on the Environment (MSJCE) monitoring program, 20 stations were also
51
monitored in 2007. Locations of DOE river and marine water quality monitoring
stations are shown in Figure 4.2. Water quality data from 1997 to 2006 is available
from DOE.
3JH13
#
S
N
3JH15
#
S
Sg.Johor
#
S
3JH16
#
S
3JH19
W
Rantau Panjang
3JH20
S
Kota Tinggi
3JH18
#
S
Sg.Seluyut
3JH08
#
S
3JH07
#
S
U
%
DOE River WQS
DOE Marine WQS
Sg.Temon
Sg.Redan
Johor Lama
Sg.Tiram
3JH06
#
S
#
S
3JH22
#
S
Sg.Berangan
3JH09
#
S
E
#
S
Sg.Papan
Sg.Layang
Teluk Sengat
#
S
3JH03
Sg.Chemangar
#
S3JH25
#
S
Sg.Layau
3JH05
#
S
#
S
3JH28
3JH30
Sg.Serai
Tg.Serindet
Sg.Tebrau
U
SJ4A %
1437951
U
%
1340973 %
U
U
%
SJ4
%
U
1438913
Pasir Gudang
Sg.Kim Kim
Tg.Surat
U
%
Sg.Johor
Sg.Belungkor
1440963
Sg.Buloh
1428939
U
%
U
%
1439965
U
%
1440916
Sg.Lebam
U
%
SJ2
SJ1%
U
P.Ubin
1441966
U
%
P.Tekong
3JH33
#
S
Sg.Sebina
1443969 Sg.Santi
U
%
Tg.Pengelih
1441967 %
U
U
%
1441968
Figure 4.2: Locations of DOE river and marine water quality stations
Water quality index (WQI) as shown in Table 4.4 is the most important
criteria in order to determine the water quality in particular waterbodies and limit to
freshwater or river. The use of an appropriate WQI would be more beneficial for
administrative and management purposes and for meaningful communication with
the public. DO, BOD, COD, AN, SS and pH are common parameters that are used
by DOE in determining WQI.
River classification for each parameter can be
measured by using Table 4.5. The percentage of the entire parameters will be
evaluated from that the classes will be determined.
52
Table 4.4: Water Quality Index (DOE, 1986)
WQI range
Pollution Degree
<31.0
Severely polluted
31.0-51.9
Slightly polluted
51.9-76.5
Moderate
76.5-92.7
Clean
>92.7
Very clean
To compare the water quality status of the rivers and their changes over the
past three years, the maximum, minimum and average values of the annual mean
WQI for each river in 2004, 2005 and 2006 are plotted in Figure 4.3. Based on WQI,
Sungai Serai, Sungai Berangan and Sungai Tiram are categorized as moderate river
or Class III and most of the rivers are categorized as clean river or Class II.
Figure 4.4 displays the water quality trend at each monitoring station along
upper Sungai Johor mainstream over the past ten years from 1997 to 2006. The
rivers’ water quality generally deteriorated towards the estuaries, indicating multiple
inputs of pollutants along the river systems (Lim and Leong, 1988). As shown in
Figure 4.4, stations 3JH19 and 3JH20 are categorized as clean or Class II except in
June 1997 and the year 2000, which have been categorized as polluted river and
moderate river, respectively. Station 3JH20 is located near Kota Tinggi bridge and
station 3JH19 is approximately 7 kilometers upward from station 3JH20. Stations
3JH13 and 3JH15 are categorized as clean river or Class II. Station 3JH13 is located
near Kampung Rantau Panjang, Kota Tinggi and station 3JH15 is located
approximately 5 kilometers downward of station 3JH13.
53
Table 4.5: The DOE Water Quality Index Classification (DOE, 1986)
Class
Parameter
Unit
I
II
III
IV
V
mg/l
<0.1
0.1-0.3
0.3-0.9
0.9-2.7
>2.7
BOD
mg/l
<1
1-3
3-6
6-12
>12
COD
mg/l
<10
10-25
25-30
50-100
>100
DO
mg/l
>7
5-7
3-5
1-3
<1
pH
-
>7
6-7
5-6
<5
>5
mg/l
<25
25-50
50-150
150-300
>300
76.5-92.7 51.9-76.5 31.0-51.9
<31.0
Ammoniacal
Nitrogen
Suspended
solids
Water quality Index
>92.7
WQI (mg/l)
WQI (mg/l)
60
Sa
nt
i
Pa
pa
n
Se
lu
yu
t
Pe
le
pa
h
La
ya
u
Le
ba
m
Te
m
on
Se
ra
i
Ti
ra
m
Se
m
an
ge
Be
r
ra
ng
an
La
ya
ng
Sa
nt
i
Pa
pa
n
Se
lu
yu
t
Pe
le
pa
h
La
ya
u
Le
ba
m
Te
m
on
Se
ra
i
Ti
ra
m
Se
m
an
ge
Be
r
ra
ng
an
La
ya
ng
WQI (mg/l)
60
Sa
nt
i
Pa
pa
n
Se
lu
yu
t
Pe
le
pa
h
La
ya
u
Le
ba
m
Te
m
on
Se
ra
i
Ti
ra
m
Se
m
an
ge
Be
r
ra
ng
an
La
ya
ng
54
100
2004
80
clean
moderate
40
20
polluted
0
River
100
2005
80
clean
moderate
40
20
polluted
0
River
100
2006
80
clean
60
moderate
40
20
polluted
0
River
Figure 4.3: Range of WQI for selected rivers along Sungai Johor
55
3JH20
3JH19
100
Water QuaIity Index
90
clean
80
70
moderate
60
50
40
30
polluted
20
10
0
1997
1998
1999
2000
2001 2002 2003
tim e (years)
3JH15
2004 2005 2006
2007
3JH13
100
Water QuaIity Index
90
clean
80
70
moderate
60
50
40
30
polluted
20
10
0
1997 1998
1999
2000
2001 2002
2003 2004
tim e (years)
2005 2006
2007
Figure 4.4: Water quality trends at DOE river monitoring stations
Water quality is the most important criterion in site selection. Marine and
river water quality data of Sungai Johor basin were used to investigate the
relationship between the longitudinal distribution of salinity, temperature and DO.
The range of values of longitudinal distribution of salinity, temperature and DO
parameters from Tanjung Pengelih to Kampung Rantau Panjang are summarized in
Tables 4.6. Eight of the marine and river water quality stations were analyzed to
investigate the temperature, dissolved oxygen and salinity concentration trend along
Sungai Johor mainstream from downstream boundary at Tanjung Pengelih to
56
upstream boundary at Kampung Rantau Panjang. Trend of temperature, DO and
salinity are shown in Figure 4.5-4.7, respectively. The longitudinal profiles of DO
concentration, salinity and temperature for year 2003 to year 2005 are shown in
Appendix A.
Table 4.6: The Ranges of Value of Water Quality Parameters in Sungai Johor
(DOE, 2006)
Station
No.
1441967
1440963
1440916
1340973
3JH20
3JH19
3JH15
3JH13
Distance from
Tanjung Pengelih (km)
0
12
19
22
62
68
78
84
Temperature (°C)
min
max average
27.53
28.34
27.94
28.48
28.72
28.60
29.54
31.57
30.56
29.25
29.98
29.62
26.86
28.72
27.67
26.07
28.07
27.17
25.84
28.10
27.22
26.17
28.01
27.17
DO (mg/l)
max average
6.71
6.59
6.71
6.57
6.62
6.48
6.32
6.21
6.40
5.59
6.67
6.36
6.85
6.54
6.75
6.57
min
6.47
6.42
6.33
6.09
4.87
5.94
6.07
6.22
min
min
29.68
31.62
24.66
30.40
0.02
0.02
0.02
0.02
max
Salinity (ppt)
max average
32.82
31.25
33.33
32.48
29.68
27.17
31.02
30.71
0.03
0.03
0.03
0.02
0.03
0.03
0.04
0.03
average
1440916
32
temperature (°C)
31
1340973
30
1440963
3JH20
29
3JH19
3JH15
3JH13
28
27 1441967
26
25
0
DOWNSTREAM
(Tg Pengelih)
10
20
30
40
50
60
distance from Tanjung Pengelih (km )
70
80
90
UPSTREAM
(Rantau Panjang)
Figure 4.5: Trend of temperature from downstream boundary at Tanjung Pengelih
towards upstream boundary at Rantau Panjang, Kota Tinggi (DOE, 2006).
57
min
max
average
1440916
8
7
1340973
DO (mg/l)
6
3JH20
1440963
5
3JH19
3JH15
3JH13
4
3 1441967
2
1
0
DOWNSTREAM
(Tg Pengelih)
10
20
30
40
50
60
70
distance from Tanjung Pengelih (km )
80
90
UPSTREAM
Rantau Panjang)
Figure 4.6: Trend of dissolved oxygen (DO) from downstream boundary at Tanjung
Pengelih towards upstream boundary at Rantau Panjang, Kota Tinggi (DOE, 2006).
min
max
average
1440916
40
35
1340973
salinity (ppt)
30
25
3JH20
1440963
3JH19
20
3JH15
3JH13
15
10 1441967
5
0
0
DOWNSTREAM
(Tg Pengelih)
10
20
30
40
50
60
distance from Tanjung Pengelih (km )
70
80
90
UPSTREAM
(Rantau Panjang)
Figure 4.7: Trend of salinity from downstream boundary at Tanjung Pengelih
towards upstream boundary at Rantau Panjang, Kota Tinggi (DOE, 2006).
4.5
Intensive Survey
Several intensive surveys were conducted by UTM research group during
January- April 2008. The location of the intensive survey stations is shown in Figure
4.8. In this study, a large number of physical, chemical and biological parameters
were collected from selected sampling sites along Sungai Johor estuarine system as
58
summarized in Table 4.7. Dissolved oxygen (DO), salinity, water temperature and
specific conductance were taken at the sampling sites in January to April 2008.
Measurements were taken vertically every one meter and about 1 foot below the
water surface and 1 foot above the bottom. The maximum depth that could be
measured at the site was 20 meters. Other water column analyses were biochemical
oxygen demand, nitrite plus nitrate nitrogen, ammonia nitrogen, total phosphorus and
total suspended solid. The samples were collected near mid channel by composting
equal-sample volumes of water collected vertically from the surface to near the
bottom.
Table 4.7: Summary of intensive field surveys data (2008)
Survey date
N
Location
January 2008
7
Sungai Tiram
February 2008
7
Sungai Tiram
February 2008
12
Sungai Lebam
February 2008
24
Sungai Johor mainstream
March 2008
20
Sungai Johor mainstream
March 2008
17
From Tg Pengelih to Tg Buai
April 2008
17
Sungai Johor mainstream
59
N
Rantau Panjang
Sg.Johor
W
#
S
#
S
#
S
#
S #
S
#
S
S
#
S
#
S
Sg.Seluyut
#
S
T
$
#
S
Sg.Berangan
U
%
#
S
V
&
#
S
'W
#
S
Sg.Redan
V
&
7
V
&
6
#
S
5
Johor Lama
Sg.Papan
V
&
1
V
&
4
V
&
3
V
&
2
Sg.Layang
Sg.Chemangar
#
S
Teluk Sengat
#
S
Sg.Tebrau
Sg.Serai
Sg.Layau
#
S
#
S
U
%
#
S
#
S
'W 'W
Pasir Gudang
'W
Sg.Buloh
'W
'W
Tg Pengelih
Sg Johor
Sg Lebam
Sg Tiram
Johor Strait
Sg.Temon
Sg.Tiram
V
&
E
Kota Tinggi
U
%
#
S
U
%
%
UU
%
U
%
U
%
U
%%
U
Tg.Serindet
Sg.Lebam
U
%
T
$
$%
T
U
T Tg.Surat
$
U
%
T
$
Sg.Kim Kim
T
Sg.Johor$
T Sg.Belungkor
$
T $
$
T
T
$
'W
'W
T
$
T
$
P.Tekong $
T
P.Ubin
T
$
T
$
Sg.Santi
T
$
T
$
Tg.Pengelih
T
$
Sg.Sebina
Figure 4.8: Locations of intensive survey stations along Sungai Johor.
Water quality data from the Sungai Johor were used to investigate the
relationship between the longitudinal distribution of salinity, temperature and DO.
Figure 4.8 shows the location of longitudinal sampling stations along Sungai Johor
estuary. The range of values of longitudinal distribution of salinity, temperature and
DO parameters are shown in Tables 4.7 and Figures 4.9- 4.11.
60
Table 4.8: Data of in situ water quality parameters from Tg Pengelih to Kota Tinggi
Station
ID
Distance from
P01 (km)
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
19
23
25
29
31
35
41
43
47
49
51
54
56
58
60
64
66
68
69
Salinity (ppt)
P01
P02
P03
P04
P05
P06
P07
P08
P09
P10
P11
P12
P13
P14
P15
P16
P17
J20
J19
J18
J17
J16
J15
J14
J13
J12
J11
J10
J09
J08
J07
J06
J05
J04
J03
J02
J01
X
(m)
Y
(m)
676636
676905
677226
677506
677651
677431
676998
676654
675852
674913
673898
673112
672253
671267
671401
671605
672180
672045
670877
670513
670278
670229
669970
667963
664151
661732
660657
659026
660300
660025
658805
657511
657709
656933
656711
655237
653974
151434
152409
153063
154082
155266
156293
157177
157778
158491
159031
159507
160100
160913
161780
162666
163268
164121
164565
164725
167102
168822
169761
171454
174875
179470
181209
182401
183439
186519
187410
187825
189121
189990
189953
190964
190927
191435
min
28.04
28.40
28.08
27.85
28.00
27.78
27.87
27.98
27.89
27.96
27.98
27.99
27.84
27.86
27.81
27.82
27.85
27.74
27.52
27.63
27.47
27.62
27.71
28.19
28.01
28.02
28.10
27.80
27.51
27.10
26.61
26.44
26.39
26.34
26.36
26.45
26.53
Temperature
(°C)
max
average
28.30
28.14
28.72
28.55
28.33
28.17
28.03
27.91
28.60
28.23
28.40
28.10
28.64
28.15
28.15
28.05
27.95
27.92
27.96
27.96
27.99
27.99
28.70
28.24
27.98
27.90
28.46
28.06
28.33
28.04
28.33
28.02
28.45
28.09
27.85
27.79
27.76
27.68
27.85
27.73
27.82
27.60
27.87
27.71
27.99
27.89
28.25
28.21
28.32
28.16
28.46
28.31
28.43
28.29
28.17
28.01
28.32
27.90
27.19
27.14
27.01
26.76
27.04
26.70
26.54
26.45
26.46
26.38
26.36
26.36
26.45
26.45
26.54
26.54
min
22.25
21.71
27.35
27.09
25.77
25.41
26.13
25.36
25.54
25.18
24.64
23.62
23.71
21.32
21.45
21.12
20.61
27.42
26.52
26.15
25.79
26.11
26.85
24.39
21.39
19.82
20.14
15.44
10.84
5.66
1.73
1.04
0.67
0.15
0.07
0.08
0.17
Salinity
(ppt)
max
23.11
28.13
28.02
28.03
27.13
27.41
26.85
26.13
25.98
25.20
24.80
24.29
24.92
24.11
23.80
23.68
23.46
27.94
27.55
27.48
27.43
28.37
28.67
27.96
26.64
23.90
22.94
19.02
19.01
6.53
6.47
5.67
2.57
1.03
0.08
0.08
0.23
30
28
26
24
22
20
18
16
14
12
10
8
6
4
2
0
average
22.81
24.26
27.76
27.63
26.63
26.60
26.49
25.80
25.76
25.19
24.71
23.93
24.33
23.02
22.99
22.80
22.24
27.64
27.17
26.68
26.46
26.93
28.03
19.25
24.17
22.39
21.41
17.43
14.91
6.19
4.07
3.15
1.56
0.47
0.07
0.08
0.20
MIN
0
Tg Pengelih
10
20
30
40
50
Distance from Tanjung Pengelih (km)
60
MAX
DO
(mg/l)
max
8.79
7.18
6.60
6.66
6.16
6.10
5.86
5.79
5.71
5.53
5.57
5.41
5.44
5.68
5.49
5.39
5.45
3.93
4.25
4.30
4.33
4.36
4.60
4.84
5.26
5.26
5.45
5.28
5.38
6.28
7.01
8.85
7.96
6.44
3.41
3.61
3.60
min
7.19
6.70
6.33
6.12
5.99
5.84
5.67
5.59
5.51
5.37
5.31
5.32
5.01
4.83
4.84
4.80
4.73
3.80
3.97
4.09
4.23
4.01
4.03
2.12
4.79
4.60
4.69
4.70
4.27
5.90
4.31
4.18
5.01
3.51
3.36
3.49
3.53
MEAN
70
80
Kota Tinggi
Figure 4.9: Trend of salinity from Tanjung Pengelih (downstream) to Kota Tinggi
(upstream)
average
7.86
6.96
6.44
6.32
6.07
5.96
5.75
5.67
5.60
5.45
5.40
5.35
5.17
5.18
5.13
5.05
5.03
3.88
4.09
4.21
4.27
4.22
4.27
3.76
5.02
4.85
5.04
4.94
4.83
6.11
5.54
6.04
6.44
4.58
3.38
3.54
3.56
61
30
MIN
MAX
MEAN
Temperature (°C)
29
28
27
26
25
0
10
20
30
40
50
60
Distance from Tanjung Pengelih (km)
Tg Pengelih
70
80
Kota Tinggi
Figure 4.10: Trend of temperature from Tanjung Pengelih (downstream) to Kota
Tinggi (upstream)
10
MIN
MAX
MEAN
Dissolved oxygen (mg/l)
9
8
7
6
5
4
3
2
Lower Sungai Johor
1
Upper Sungai Johor
0
0
10
Tg Pengelih
20
30
40
50
Distance from Tanjung Pengelih (km)
60
70
80
Kota Tinggi
Figure 4.11: Trend of dissolved oxygen from Tanjung Pengelih (downstream) to
Kota Tinggi (upstream)
As shown in Figure 4.11, dissolved oxygen (DO) at upper and lower Sungai
Johor estuary can be described as low DO a consequence of high nutrient enrichment
as shown in Figure 4.12 and Figure 4.13 and becomes one of the most prominent
stressor of estuarine and coastal aquatic biota. This condition of low DO is known as
hypoxia.
62
Dissolved oxygen (DO) deficiency, as an index of deteriorated water quality,
has been widely observed in estuarine and coastal waters.
Hypoxia (deficient
dissolved oxygen) has been observed frequently in the bottom water of the lower
portion of the Rappahannock River tidal, a western shore tributary of Chesapeake
Bay (Park et al., 1996). Several studies have been conducted to understand the
controlling processes for hypoxia in Malaysian estuaries system such as Sungai
Langat estuary and Sungai Skudai estuary (Hashim et al., 2005).
Oxygen depletion was linked to nutrient loading such as eutrophication
through the accumulation, deposition and decomposition of phytoplankton biomass
(Smith et al., 1992). As stated by Vollenweider et al. (1992), the basic cause of
nutrient problem in estuaries and near shore coastal waters is the enrichment of
freshwater with nitrogen and phosphorus on their way to the sea and by direct input
within the tidal systems.
Coastal water receive pollutants either through direct
discharges from coastal activities or indirectly through rivers which may be polluted
from inland activities. Figure 4.12 shows the nitrogen (N) and phosphorus (P) loads
from upper boundary of Sungai Johor estuary at Kampung Rantau Panjang. The
most common effect of increased N and P supplies to aquatic ecosystems is an
increase in the abundance of algae and aquatic plants. As stated by Smith (2003),
these nutrient inputs have had profound negative effects upon the quality of surface
waters worldwide. The maximum value of N load at Rantau Panjang in April 2006 is
3292 kg/day and minimum value of N load in August 2006 is 270 kg/day. The
maximum value of P load in April 2006 is 32 kg/day and minimum value of P load in
February 2006 is 10 kg/day.
According to DHI Water and Environment (2004), Chlorophyll a
measurements show relative high values with 15-20 µg/l for Malaysian rivers and 2540 µg/l in Singapore rivers, which could be due to the growth of phytoplankton in the
upstream freshwater system. The nutrient levels in the same rivers are also high.
63
The maximum, minimum and average values of nitrate nitrogen, ammonia
nitrogen and phosphorus concentrations of Sungai Johor tributaries are shown in
Figure 4.13.
Nitrate nitrogen concentration is high according to Interim Water
Quality Standards (INWQS) at Sungai Serai and Sungai Berangan with average
values of 3.94 mg/l and 4.85 mg/l, respectively. Ammonia nitrogen concentration is
high in value at Sungai Serai, Sungai Tiram and Sungai Berangan with average
values of 6.14 mg/l, 2.08 mg/l, and 3.30 mg/l, respectively. Phosphate concentration
is high in value at Sungai Serai and Sungai Berangan with average values of 0.52
mg/l and 0.45 mg/l, respectively. The temporal profiles of DO concentrations,
nitrogen and phosphorus loads of Sungai Johor tributaries from year 1996 to 2007
are shown in Appendix D.
2006
3500
350
3000
300
2500
250
2000
200
1500
150
1000
100
500
50
0
Jan
(b)
0
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
2006
40
P load (kg/day)
400
Flow (m3/s)
N load (kg/day)
4000
400
35
350
30
300
25
250
20
200
15
150
10
100
5
50
0
Jan
Flow (m3/s)
(a)
0
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Figure 4.12: Nitrogen and Phosphorus Loads at Kampung Rantau Panjang
Se
ra
i
Ti
r
Se am
m
an
ge
Be
r
ra
ng
an
Te
m
on
La
ya
u
Le
ba
m
La
ya
ng
NH3-N (mg/l)
River
(c)
River
Sa
nt
i
Pa
pa
n
Se
lu
yu
t
Pe
R
an
le
pa
ta
u
h
Pa
nj
an
g
Se
ra
i
Ti
r
Se am
m
an
ge
Be
r
ra
ng
an
Te
m
on
La
ya
u
Le
ba
m
La
ya
ng
NO3 (mg/l)
(a)
(b)
Sa
nt
i
Pa
pa
n
Se
lu
yu
t
Pe
R
le
an
p
ta
ah
u
Pa
nj
an
g
River
Sa
nt
i
Pa
pa
n
Se
lu
yu
Pe t
R
an
le
pa
ta
u
h
Pa
nj
an
g
La
ya
ng
Se
ra
i
Ti
ra
m
Se
m
an
Be ger
ra
ng
an
Te
m
on
La
ya
u
Le
ba
m
PO4 (mg/l)
64
MIN
MIN
MIN
(a) Nitrate nitrogen (b) Ammonia nitrogen (c) Phosphate
MAX
MAX
MEAN
14
12
10
8
6
4
2
0
MAX
MEAN
16
14
12
10
8
6
4
2
0
2.00
MEAN
1.50
1.00
0.50
0.00
Figure 4.13: Water Quality Concentration in Sungai Johor tributaries.
65
4.6
Discrete Water Chemistry Samples
The sampling program consists of water sample collection at each of the
sampling locations along Sungai Johor estuary system. Laboratory data analysis has
been conducted by the Faculty of Science, UTM Skudai. Figure 4.14 shows the water
quality concentration trend along Sungai Johor estuary system from downstream
boundary at Tanjung Pengelih to upstream boundary at Kota Tinggi.
Figure 4.14(a) shows the profile of BOD concentration along Sungai Johor
estuary system. The maximum value of BOD concentration is 1.1 mg/l and minimum
value of BOD concentration is 0.9 mg/l. The maximum value of ammonia nitrogen
concentration is 0.21 mg/l and minimum value is 0.07 mg/l. The maximum value of
nitrate nitrogen concentration is 1.3 mg/l and minimum value is 0.1 mg/l. The
maximum value of phosphate concentration is 0.47 mg/l and minimum value is 0.18
mg/l.
(a)
2.0
WQ01 WQ02
WQ03
WQ04 WQ05 WQ06
WQ07
WQ08
1.8
1.6
BOD5 (mg/l)
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50
Tanjung Pengelih
distance (km )
Kota Tinggi
66
(b)
1.0
WQ01 WQ02
WQ03
WQ04 WQ05 WQ06
WQ07
WQ08
0.9
0.8
NH4-N (mg/l)
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50
distance (km )
Tanjung Pengelih
(c)
2.0
WQ01 WQ02
WQ03
WQ04 WQ05 WQ06
Kota Tinggi
WQ07
WQ08
1.8
NO3-N (mg/l)
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50
distance (km )
Tanjung Pengelih
(d)
1.0
WQ01 WQ02
WQ03
WQ04 WQ05 WQ06
Kota Tinggi
WQ07
WQ08
0.9
0.8
PO4 (mg/l)
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50
Tanjung Pengelih
distance (km )
Kota Tinggi
Figure 4.14: Water quality concentration along Sungai Johor estuary system
a) BOD b) NH3-N c) NO3-N and d) PO4 concentrations
67
4.7
Continuous Monitoring Data
For the purpose of model calibration, an intensive field-monitoring to
quantify the hydrodynamic conditions within the Sungai Johor Estuary was
conducted on December 2007 to February 2008. Under this monitoring effort, a total
of five continuous monitoring stations measured water surface elevation, salinity, DO
and temperature.
The continuous monitoring stations consists of permanently
mounted instruments which recorded a 10-minute intervals of data. The instruments
were placed near the bottom, approximately 3 feet above the river bed. The stations
are located at Teluk Sengat, Johor Lama, Tanjung Serindet, Tanjung Buai and
Tanjung Surat as shown in Figure 4.15 and summarized in Table 4.9.
Table 4.9: Summary of Continuous Monitoring Stations of Sungai Johor
Location
Install date
Removal date
Duration
Johor Lama
13/12/2007
06/01/2008
25 days
Tanjung Buai
09/02/2008
08/02/2008
29 days
Tanjung Serindet
13/02/2008
06/02/2008
25 days
Tanjung Surat
13/12/2007
27/12/2007
15 days
13/01/2008
12/02/2008
31 days
13/12/2007
05/01/2008
24 days
13/01/2008
12/02/2008
31 days
Teluk Sengat
68
Sg.Johor
Rantau Panjang
Vertical profile stations
Continuous monitoring
stations
Kota Tinggi
Sg.Seluyut
N
W
Sg.Berangan
E
S
Sg.Redan
Sg.Temon
#
S
V16
Johor Lama
Sg.Tiram
Sg.Papan
V15
#
S
Sg.Layang
Teluk Sengat
V14
Sg.Chemangar
Sg.Layau
#
S
Sg.Serai
V11
#
S
Sg.Tebrau
#
S
V12
#
S
V13
Tg.Serindet
Sg.Lebam
V10#S
Sg.Kim Kim
Pasir Gudang
Sg.Buloh
V5
#
S
Sg.Johor
Tg.Surat
Sg.Belungkor
V4#S
#
S
V9
#
S
V8
#
S
V7
P.Ubin
#
SV6
V3
#
S
P.Tekong
Sg.Sebina
V2#S
Sg.Santi
Tg.Pengelih
V1 #S
Figure 4.15: Locations of vertical profile and continuous monitoring stations along
Sungai Johor
4.8
Vertical Profiles
Vertical profiles of salinity, temperature and dissolved oxygen were taken at
each sampling stations as shown in Figure 4.15. Data were grouped with a one meter
vertical resolution. Cruises typically started from Station Tanjung Pengelih and
proceeded north towards Kota Tinggi or east to Pasir Gudang.
The vertical salinity distribution characterized the degree of stratification and
controled the vertical mixing process. A direct relationship between the occurrence
of stratification and the depletion of oxygen below the pycnocline was found
69
throughtout the literature for many estuaries (Reynolds Fleming, 2003).
Stratification separated the water column into vertical sections where the bottom
layer was effectively cut off from reaeration due to the lack of mixing processes.
This separation allowed benthic respiration to deplete DO in the lower layers.
Increased benthic respiration and hence DO depletion were directly attributed to
eutrophication.
Hypoxia was defined as the reduction of DO concentration to less
than 4 mg/l for a period lasting 24 hours or more (Thursby et al., 1999). The term
severe hypoxia represented DO values less than 2 mg/l.
Figure 4.16 which shows the measured vertical profiles of salinity,
temperature and dissolved oxygen at Tanjung Pengelih, Tanjung Buai, Teluk Sengat,
Pasir Gudang Port and Tanjung Belungkor stations. The profiles were measured on
the same day as the longitudinal survey illustrated in Figure 4.8.
The salinity
increases with depth, while oxygen decreases with depth. The temperature does not
changes with depth.
The plots below indicate that the circulation pattern in a
stratified estuarine system sustains the depleted oxygen condition, making estuaries
fundamentally different in their oxygen budgets from freshwater rivers.
Station: WQ01-Tg Pengelih
20
Salinity (ppt) and tem perature (°C)
22
24
26
28
30
32
0
2
4
Depth (m)
(a)
6
8
10
temperature
salinity
DO
12
14
0
1
2
3
4
5
Dissolved oxygen (m g/l)
6
7
8
70
(b)
Station: Tg Buai
Salinity (ppt) and tem perature (°C)
20
22
24
26
28
30
32
0
1
2
Depth (m)
3
4
5
6
7
temperature
salinity
DO
8
9
0
2
Station: WQ07- Teluk Sengat
20
3
4
5
Dissolved oxygen (m g/l)
6
7
8
Salinity (ppt) and tem perature (°C)
22
24
26
28
30
32
0
1
2
Depth (m)
(c)
1
3
4
5
temperature
salinity
DO
6
7
0
1
2
3
4
5
Dissolved oxygen (m g/l)
6
7
8
71
(d)
Station: WQ05- Pasir Gudang Port
20
22
Salinity (ppt) and tem perature (°C)
24
26
28
30
32
0
2
Depth (m)
4
6
8
10
temperature
salinity
DO
12
14
0
2
Station: Belungkor
3
4
5
Dissolved oxygen (m g/l)
6
22
24
26
8
Date: 25 April 2008
Salinity (ppt) and tem perature (°C)
20
7
28
30
32
0
temperature
salinity
2
DO
4
Depth (m)
(e)
1
6
8
10
12
14
0
1
2
3
4
5
Dissolved oxygen (m g/l)
6
7
8
Figure 4.16: Vertical profiles of observed salinity, temperature and DO at Tanjung
Pengelih, Tanjung Buai, Teluk Sengat, Pasir Gudang Port and Tanjung Belungkor
stations
72
CHAPTER 5
RESULTS AND DISCUSSIONS
5.0
Introduction
Calibration resulted in a consistent set of model coefficients that reproduced
the observed data for all state variables considered. All model coefficients should be
consistent between the calibration period and the verification period. The method
used in determining the values for the model coefficients is essentially one of trial
and error. The criteria for selecting an appropriate calibration and verification data
set are adequate temporal and spatial coverage, and available data for all variables
considered in the computation. An intensive survey conducted in January 14-30,
2008 was used as calibration data set. This calibration data set was selected because
of the availability of a comprehensive set of data and adequate description of
boundary conditions during the study period.
Hsu et al. (1999) stated that it is an accepted requirement that a numerical
model of estuarine hydrodynamics be calibrated and verified before being put into
any practical usage. However, there is no widely accepted procedure for carrying out
these tasks.
Model calibration appears in various forms, dependent on data
availability, characteristics of water body, and most of all, the perceptions and
opinions of modelers. It often happens that calibrated model results were compared
with observed data to demonstrate agreements, with little explanation of which
coefficients were adjusted in arriving at the calibrated model.
73
Wang et al. (1990) reported that a model was initially calibrated to surface
elevations and currents of astronomical tides, followed by calibration of exchange
processes and turbulence closures.
They further emphasized the quantitative
measure of model performance in comparison with prototype data but they did not
propose any procedure for carrying out model calibration and verification. Hsu et al.
(1999) calibrated their vertical (lateral averaged) 2D model in three steps;
preliminary calibration, fine tune calibration of friction coefficient and calibration of
mixing processes. They recommended using the distribution of the tidal range as a
function of distance from the river mouth to calibrate the friction coefficient.
Hsu et al. (1999) gave a rational approach to calibrate and verify a
hydrodynamic model of partially stratified estuaries. They applied their vertical
(laterally averaged) 2D model to the tidal Tanshui River system, which is the largest
tidal river in Taiwan. The friction coefficient was calibrated and verified with model
simulation of barotropic flow, and the turbulent diffusion and dispersion coefficients
were calibrated through comparison of salinity distribution.
5.1
Hydrodynamic Model Calibration
The calibration methodology for the Sungai Johor EFDC model included
graphical time series comparisons (qualitative) and statistical calculations
(quantitative). The statistical calculations include a variety of statistical calculations.
The calibration methodology is also parameter specific starting with the following:
i.
Water Surface Elevation
ii.
Temperature
iii.
Salinity
Each one of these parameters has its importance in the determination of
success for the model calibration and verification.
The order in which the
hydrodynamic model is calibrated is performed to address issues such as bathymetry,
friction, tidal volume, cross-sectional area, and heat budget before salinity is
74
calibrated. Salinity is the predominant signal in the model to ensure that mass is
being moved horizontally and vertically with the appropriate timing and direction.
5.1.1
Water Surface Elevation Calibration
The water surface calibration was performed by modifying the downstream
elevation boundary until the model closely fits the data at Tanjung Surat. Then, the
bottom roughness was modified to reach an appropriate phase shift of the elevation
signal at each of the stations.
The model was executed using the boundary conditions for the calibration
period of 30 days (January 1-30, 2008). During the calibration process, bottom
roughness was determined to be the most influential modeling parameter for
hydrodynamic modeling (Hashim, 2001; Liu et al., 2008). Bottom roughness values
were adjusted until the predicted results reasonably matched the observed data. After
several adjustments, a bottom roughness of 2 cm was selected for the use in this
study. The model results were compared against observed data in the study period
(January 14-30, 2008). The temporal profiles of observed and predicted tide level
are compared in for the calibration period. As shown in Figures 5.1 and 5.2, the
predicted tide levels reasonably matched the observed data at two sampling stations:
Station Teluk Sengat and Tanjung Surat. These figures indicate that the model
reasonably simulated the tide range and phase at a number of locations throughout
Sungai Johor estuary.
A study by Mohd Zaharifudin (2008) showed that after several adjustments, a
bottom roughness of 4 cm was selected for use in upper Sungai Johor near Kota
Tinggi Bridge. The modeled results were compared against DID data in his study
period (July 1-31, 2006).
The simulated water level showed a good and fairly good agreement with
observations, indicated by the low values of root mean square as shown in Figures
5.1, 5.2, and 5.3 respectively and summarized in Table 5.1.
75
Table 5.1: Error analysis for observed and simulated water surface elevations
Station
Average error
Average absolute
Root Mean Square
difference
error difference
difference
Tanjung Surat
0.141
0.249
0.321
Teluk Sengat
0.136
0.231
0.281
Kota Tinggi
0.002
0.421
0.500
DS-INTL
DS-INTL
Water Quality Modeling for Sg Johor - Min value
Calibration Results: Tanjung Surat (January 2008)
3.50
3.25
3.00
Water Surface Elevation (m)
2.75
2.50
2.25
2.00
1.75
1.50
1.25
1.00
Legend
0.75
tgsurat_WSE-Model
tgsurat_WSE-Data
0.50
0.25
0.00
13
15
17
19
21
23
25
27
29
31
Time (days)
DS-INTL
DS-INTL
Figure 5.1: Observed and simulated water surface elevations at Station Tanjung Surat
DS-INTL
DS-INTL
Water Quality Modeling for Sg Johor - Min value
Calibration Results: Teluk Sengat (January 2008)
3.50
3.25
3.00
Water Surface Elevation (m)
2.75
2.50
2.25
2.00
1.75
1.50
1.25
1.00
0.75
Legend
0.50
tlksengat_WSE-Model
tlksengat_WSE-Data
0.25
0.00
13
15
17
19
21
23
25
27
29
31
Time (days)
DS-INTL
DS-INTL
Figure 5.2: Observed and simulated water surface elevations at Station Teluk Sengat
76
Figure 5.3: Observed and simulated water surface elevations at Kota Tinggi
(Mohd Zaharifudin, 2008).
5.1.2
Temperature Calibration
Temperature was the hydrodynamic parameter with the least calibration. The
temperature data were used at Tanjung Pengelih for the downstream boundary. The
summary statistics, time series plots of the draft calibration for temperature at Station
Tanjung Surat and Teluk Sengat are shown in Table 5.2 and Figures 5.4-5.5,
respectively.
The simulated water temperature showed good and fairly good agreement
with observations, indicated by the low values of root mean square as shown in
Figures 5.4 and 5.5 and summarized in Table 5.2.
77
Table 5.2: Error analysis for observed and simulated water temperature
Station
Average error
Average absolute error
RMS
Tanjung Surat
0.595
0.941
1.126
Teluk Sengat
-0.158
0.428
0.524
Figures 5.6 and 5.7 present the longitudinal comparisons of the simulated
water temperature and the measured data. The pattern of water temperature along
Sungai Johor can be seen in Figure 5.6 from Tanjung Pengelih to Johor Lama. The
pattern of water temperature along Johor Straits can be seen in Figure 5.7 from
Tanjung Kopok to Sungai Tebrau estuary. The simulated water temperature is
compared with DOE marine water quality data at Station SJ1, SJ2, SJ4 and SJ4A.
NTL
DS-
Temperature Calibration for Sg Johor
Station: Tanjung Surat
35
34
33
Legend
Measured
Simulated
Temperature (°C)
32
31
30
29
28
27
26
25
14
15
16
17
18
19
20
21
22
Time (days)
NTL
DS-
Figure 5.4: Observed and Simulated Temperatures at Station Tanjung Surat
78
Figure 5.5: Observed and Simulated Temperatures at Station Teluk Sengat
measured
simulated
35 WQ01
PF03
BLKR
WQ02
WQ03
WQ06
WQ08
WQ07
Temperature,°C
30
25
20
15
10
0
2
4
Tanjung Pengelih
6
8
10
12
14
16
18
20
22
distance, km
Figure 5.6: Temperature profile along Sungai Johor
24
26
28
Johor Lama
79
MIN
SJ1
MAX
MEAN
model
34
SJ2
Temperature (°C)
33
SJ4A
32
31
30
29
28
27
0
2
4
6
Kg Tg Kopok
8
10
12
14
16
18
20
distance from Tg Kopok
22
24
26
28
Sg Tebrau Estuary
Figure 5.7: Temperature profile along Johor Straits
5.1.3
Salinity Calibration
In most estuaries with a significant freshwater discharge, salinity may serve
as an ideal natural tracer for mixing processes calibration. Salinity distribution in an
estuary is affected by the tidal current, freshwater discharge, density circulation, as
well as turbulent mixing processes. Therefore the salinity distribution reflects the
combined results of all processes, and in turn it controls density circulation and
modifies mixing processes.
Salinity is the key parameter of concern because it can dictate how well the
model is transporting mass in the system. Hsu et al. (1999) calibrated the coefficient
in the turbulence model by matching the observed and computed salinity
distributions. The summary statistic and time series plot of the preliminary
calibration for salinity are shown in Figure 5.8 and Table 5.3.
80
DS-INTL
DS-IN
Salinity Calibration for Station Tanjung Surat
Calibration Results: Time Series Summary
30
28
Legend
Simulated
Measured
Salinity (ppt)
26
24
22
20
14
16
18
20
22
24
26
28
30
Time (days)
Figure 5.8: Observed and Simulated Salinity at Station Tanjung Surat
To provide a quantitative assessment of the EFDC predictive capability, each
time series comparison underwent a simple statistical treatment to determine a mean
difference, mean absolute difference, and root mean square difference as shown in
Table 5.3.
Based on the error analysis as shown in Table 5.3, the model is
underpredicted in response to the natural system at Station Tanjung Surat. This might
be due to the errors in the bathymetry data or the errors associated to the prescribed
boundary conditions as stated by Liu et al., (2008).
In Figure 5.9, the model
reproduced seasonal trends of measured salinity for the Sungai Johor estuarine
system. Figure 5.10 present the longitudinal comparison of the simulated salinity
against the DOE marine water quality monitoring data at SJ1, SJ2, SJ4 and SJ4A
stations.
Table 5.3: Error analysis for observed and simulated salinity
Station
Tanjung Surat
Average error
-0.495
Average absolute error
1.026
RMS
1.206
81
measured
35
WQ01
PF03
30
WQ02
simulated
WQ06
BLKR
WQ03
WQ07
WQ08
Salinity, ppt
25
20
WQ09
15
10
5
0
-5
0
2
4
6
8
10
12
14
16
18
20
22
24
26
28
30
32
34
36
38
40
distance, km
Figure 5.9: Salinity profile along Sungai Johor
MIN
SJ1
MAX
MEAN
model
30
Salinity (ppt)
28
SJ2
26
24
SJ4A
22
20
18
0
Kg Tg Kopok
2
4
6
8
10
12
14
16
18
distance from Tg Kopok
20
22
24
26
28
Sg Tebrau Estuary
Figure 5.10: Salinity profile along Johor Straits
5.2
Water Quality Model Calibration
The measured values from the collected survey data were used for calibrating
the EFDC water quality model. The data consists of dissolved oxygen, BOD,
ammonia, other nutrients, and chlorophyll-a concentrations. Specifically, for the
EFDC model calibration, dissolved oxygen concentration at Station Tanjung Surat
was used.
82
5.2.1
Dissolved Oxygen
Predicted and measured DO concentration was compared at Station Tanjung
Surat as presented in Figure 5.11. To provide a quantitative assessment of the EFDC
predictive capability, time series comparison underwent a simple statistical treatment
to determine a mean difference, mean absolute difference, and root mean square
difference as shown in Table 5.4. DO calibration result as presented in Table 5.4
shows that the response of the natural system of the DO concentration at Station
Tanjung Surat was overpredicted.
Table 5.4: Error analysis for observed and simulated dissolved oxygen
Station
Tanjung Surat
Average error
Average absolute error
0.795
0.826
RMS
0.949
Dissolved oxygen longitudinal profile along Sungai Johor as shown in Figure
5.12 shows that DO concentration increased at upper Sungai Johor estuary and
decreased at lower Sungai Johor estuary. A component analysis for DO indicated
that, on average, the major source of DO was reaeration, algal production, while
sinks included sediment oxygen demand (SOD) and respiration.
Figure 5.13 present the longitudinal comparison of the simulated dissolved
oxygen concentration to the DOE marine water quality monitoring data at SJ1, SJ2,
SJ4 and SJ4A stations.
83
INTL
DS-IN
DO calibration profile for Sungai Johor
Station: Tanjung Surat
8
7
Legend
Dissolved Oxygen (mg/l)
6
measured
data
5
4
3
2
1
14
15
16
17
18
19
20
21
22
Time (days)
DS-IN
INTL
Figure 5.11: Observed and Simulated DO at Station Tanjung Surat
measured
simulated
10
9
8
DO, mg/l
7
6 WQ01
PF03 WQ02 BLKR
5
WQ09
WQ03
WQ06
WQ07
4
WQ08
3
2
1
0
2
Tg Pengelih
4
6
8
10
12 14 16
18 20
22 24
26 28
30 32 34
36 38
distance, km
Figure 5.12: DO concentration profile along Sungai Johor
40 42
44 46
Kota Tinggi
84
MIN
Dissolved oxygen (mg/l)
SJ1
10
9
8
7
6
5
4
3
2
1
0
MAX
MEAN
model
SJ2
0
2
4
6
8
SJ4A
10
12
14
16
18
20
24
26
28
Sg Tebrau Estuary
distance from Tg Kopok
Kg Tg Kopok
22
Figure 5.13: Dissolved oxygen concentration profile along Johor Straits
5.2.2
Ammonia Nitrogen
Model prediction and simulation for ammonia nitrogen concentration were
compared as shown in Figure 5.14. The predicted ammonia nitrogen was compared
to the measured values from the sampling data in January 2008. The longitudinal
profile of simulated and observed ammonia nitrogen are shown in Figure 5.14. The
predicted ammonia nitrogen is in the range of observed data.
NH4-N longitudinal profile along Sungai Johor
1.0
measured
0.9
simulated
0.8
NH4-N, mg/
0.7
0.6
0.5
0.4
0.3
0.2
WQ01
WQ02
0.1
WQ03
WQ06
WQ07
WQ08
WQ09
WQ10
0.0
0
2
4
6
Tanjung Pengelih
8
10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50
distance, km
Kota Tinggi
Figure 5.14: Ammonia nitrogen concentration profile along Sungai Johor
85
5.2.3
Nitrate Nitrogen
Model prediction and simulation for nitrate nitrogen concentration were
compared as shown in Figure 5.15.
The longitudinal profile of simulated and
observed nitrate nitrogen are shown in Figure 5.15. The predicted nitrate nitrogen is
in the range of observed data.
NO3-N longitudinal profile along Sungai Johor
NO3-N, mg/
measured
2.0
1.8
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
simulated
WQ10
WQ06
WQ07
WQ02
WQ03
WQ08
WQ09
WQ01
0
2
4
6
8
10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50
Tanjung Pengelih
distance, km
Kota Tinggi
Figure 5.15: Nitrate nitrogen concentration profile along Sungai Johor
5.2.6
Total Phosphate
Model predictions and simulations for total phosphorus concentration were
compared as shown in Figure 5.16.
The longitudinal profile of simulated and
observed phosphate are shown in Figure 5.14. The predicted phosphate is in the
range of observed data.
86
PO4 longitudinal profile along Sungai Johor
measured
simulated
1.0
PO4, mg/l
0.8
0.6
0.4
WQ08
WQ01
WQ02
0.2
WQ03
WQ06
WQ07
WQ10
WQ09
0.0
0
2
4
6
8
10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50
distance, km
Figure 5.16: Total Phosphate concentration profile along Sungai Johor
5.2.7
Chlorophyll a
According to DHI Water and Environment (2004), Chlorophyll a
measurements show relatively high values (15-20 µg/l for Malaysian rivers and 2540 µg/l for Singapore rivers) which could be due to the growth of phytoplankton in
the upstream freshwater system. The nutrient levels in the same rivers are also high.
Mixing with marine-based Chlorophyll a from eutropied waters is also a possible
factor for the high readings.
Model simulation for Chlorophyll a concentration along Sungai Johor is
shown in Figure 5.17. As shown in Figure 5.17, Chlorophyll a concentration along
Sungai Johor is affected by the change of maximum algal growth rate.
87
baseline
algal gro wth rate=1
algal gro wth rate=3
10
9
8
Chl a (ug/l)
7
6
5
4
3
2
1
0
0
5
Tg P engelih
10
15
20
25
30
35
40
distance, km
45
Ko ta Tinggi
Figure 5.17: Chlorophyll a concentration profile along Sungai Johor
5.3
Hydrodynamic Model Sensitivity Analysis
Through model calibration, the sensitivity of each parameters can be
identified in EFDC model. Obviously, bathymetry is one of the most critical pieces
of information for the model and a significant amount of time and effort were placed
on capturing the bathymetry in the model with limited smoothing and averaging. As
for the other model inputs and calibration parameters, the Sungai Johor upstream
flow, downstream/ocean salinity boundary, and bottom roughness rank as the most
sensitive parameters for the EFDC model. The horizontal eddy viscosity was also run
for comparisons. These parameters were changed in the EFDC model while holding
the bathymetry constant.
5.3.1
Sensitivity of Water Surface Elevation to Bottom Roughness
The EFDC model uses bottom roughness to account for the energy loss due to
the friction exerted from the sides and bottom of the channel and any other forms of
energy loss, such as meandering channels. Based on the sensitivity analysis, the
bottom roughness was constant throughout most of the model domain. Bottom
88
roughness is very sensitive to water surface elevation at upper Sungai Johor and
insensitive to lower Sungai Johor as presented in Tables 5.5, 5.6, and 5.7,
respectively.
Table 5.5: Sensitivity analysis of bottom roughness with water surface elevation at
Station Tanjung Surat
Scenarios
At Station
Tanjung Surat
Average error
difference
Average absolute
error difference
Root Mean Square,
RMS difference
Z= 0.01
Z= 0.02
Z= 0.03
Z= 0.04
0.143
0.141
0.145
0.143
0.252
0.249
0.252
0.251
0.327
0.321
0.324
0.323
Table 5.6: Sensitivity analysis of bottom roughness with water surface elevation at
Station Teluk Sengat
Scenarios
At Station
Teluk Sengat
Average error
difference
Z= 0.01
Z= 0.02
Z= 0.03
Z= 0.04
0.135
0.136
0.136
0.136
Average absolute Root Mean Square,
error difference
RMS difference
0.236
0.231
0.234
0.235
0.288
0.281
0.286
0.287
Table 5.7: Sensitivity analysis of bottom roughness with water surface elevation at
Station Kota Tinggi (Mohd. Zaharifudin, 2008)
Scenarios
At Station Kota Tinggi
Average error
difference
Average absolute
error difference
Root Mean
Square, RMS
difference
Z= 0.02
Z= 0.03
Z= 0.04
Z= 0.05
Z= 0.06
Z= 0.07
2.374
0.932
0.002
0.008
0.015
0.021
2.387
1.129
0.421
0.440
0.458
0.474
2.498
1.408
0.500
0.523
0.543
0.561
89
5.3.2
Sensitivity of Salinity to Bottom Roughness
Based on the error analysis at Tanjung Surat as shown in Table 5.8 and
Figure 5.18, mean error values are less than zero. It shows that the model is
underpredicted in response to the natural system.
Table 5.8: Error analysis for salinity sensitivity analysis with bottom roughness at
Station Tanjung Surat
Station
Average error
Average absolute error
RMS
Z= 0.01
-0.355
0.987
1.203
Z= 0.02
-0.314
0.947
1.158
Z= 0.03
-0.410
0.964
1.191
Z= 0.04
-0.450
1.002
1.243
DS-INTL
DS-INTL
Salinity sensitivity analysis with bottom roughness
Station: Tanjung Surat
26
25
24
Salinity (ppt)
23
22
21
Legend
20
z= 0.02
z=0.01
z=0.03
z=0.04
19
18
14.00
14.80
15.60
16.40
17.20
18.00
18.80
19.60
20.40
21.20
22.00
Time (days)
DS-INTL
DS-INTL
Figure 5.18: Sensitivity of salinity to bottom roughness at Station Tanjung Surat
5.3.3
Sensitivity of Salinity to Horizontal Eddy Viscosity
The EFDC model was
sensitive to the horizontal eddy viscosity. The
calibration run was made with the horizontal eddy viscosity at 25 m2/s. This
90
parameter was varied from 0.005 m2/s to 75 m2/s. The profile of various values of
horizontal eddy viscosity at Station Tanjung Surat are shown in Tables 5.9 and
Figure 5.19.
DS-INTL
DS-INTL
Salinity Calibration for Sg Johor - horizontal eddy visc=25, decrease water temp
Calibration Results: Time Series Summary
25
24
Salinity (ppt)
23
22
Legend
sal1_eddy0.005
sal1_eddy50
sal1_eddy75
sal1_eddy0.01
sal1_eddy0.05
21
20
14
15
16
17
18
19
20
21
22
Time (days)
DS-INTL
DS-INTL
Figure 5.19: Sensitivity of salinity to horizontal eddy viscosity at Station Tanjung
Surat
Table 5.9: Error analysis for salinity sensitivity analysis with horizontal eddy
viscosity at Station Tanjung Surat
Scenarios
Horizontal eddy
viscosity
0.005
0.01
25.00
50.00
75.00
RMS error
Average
absolute error
Average error
1.145
1.534
0.428
1.451
1.491
0.797
1.090
0.329
1.160
1.205
0.509
0.734
0.291
-0.441
-0.488
91
5.3.4 Sensitivity of Salinity to Upstream Boundary Condition
As shown in Figure 5.20, the salinity concentrations are sensitive to the
upstream boundary condition at Rantau Panjang. The model was run for the worse
case scenario during the flood event at Kota Tinggi in January 2007. The freshwater
inflows from upstream boundary at Rantau Panjang has significantly changed the
salinity concentrations at upper Sungai Johor estuarine system.
no rmal flo w
heavy flow
30
salinity, ppt
25
20
15
10
5
0
0
Tg Pengelih
5
10
15
20
25
30
35
distance, km
40
45
Kota Tinggi
Figure 5.20: Sensitivity of salinity to upstream boundary conditions
5.3.5 Sensitivity of Salinity to Downstream Salinity Boundary
The downstream salinity boundary was increased by 1 ppt and 2 ppt when the
salinity boundary changed by 10% and 20% from the baseline run. As shown in
Figure 5.21, the salinity concentration is sensitive to dowstream salinity boundary.
92
10%
20%
28
26
24
salinity, ppt
22
20
18
16
14
12
10
8
6
4
2
0
0
5
10
15
Tg Pengelih
20
25
distance, km
30
35
40
45
Kota Tinggi
Figure 5.21: Sensitivity of salinity to downstream salinity boundary
5.4
Water Quality Model Sensitivity Analysis
Since a mathematical model is a simplified representation of the real world,
its prediction is often subject to considerable uncertainty from a variety of sources.
These sources include over simplification of modeling assumptions and formulations,
noise-distorted data, and model parameter values. It is important to gain a better
understanding of a model’s reliability by analyzing the uncertainties associated with
a model. Sensitivity analysis is a prime method in measuring a model’s uncertainty
and reliability. In this study, the sensitivity of the DO concentration to various
parameters was evaluated. The analysis was performed by assessing the impacts of
the following factors on DO:
i)
COD decay rates
ii)
Nitrification rates
iii)
Sediment oxygen demand (SOD)
iv)
Maximum algal growth rate
v)
Loads from point sources
vi)
Reaeration rate
93
Sensitivity tests were run on a baseline model simulation in January 2008.
The sensitivity of the dissolved oxygen to the parameters changes were estimated by
comparing it with the baseline model scenario. The sensitivity analysis results are
based upon the varying parameters are shown in tables and figures below. The
following sections discuss the results of the tested parameters.
5.4.1
Sensitivity to COD Decay Rate
The basic COD decay rate for the baseline run was 0.10 l/day. The variations
of the parameter were 0.04 and 0.40 1/day at Station Tanjung Surat and Teluk
Sengat. As shown in Figures 5.22 and 5.23, with the same magnitude of variation in
parameter values, the DO concentrations are sensitive to the COD decay rate. Table
5.10 and 5.11 show the error analysis of DO concentration to COD decay rate.
DS-INTL
DS-INTL
DO sensitivity analysis with COD decay rate
Station: Tanjung Surat
8
Legend
7
Dissolved Oxygen (mg/l)
6
KCD=0.1
KCD=0.4
KCD=0.04
KCD=0.06
KCD=0.08
5
4
3
2
1
14
15
16
17
18
19
20
21
22
Time (days)
DS-INTL
DS-INTL
Figure 5.22: Sensitivity of DO to COD decay rate at Station Tanjung Surat
94
Table 5.10: Error analysis of DO sensitivity analysis to COD decay rate at Station
Tanjung Surat
Scenarios
RMS error
Average absolute error
Average error
KCD = 0.10
0.571
0.406
0.380
KCD = 0.40
0.769
0.552
0.258
KCD=0.04
0.528
0.449
-0.373
KCD=0.06
0.365
0.283
-0.079
KCD= 0.08
0.374
0.265
0.134
DS-INTL
DS-INTL
DO sensitivity analysis with COD decay rate
Station: Teluk Sengat
8
7
Dissolved Oxygen (mg/l)
6
5
4
Legend
KCD=0.1
KCD=0.4
KCD=0.04
KCD=0.06
KCD=0.08
3
2
1
14
15
16
17
18
19
20
21
22
Time (days)
DS-INTL
DS-INTL
Figure 5.23: Sensitivity of DO to COD decay rate at Station Teluk Sengat
Table 5.11: Error analysis of DO sensitivity analysis to COD decay rate at Station
Teluk Sengat
Scenarios
RMS error
Average absolute error
Average error
KCD= 0.04
0.580
0.556
-0.556
KCD= 0.06
0.234
0.211
-0.211
KCD= 0.08
0.267
0.239
-0.237
KCD = 0.10
0.000
0.000
0.000
KCD = 0.40
0.645
0.533
0.075
95
5.4.2
Sensitivity of DO to Nitrification Rate
The basic nitrification rate for the baseline run was 0.07 1/day. The sensitivity
of the instream DO concentrations to the nitrification rate was analyzed by changing
the value of the parameter with 0.01 and 0.10 1/day. The resulting sensitivity for
Station Tanjung Surat and Teluk Sengat is shown in Figures 5.24 and 5.25,
respectively. DO regime in Sungai Johor estuary is not sensitive to this parameter
variation. Table 5.12 and 5.13 show the error analysis of DO concentration to
nitrification rate.
DS-INTL
DS-INTL
DO sensitivity analysis with Kn
Station: Tanjung Surat
8
Legend
7
Kn=0.07
Kn= 0.1
Kn=0.01
Dissolved Oxygen (mg/l)
6
5
4
3
2
1
0
14
15
16
17
18
19
20
21
22
Time (days)
DS-INTL
DS-INTL
Figure 5.24: Sensitivity of DO to nitrification rate at Station Tanjung Surat
Table 5.12: Error analysis of DO sensitivity analysis to nitrification rate at Station
Tanjung Surat
Scenarios
RMS error
Average absolute error
Average error
Kn = 0.07
0.296
0.224
0.202
Kn = 0.10
0.487
0.388
-0.022
Kn = 0.01
0.268
0.197
0.160
96
DS-INTL
DS-INTL
DO sensitivity analysis with nitrification rate, Kn
Station: Teluk Sengat
8
Legend
Kn=0.07
Kn=0.1
Kn=0.01
7
Dissolved Oxygen (mg/l)
6
5
4
3
2
1
0
14
15
16
17
18
19
20
21
22
Time (days)
DS-INTL
DS-INTL
Figure 5.25: Sensitivity of DO to nitrification rate at Station Teluk Sengat
Table 5.13: Error analysis of DO sensitivity analysis to nitrification rate at Station
Teluk Sengat
Scenarios
RMS error
Average absolute error
Average error
Kn = 0.07
0.000
0.000
0.000
Kn = 0.10
0.204
0.170
-0.057
Kn = 0.01
0.267
0.237
-0.233
5.4.3
Sensitivity of DO to Sediment Oxygen Demand
Benthic sediment is recognized as a potentially large oxygen sink in water
bodies (Bowman and Delfino, 1980). Past research has indicated that the sediment
oxygen demand can account for a major portion of the overall oxygen uptake for
some surface waters.
The SOD rate is an important parameter in water quality modelling especially
for the modelling of dissolved oxygen in surface waters. The predictions from these
models were used by some government agencies such as the United States
97
Environmental Protection Agency (USEPA) for determining the specifications for
wastewater discharge permits from municipal and industrial effluent discharges
(Brown and Barnwell, 1987).
Some of the earliest work on the magnitude of the SOD was done by Baity
(1938) and Fair et al. (1941). A variety of other in situ measurements have been
made for numerous water bodies in recent years. Table 5.14 summaried the range of
SOD values. For Malaysian river application, an example of the SOD values for
Sungai Tebrau is presented in Table 5.15.
Table 5.14: Some sediment oxygen demand values
Bottom type and location
Sphaerotilus – (10 g dry wt/m2)
Municipal sewage sludge-outfall vicinity
Municipal sewage sludge- “aged”, downstream of
outfall
Estuarine mud
Sandy bottom
Mineral soils
Source: Thomann (1972)
SOD (g O2/m2.day)@20ºC
Range
Approximate
average
7
2-10.0
4
1-2
1.5
1-2
0.2-1.0
0.05-0.1
1.5
0.5
0.07
Table 5.15 :Sediment Oxygen Demand for Sungai Tebrau (IPASA, 2002)
Location
SOD (g/m2/day)
Senai 1
0.144
Senai 2
1.987
Kg Maju Jaya
0.720
Pasar Borong Pandan
2.074
SOD is the total oxygen consumption incurred by the biochemical processes
in the sediment layer. Generally, SOD is positively correlated to nutrient and organic
loadings for a specific water body (Di Toro et al., 1990; Chapra, 1997; DNREC,
2001). The basic SOD value for the baseline run was 2.00 g/m2/day. The variations
of the parameter were 1.00 and 5.00 g/m2/day at Station Tanjung Surat as presented
98
in Figure 5.26 and Table 5.16. Figure 5.27 shows that the DO concentration is
moderately sensitive to changes in SOD along the Sungai Johor.
Table 5.16: Sensitivity of DO to SOD
Scenarios
SOD= 1.00
SOD= 2.00
SOD= 3.00
SOD= 4.00
SOD= 5.00
RMS error
0.456
0.237
0.270
0.503
0.770
DS-INTL
Average absolute error
0.401
0.191
0.204
0.458
0.732
Average error
-0.382
-0.100
0.732
0.180
0.458
DS-INTL
DO sensitivity analysis with SOD
Station: Tanjung Surat
8
Legend
SOD=1
SOD=2
SOD=5
SOD=3
SOD=4
7
Dissolved Oxygen (mg/l)
6
5
4
3
2
1
14
15
16
17
18
19
20
21
22
Time (days)
DS-INTL
DS-INTL
Figure 5.26: Sensitivity of DO to SOD at Station Tanjung Surat
baseline
SOD=1
SOD=3
10
9
8
DO (mg/l)
7
6
5
4
3
2
1
0
0
Tg P engelih
5
10
15
20
25
30
35
distance, km
Figure 5.27: Sensitivity of DO to SOD
40
45
Kota Tinggi
99
5.4.4
Sensitivity of DO to Maximum Algal Growth Rate
The basic maximum algal growth rate value for the baseline run was 2.00
day-1. The variations of the parameter were 1.00 and 3.00 g/m2/day. As shown in
Figure 5.28, DO concentration is sensitive to the changes in algal growth rate.
baseline
algal gro wth rate=1
algal gro wth rate=3
10
9
8
DO (mg/l)
7
6
5
4
3
2
1
0
0
5
Tg P engelih
10
15
20
25
30
distance, km
35
40
45
Ko ta Tinggi
Figure 5.28: Sensitivity of DO to maximum algal growth rate
5.4.5
Sensitivity of DO to Loads from Point Sources.
The sensitivity of DO concentration to the pollutant loads from the point
sources was analyzed by changing the load of NH3-N, NO3-N, PO4, COD and DOC
from the dischargers. The resulting sensitivity is shown in Figure 5.29. As shown in
Figure 5.29, DO concentration is sensitive to changing loads from point sources.
100
10
baseline
maximum lo ad
minimum lo ad
9
8
DO (mg/l)
7
6
5
4
3
2
1
0
0
5
10
15
20
25
30
35
40
distance, km
Tg P engelih
45
Ko ta Tinggi
Figure 5.29: Sensitivity of DO to loads from point sources.
5.4.6
Sensitivity of DO to Reaeration Rate.
The effect of reaeration rate on DO distribution was analyzed by changing the
value of the reaeration rate with factors of 1.00 and 3.00. Figure 5.30 plots the
relative sensitivity of DO to the reaeration rate constant by distance from
downstream boundary at Tanjung Pengelih. It is observed that the DO concentration
is moderately sensitive to the reaeration rate coefficient.
baseline
reaeratio n=1
reaeratio n=3
11
10
9
DO (mg/l)
8
7
6
5
4
3
2
1
0
0
Tg P engelih
5
10
15
20
25
30
35
40
distance, km
Figure 5.30: Sensitivity of DO to reaeration rate
45
Ko ta Tinggi
101
CHAPTER 6
CONCLUSIONS AND RECOMMENDATIONS
6.1
Conclusions
A numerical model has been applied to study the hydrodynamic and water
quality characteristics of estuary. The hydrodynamic model is based on the principles
of conservation of volume, momentum and mass, and the water quality on the
conservation of mass. The model was successfully applied to Sungai Johor estuarine
system.
Sensitivity analysis method was used in this study to identify the sensitivity
of EFDC hydrodynamic and water quality parameters of the Sungai Johor model.
This identification enabled modellers to put more emphasis on sensitive model
parameters during model calibration.
With the available data and time constraints, reasonable results were obtained
for hydrodynamic and water quality model calibration. Therefore, EFDC model
proved to be quite effective in the hydrodynamic and water quality simulation of
water surface elevation, water temperature, salinity, DO and nutrients concentration
of Sungai Johor estuarine system. Model calibration was conducted to compare the
observed and simulated results using graphical comparisons, absolute mean errors,
and RMSEs.
102
6.2
Recommendations
During the present study, the following limitations have been noted.
These need to be further investigated and be included in future model calibration
and applications to improve the model predictive capability.
i)
In the future research, the impacts of wind data and finer grids on the
velocity simulation will be explored.
ii)
As almost all other investigators have noticed, it was the turbulence
closure model that limited the predictive capability of the hydrodynamic
model and thus its capability to other systems. More understanding and
better numerical representation of the turbulent mixing processes are
essential to improve the model capability.
iii)
Coupling of the water quality model with a sediment transport model and
a sediment diagenesis model is important in predicting the nutrient
movement, particularly for phosphate and sediment nutrient exchanges.
The mechanisms that appear to be of significant include the absorption of
phosphate to sediment particles and subsequent settling, transportation of
sediment in response to high freshwater flow and the release of sediment
phosphate.
iv)
The demise of freshwater phytoplankton in the presence of salt is thought
to be a possible mechanism that limits the characteristic of high
chrolophyll a concentration to the tidal freshwater portion of the river, as
has been frequently observed in many estuarine environments including
the Sungai Johor river system.
v)
The current calibration of the water quality model has a shortcoming in
the prediction of nutrients. It was not because of the model but because of
the quality and quantity of the field data used for the current calibration.
The present model needs to be calibrated with more detailed field data to
perform the sensitivity analysis pertaining to nutrient limitation.
vi)
Future works is to link the output of wave model such as SWAN
nearshore modeling with EFDC hydrodynamic model to improve river
103
and coastal hydrodynamic with wave parameter effect at Sungai Johor
estuarine system.
vii)
This modeling study should involve watershed hydrology modeling. A
large number of modeling parameters have been defined based upon
previous similar studies, the best available data, standard modeling
assumptions, and comparison with with relevance literature. It is
anticipated and recommended that the development of this model be
continued to synthesis additional field data into the modeling process as
that data become available.
104
REFERENCES
Ali, N., Elshafie, A., Karim, O.A. and Jaafar, O. (2009). Prediction of Johor River
Water Quality Parameters Using Artificial Neural Network. Europian
Journal of Scientific Research. Vol.28 (3), 422-435.
Ambrose, R.B. (1987). Modeling of Volatile Organics in the Delaware Estuary.
ASCE, Journal of Environmental Engineering. Vol. 113(4), 703-721.
Ambrose, R.B. and Martin, J.L.(1990) Technical Guidance Manual for Performing
Waste Load Allocations, Book III: Estuaries, Part I, Estuaries and Waste
Load Allocations. U.S. Environmental Protection Agency, Washington, D.C.
Ambrose, R.B., Wool, T.A., and Martin, J.L. (1993) The Water Quality Analysis
Simulation Program, WASP5, Part A: Model Documentation. U.S. EPA
Center for Exposure Assessment Modeling, Athens, GA.
Beman, J.M., Arrigo, K.R. and Matson, P.A. (2005) Agricultural runoff fuels large
phytoplankton blooms in vulnerable areas of the ocean. Nature 434, 211-214
Blumberg, A. and Mellor, G.L. (1987). A description of a three dimensional coastal
ocean circulation model. Three Dimensional Coastal Ocean Models, N.S.
Heaps, ed., American Geophysical Union, Washington, D.C., 1-16.
Bowman, G.T., and Delfino, J.J. (1980). Sediment Oxygen Demand Techniques: A
Review and Comparison of Laboratory and In situ Systems. Water Research,
14,491-499.
Brown, L.C. and Barnwell Jr., T.O. (1987). The Enhanced Stream Water Quality
Models QUAL2E and QUAL2E-UNCAS: Documentation and User Manual.
EPA 600/3-
87/007, U.S. Environmental Protection Agency.
Chapra, S.C. (1997). Surface Water Quality Modeling. New York: McGraw-Hill
Companies, Inc.
Cerco, C.F., and Cole, T.M., (1994). Three-Dimensional Eutrophication Model of
Chesapeake Bay. Volume I: Main Report. USACE, Baltimore, MD, pp.658
105
Clean Water Noor Enterprise (2007). River Water Quality Modelling and Study of
Sewage Pollutant Impacts for Sungai Skudai and Sungai Tebrau River Basin.
Draft Final Report. Clean Water Noor Enterprise.
Chua, L.W. (2002). Application of WASP5 for Skudai River Estuarine System.
MEng Degree Thesis. Universiti Teknologi Malaysia, Skudai.
Cole, R.W. and Buchak, E.M. (1994). CE-QUAL-W2: A two dimensional, laterally
averaged, hydrodynamic and water quality model, version 2.0. Instructional
Report, EL-95-1, Army Engineer Waterways Experiment Station, Viksburg,
MS.
Craig, P.M. (2004). “User’s Manual for EFDC_Explorer: A Pre/Post Processor for
the Environmental Fluid Dynamics Code.” Dynamic Solutions, LLC,
Knoxville, Tennessee USA & Hanoi, Vietnam.
Danish Hydraulics Institute (2004a). Detailed Studies on Tuas View Extension and
Pulau Tekong Land Reclamation. Volume III: Hydrodynamic Impact
Assessment. Draft Final Report. DHI Water and Environment
Danish Hydraulics Institute (2004b). Detailed Studies on Tuas View Extension and
Pulau Tekong Land Reclamation. Volume VIII: Water Quality Impact
Assessment. Draft Final Report. DHI Water and Environment
Department of Environment (1986a). Classification of Malaysian Rivers. DOE,
Malaysia
Department of Environment (1986b). Water Quality Criteria and Standards for
Malaysia. DOE, Malaysia
Department of Environment (2007). Malaysia Environment Quality Report. DOE,
Malaysia
Dyer, K.R. (1986). Coastal and Estuarine Sediment Dynamics, John Wiley and Sons,
London.
Fatimatuzahrah Mohd Mahir (2008). Numerical Simulation of DO and COD of
Sungai Johor Estuary. Undergraduate Degree Thesis. Universiti Teknologi
Malaysia, Skudai.
Farah Salwa Ismail (2008). Water Quality Data Observation and Management for
Sungai Johor. Undergraduate Degree Thesis. Universiti Teknologi Malaysia,
Skudai.
106
Galperin B., Kantha L.H., Hassid S. and Rosati A. (1988) A quasi-equilibrium
turbulent energy model for geophysical flows. Journal of Atmospheric
Science. 45, 55-62.
Hammond, A.J. (2004). Hydrodynamic and Water Quality Simulation of Fecal
Coliform in the Lower Appomattox River, Virginia. MSc. Thesis. Virginia
Polytechnic Institute and State University
Hamrick, J.M.
(1992). A Three Dimensional Environmental Fluid Dynamics
Computer Code: Theoretical and Computational Aspects. SRAMSOE#317.
SMS/VIMS, College of William and Mary, Gloucester Point, VA. 63 pp.
Hamrick, J.M.
(1996). User’s Manual for the Environmental Fluid Dynamics
Computer Code. SRAMSOE#331. SMS/VIMS, College of William and
Mary, Gloucester Point, VA. 223 pp.
Hamrick, J.M. and Wu, T.S. (1997) Computational Design and Optimization of the
EFDC/ HEM-3D Surface Water Hydrodynamic and Eutrophication Models.
Next Generation Environmental Models and Computational Methods, G.
Delic and M.F. Wheeler, eds., Society for Industrial and Applied
Mathematics, Philadelphia.
Harleman, D.R.F. (1971). One-dimensional models in estuarine modeling: an
assessment, Tracor, Inc. Project 16070DZV, Water Quality Office, U.S.
Environmental Protection Agency, Stock No. 5501-0129, U.S. Government
Printing Office, Washington, DC.
Hashim, N.B (2001). Watershed, Hydrodynamic and Water Quality Models for
Total Maximum Daily Load St. Louis Bay Watershed Mississippi. Ph.D.
Thesis. Mississippi State University.
Hashim, N.B., Hassan, R.M., Ismail, R., Nasir, K.A., Chua, L.H. and Ismail, M.
(2004). Simplified Analytical Solution Approach for Upper Langat River.
Conference on Water Management for Sustainable Environment, Hotel,
UiTM Shah Alam (August 25-26)
Hashim, N.B., Hassan, R.M., Ibrahim, Z., Ismail, R., Malek, A.R. and Ismail, M.
(2004). Baseline Study of Water Quality in Skudai River Estuary,. Conference
on Water Management for Sustainable Environment, Hotel, UiTM Shah
Alam (August 25-
26)
Hashim, N.B., Hassan, R.M., Ibrahim, Z., Ismail, R., Malek, A.R. and Ismail, M.
(2004). Baseline Study of Water Quality in Langat River Estuary, Seminar
107
Penyelidikan Kejuruteraan Awam (SEPKA 2004), Rumah Alumni Universiti
Teknologi Malaysia, Skudai (September 1-2).
Hashim, N.B., Fadzil, A.B. and Ismail, T. (2004). “Application of Environmental
Fluid Dynamics Code (EFDC) Model In Impact Assessment”. Universiti
Teknologi Malaysia.
Hashim, N.B., Ismail, R., Nasir, K.A., Ismail, M. and Malek, A.R. (2004).“Baseline
Study of Water Quality In Langat River Estuary.” Faculty of Civil
Engineering & Faculty of Science, Universiti Teknologi Malaysia, Johor
Bahru, Malaysia.
Hashim, N.B., Hassan, R.M. and Ismail, M. (2005), “A Numerical Study of Hypoxia
in Langat River Estuarine System”, Seminar Penyelidikan Kejuruteraan
Awam SEPKA 2005, 5-6 July 2005, Pulai Springs, Skudai, Johor.
Hashim, N.B., Ismail, M. and Lee, T.Y. (2007), Hydrodynamic Modeling of Sungai
Johor Estuary Using EFDC, Prosiding Kebangsaan Awam ‘07, Langkawi,
Kedah, 29-31 May.
Hashim, N., Jalil, A., and Soon, T.T. (2007). Development of water quality
modeling of Juru River. 2nd International Conference on
Managing Rivers
st
in the 21 Century: Solution Towards Sustainable River Basins. June 6-8.
Kuching Sarawak.
Hashim, N.B., Ismail, R. and Ismail, R. (2008). Numerical Simulation of Coastal
Pollution in Sungai Johor Estuarine System. 2nd Engineering Conference
(EnCon2008) on Sustainable Engineering Infrastructure Development and
Management. December 18-19, Kuching, Sarawak.
Hayter, E.J. (2006). Evaluation of the State of the Art Contaminated Sediment
Transport and Fate Modeling System. U.S. Environmental Protection
Agency, Washington, D.C.
Heaps, N.S. (1972). On the numerical solution of the three-dimensional
hydrodynamical equations for tides and storm surges. Memories de al Societe
Royale des Sciences de Liege, Series 6, v.2.
Hsu, M.H., Kuo, A.Y., Kuo, J.T., and Liu, W.C. (1999). Procedure to calibrate and
verify numerical models of estuarine hydrodynamics. Journal of hydraulic
engineering. ASCE, 125(2), 166-182
Idi Shah Rizal Idris (2007). Preliminary Hydrodynamic Modeling of Sungai Johor.
Undergraduate Thesis. Universiti Teknologi Malaysia, Skudai.
108
Institute of Environmental and Water Resource Management (IPASA) (2002). The
study of Pollution Prevention and Water Quality Improvement of Sungai
Tebrau and Sungai Segget. Volume II. Draft Final Report. IPASA UTM.
Ji, Z.G., Morton, M.R., and Hamrick, J.K. (2001). Wetting and drying simulation of
estuarine process. Estuarine, Coastal and Shelf Science, 53, pp 683-700.
Jin, K.R., Hamrick, J.H. and Tisdale, T. (2000). Application of Three-Dimensional
Hydrodynamic Model for Lake Okeechobee. Journal of Hydraulic
Engineering. Vol. 126 (10), 758-771.
Khairul Ikhsan Shamsudin (2008). Numerical Modeling of Salinity Transport for
Sungai Lebam, Johor Darul Takzim. Undergraduate Thesis. Universiti
Teknologi Malaysia, Skudai.
Koh, H.L., Lim, P.E. and Midun, Z. (1991) Management and Control of Pollution in
Inner Johore Strait. Environmental Monitoring and Assessment 19: 349-359,
Netherlands.
Koh, H.L. and Lee, H.L. (2001). Modeling Water Quality in Sg. Lebum, Johor,
Malaysia. In: Recent Advances in Marine Science and Technology, 2000. Ed.
Narendra Saxena, p.199-208. PACON International, USA, June 2001.
Land clearing, sand mining affect rivers (2007, April 16). New Straits Times. p.
Lee, T.Y. (2006). Two Dimensional Modeling
for Johor River Estuary.
Undergraduate Thesis. Universiti Teknologi Malaysia, Skudai
Leendertse, J.J., Alexander, R.C., and Liu, S.K. (1973). A Three dimensional model
for estuaries and coastal seas. Volume I, principles of computation: Rand
Corporation, Santa Monica, California, Report R-1417-OWRR.
Liu, Z., Kieffer, J.M., Kingery,W.L., Huddleston, D.H. and Hossain, F. (2007)
Watershed modeling of dissolved oxygen and bio-chemical oxygen demand
using Hydrological Simulation Program Fortran. Journal of Environmental
Science and Health. 42:2023–2032.
Liu, Z., Hashim, N.B., Kingery, W.L., Huddleston, D.H. and Xia, M. (2008a).
Hydrodynamic modeling of St. Louis Bay estuary and watershed using EFDC
and HSPF. Journal of Coastal Research. Volume S52. 107-116.
Liu, Z., Kingery, W.L., Huddleston, D.H., Hossain, F., Hashim, N.B. and Kieffer,
J.M. (2008b). Assessment of Water Quality Conditions in the St. Louis Bay
109
Watershed. Journal of Environmental Science and Health. Vol. 43 (5), 468477
Liu, Z., Kingery, W.L., Huddleston, D.H., Hossain, F., Chen, W., Hashim, N.B. and
Kieffer, J.M. (2008c). Modeling Nutrient Dynamic Under Critical Flow
Conditions in Three Tributaries of St. Louis Bay. Journal of Environmental
Science and Health. Vol. 43 (6), 633-645
Liu, Z., Kingery, W.L., Huddleston, D.H., Hossain, F., Hashim, N.B. and Kieffer,
J.M. (2008d). Application and Evaluation of Two Nutrient Algorithms of
Hydrological Simulation Program Fortran in Wolf River Watershed. Journal
of Environmental Science and Health. Vol. 43 (7), 738-748
Lung, W.S. and Hwang, C.C. (1989). Integrating hydrodynamic and water quality
models for the Patuxent Estuary, p.420-429, in M.Spaulding (ed), Estuarine
and Coastal Modeling, American Society of Civil Engineering, New York.
Lung, W.S. (1993) Water Quality Modeling, Volume III, Application to Estuaries,
CRC Press, Boca Raton, FL., pp. 1-4.
Lung, W.S., and C.E. Larson. (1995) Water quality modeling of Upper Mississippi
River and Lake Pepin. ASCE J. Environ. Eng. 121(10): 691-699.
Lung, D.S. (2001). Kajian Kualiti Air Sungai Kempas.
Undergraduate Thesis.
Universiti Teknologi Malaysia, Skudai
Martin, J.L. (1988) Application of two dimensional water quality model. Journal of
Environmental Engineering 114(2): 317-336.
Martin, J.L. and Mccutcheon, S.C. (1999). Hydrodynamics and Transport for Water
Quality Modeling. (1st ed.) CRC Press, Inc.
Mellor, G.L. and Yamada, T. (1982). Development of a turbulence closure model for
geophysical fluid problems. Rev. Geophys. Space Phys., 20, 851-875.
Mendelsohn, D.L., Peene, S., Yassuda, E. and Davie, S. (2000). The Gaspe Current
and Cyclonic Motion Over the Northwestern Gulf of St. Lawrence. 6th
International Conference on Estuarine and Coastal Modeling, ASCE, New
Orleans, Louisiana, November 3-5, 1999
Mobley, C.D., and Stewart, R.J. (1980). On the numerical generation of boundaryfitted orthogonal curvilinear coordinate systems. Journal of Computer
Physical, 34, 124-125.
110
Moustafa, M.Z., and Hamrick, J.M. (2000). Calibration of the wetland hydrodynamic
model to the Everglades nutrient removal project. Water Quality and
Ecosystem Modeling, 1, 141-167.
Mohamad Faazil Fakharazi (2008) The Impact of Land Reclamation Activity or
Flood Problem at Sungai Batu Pahat. Undergraduate Thesis.
Universiti
Teknologi Malaysia, Skudai.
Mohd Shafiq Othman (2008). Permodelan Kualiti Air bagi Sungai Tiram
Menggunakan Perisian EFDC. Undergraduate Thesis. Universiti Teknologi
Malaysia, Skudai
Mohd Zaharifudin Muhamad Ali (2008). The Evaluation of Hydrodynamic and
Sediment Transport Model Using Environmental Fluid Dynamic Code. MEng
Thesis. Universiti Teknologi Malaysia, Skudai.
Muhamad bin Abdullah (2008). The Impact of Sand Dredging Activities to the
Sediment Transport on Sungai Batu Pahat. Undergraduate Thesis. Universiti
Teknologi Malaysia, Skudai.
Muhammad Hanif Ab. Llah Zawawi (2008). Kajian Kualiti Air bagi Sungai Lebam.
Undergraduate Thesis. Universiti Teknologi Malaysia, Skudai
Nurhidayah Hamzah (2007). Assessment on Water Quality and Biodiversity within
Sungai Batu Pahat. MEng Thesis. Universiti Teknologi Malaysia, Skudai
O’Connor, D.J. (1960). Oxygen balance of an estuary, ASCE Journal Sanitary Engr.
Div., 86 (SA3)
Officer, C.B. (1976). Physical oceanography of estuaries and associated coastal
waters, John Wiley and Sons, New York.
Park K. (1993). A Model Study of Hydrodynamic and Water Quality Characteristics
of the Rappanannock Estuary, Virginia. Ph.D. Thesis. School of Marine
Science/ Virginia Institute of Marine Science, College of William and Mary,
VA.
Park, K., Kuo, A.Y., Shen, J. and Hamrick, J.M. (1995). A three-dimensional
hydrodynamic eutrophication model (HEM-3D): Description of water quality
and sediment process submodels, Special report No. 327, The College of
William and Mary, Virginia Institute of Marine Science.
Park, K, Kuo, A.Y. and Neilson, B.J. (1996). A numerical model study of hypoxia in
the tidal Rappahannock River of Chesapeake Bay. Estuarine, Coastal and
Shelf Science 42 ,563-581
111
Pritchard, D. W. (1967) What is an estuary: physical viewpoint. p. 3–5 in: G. H. Lauf
(ed.) Estuaries, A.A.A.S. Publ. No. 83, Washington, D.C.
Rosati, A.K. and Miyakoda, K. (1988) A general circulation modle for upper ocean
simulation. Journal Physical Oceanography, 18, 1601-1626.
Roslan Mohamed Yusof (2008). Kajian Kesan Enapan Sedimen kepada Kualiti Air
di Sungai Tiram.. Undergraduate Thesis. Universiti Teknologi Malaysia,
Skudai
Reynolds Fleming, J.V. (2003). The hydrodynamics of the upper neuse river estuary,
NC and their influence on dissolved oxygen distribution. Ph.D. Thesis.
University of North Carolina.
Ryskin, G. and Leal, L.G. (1983). Orthogonal mapping. Journal of Computational
Physical. Volume 50, 71-100.
Said, S., Mah, D. Y. S., Sumok, P. and Lai, S. H. (2009) Water quality monitoring of
Maong River, Malaysia. Proceedings of the Institution of Civil Engineers,
Water Management. Thomas Telford, London, 162, No. 1, 37–42
Sheng, J. (2000). The Gaspe Current and Cyclonic Motion Over the Northwestern
Gulf of St. Lawrence. 6th International Conference on Estuarine and Coastal
Modeling, ASCE, New Orleans, Louisiana, November 3-5, 1999
Shoemaker, L., Lahlou, M., Bryer, M., Kumar, D. and Kratt, K. (1997). Conpendium
of tools for watershed assessment and TMDL development, United States
Environmental Protection Agency, Office of Water, EPA841-B-97-006,
Washington, DC.
Sisson, G.M., Shen, J., Kim, S.C., Boon, J.D. and Kuo, A.Y. (1997). VIMS Three-
Dimensional Hydrodynamic-Eutrophication Model (HEM-3D): Application
of the Hydrodynamic Model to the York River System. Special Report in
Applied Marine Science and Ocean Engineering No. 341. College of William
and Mary, Virginia Institute of Marine Science.
Smith, P.E. (1997). A Three-Dimensional, finite difference model for estuarine
circulation. Ph.D Dissertation, Universiti of California Davis, Davis, CA.
Smolarkiewicz, P.K. and Clark, T.L. (1986). The Multidimensional positive definite
advection transport algorithm: Further development and applications. Journal
Comp. Phys., 67, 396-438.
Smolarkiewicz, P.K.and Margolin, L.G. (1993) On forward in time differencing for
fluids: extension to a curvilinear framework.
112
Smorlarkiewicz, P. K. and Grabowski, W. W. (1990). The multidimensional positive
definite advection transport algorithm: Nonoscillatory Option . Journal
Comp. Phys., 86, 355-375.
Spaulding M. (2000). Proceeding of Conference on Estuarine and Coastal Modeling,
M.L. Spaulding, ASCE.
Sumok, P., Said, S. and Mah, D.Y.S (2007). Water quality modeling of the urbanized
Sungai Maong, Kuching City. 2nd International Conference on Managing
Rivers in the 21st Century: Solution Towards Sustainable River Basins. June
6-8. Kuching Sarawak.
Sundermann, J. (1975). A three-dimensional model of a homogeneous estuary:
ASCE, Proceedings of the Fourteenth Coastal Engineering Conference, V3,
June 24-28, Copenhagen, Denmark.
Syamsidik and Koh, H.L (2003). Impact Assesment Modeling On Coastal
Reclamation At Pulau Tekong . Regional Conference on Integrating
Technology in the Mathematical Sciences, Conference Hall Universiti
Sains Malaysia, (14-15 April).
Syed Mohd Naim Syed Rozali (2008). Simulation of Dissolved Oxygen in Straits of
Johor. Undergraduate Thesis. Universiti Teknologi Malaysia, Skudai.
Teoh, Y.L. (2006). “Water Quality Analysis Of Johor River Estuary”. Undergraduate
Thesis. Universiti Teknologi Malaysia, Skudai
Tetra Tech, Inc. (1994). Indian River Lagoon hydrodynamic and salinity model:
calibration and verification, a report to Florida Institute of Technology and St.
Johns River Management District, Tetra Tech, Inc., Fairfax, VA.
Tetra Tech, Inc. (1995) Hydrodynamic and water quality mathematical modeling
study of Norwalk Harbor, Connecticut. Technical report for City of Norwalk,
Department of Public Works.
Thomann, R.V. (1962). The use of systems analysis to describe the time variation of
dissolved oxygen in a tidal stream. Ph.D. Dissertation. Department of
Meteorology and Oceanography, New York University, New York.
Thomann, R.V. and Mueller, J. A. (1987). Principles of Surface Water Quality
Modeling and Control, Harper and Row Publishing Co., New York, NY.
University of Malaya Consultancy Unit (2002). Program Pencegahan Pencemaran
dan Peningkatan Kualiti Air Sungai Langat. Draf Laporan Akhir. UPUM
113
USEPA (1997). Conpendium of Tools for Watershed Assessment and TMDL
Development. USEPA, Office of Water, Washington, DC.118 pp.
Vollenweider, R.A., Marchetti, R. and Viviani, R. (1992). Coastal Marine
Eutrophication. London:Elsevier. pp 1-20
Wang, P.F., Martin, J.L. and Morrison, G. (1999). Water Quality and Eutrophication
in Tampa Bay, Florida. Estuarine, Coastal and Shelf Science. Vol. 49, 1-20.
Wang, J.D., Blumberg, A.F., Butler, H.L., and Hamilton, P. (1990). Transport
prediction in partially stratified tidal water. Journal Hydraulic Engineering,
ASCE, 116(3), 380-396.
Warwick, J.J. and Heim, K.J. (1995). Hydrodynamic modeling of the Carson River
and Lahontan Reservoir, Nevada, Water Resources Bulletin, vol. 31(1), p. 6777.
Wool, T.A., Ambrose, R.B., and Martin, J.L., and Comer, E.A (2003). Water Quality
Analysis Simulation Program, WASP version 6.0. Draft: User Manual,
USEPA, Region IV, Atlanta, GA. 267 pp.
Zhang, Q.Y. and Gin, K.Y.H (1998). Three Dimensional Numerical Simulation for
Tidal Motion in Singapore’s Coastal Waters, Coastal Engineering, Volume
39. Issue 1. pp 71-92
114
APPENDIX A.1:
DO concentration profiles along Sungai Johor
2005
min
max
average
10
1440916
9
DO (mg/l)
8
1440963
3JH15
7
3JH20
6
5
1441967
3JH13
3JH19
1340973
4
3
2
1
0
Tanjung
Pengelih
10
20
30
40
50
60
70
80
90
Rantau
Panjang
distance from Tanjung Pengelih (km )
2004
min
max
average
10
9
1440963
1340973
DO (mg/l)
8
3JH20
3JH15
7
6
5 1441967
3JH19
1440916
3JH13
4
3
2
1
0
Tanjung
Pengelih
10
20
30
40
50
60
70
80
90
Rantau
Panjang
distance from Tanjung Pengelih (km )
2003
min
max
average
10
9
3JH15
DO (mg/l)
8
7
6
1440963 1440916
3JH20
1441967
1340973
5
3JH13
3JH19
4
3
2
1
0
Tanjung
Pengelih
10
20
30
40
50
60
distance from Tanjung Pengelih (km )
70
80
90
Rantau
Panjang
115
APPENDIX A.2:
Salinity profiles along Sungai Johor
2005
min
max
average
40
35
1440963
1340973
Salinity (pptl)
30
25
1441967
20
1440916
15
10
5
3JH20 3JH19 3JH15 3JH13
0
0
Tanjung
Pengelih
10
20
30
40
50
60
70
80
90
Rantau
Panjang
distance from Tanjung Pengelih (km )
2004
min
max
average
40
35
Salinity (pptl)
30
25
1440963
1340973
1441967
20
15
10
5
1440916
3JH20 3JH19 3JH15 3JH13
0
0
Tanjung
Pengelih
10
20
30
40
50
60
70
80
90
Rantau
Panjang
distance from Tanjung Pengelih (km )
2003
min
max
average
40
35
Salinity (pptl)
30
25
1440963
1340973
20
15
1441967
1440916
10
5
3JH20 3JH19 3JH15 3JH13
0
0
Tanjung
Pengelih
10
20
30
40
50
60
distance from Tanjung Pengelih (km )
70
80
90
Rantau
Panjang
116
APPENDIX A.3:
Temperature profiles along Sungai Johor
2005
32
Temperature (°C)
30
min
max
average
1440916
1441967
3JH20
1340973
3JH15
28
1440963
26
3JH13
3JH19
24
22
20
0
Tanjung
Pengelih
10
20
30
40
50
60
70
80
90
Rantau
Panjang
distance from Tanjung Pengelih (km )
2004
min
max
average
1440916
32
1441967
Temperature (°C)
30
3JH20
3JH15
28
1440963
1340973
26
3JH13
3JH19
24
22
20
0
Tanjung
Pengelih
10
20
30
40
50
60
80
90
Rantau
Panjang
distance from Tanjung Pengelih (km )
2003
min
max
average
1440916
32
1441967
3JH20
30
Temperature (°C)
70
28
1440963
3JH15
1340973
26
3JH13
3JH19
24
22
20
0
Tanjung
Pengelih
10
20
30
40
50
60
distance from Tanjung Pengelih (km )
70
80
90
Rantau
Panjang
117
APPENDIX B
Methods of Analysis
The APHA Standard Methods for the examination of water and wastewater
(1992) have been adopted. Five of the water quality parameters namely DO, salinity,
conductivity, turbidity and pH were measured in situ. All other parameters were
analyzed in the laboratory at the Department of Chemistry, Faculty of Science, UTM
Skudai. The water quality parameters are summarized in Table B.1
Table B.1: Range of Parameters and Methods of Analysis
Parameters
Methods of analysis
pH
YSI 600R / YSI 6600
Conductivity
YSI 600R / YSI 6600
Salinity
YSI 600R / YSI 6600
Turbidity
YSI 6600
DO
YSI 600R / YSI 6600
BOD
Incubation at 20°C for 5 days
TSS
APHA Method 2540-D
COD
HACH DR4000 Method 8000
Ammonia nitrogen
HACH DR4000 Method 8038
Total Phosphorus
HACH DR4000 Method 8190
Nitrate
HACH DR4000 Method 8171
118
APPENDIX C
Temporal Profiles of Stream Flow
Stream Flow at Station Rantau Panjang (1997-2006)
350
300
Flow (m3/s)
250
200
150
100
50
0
Jan-97 Jan-98 Jan-99 Jan-00 Jan-01 Jan-02 Jan-03 Jan-04 Jan-05 Jan-06 Jan-07
Tim e (m onth,year)
Figure C.1: Stream flow at Station Rantau Panjang (1997-2006)
Flow at Station Rantau Panjang (1997-2006)
Mean
Sta.dev
100
Flow (m 3/s)
80
60
40
20
0
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
Tim e (m onth)
Figure C.2: Stream flow analysis at Station Rantau Panjang
DEC
119
APPENDIX D
Temporal Profiles of DO Concentrations and Nutrient Loads
Station 3JH13: Sg Johor at Rantau Panjang
Class I
Dissolved oxygen (mg/l)
8
7
Class II
6
5
4
Class III
3
Min:2.9
Max: 7.4
Mean: 6.2
2
1
Class IV
Class V
0
Jan-97 Jan-98 Jan-99 Jan-00 Jan-01 Jan-02 Jan-03 Jan-04 Jan-05 Jan-06 Jan-07 Jan-08
Tim e (year)
Station 3JH13: Sg Johor at Rantau Panjang
14000
N loads (kg/day)
12000
Min:19
Max: 13 133
Mean: 1 610
10000
8000
6000
4000
2000
0
Jan97
Jan98
Jan99
Jan00
Jan01
Jan02
Jan03
Jan04
Jan05
Jan06
Jan07
Jan08
Tim e (year)
Min:1
Max: 1 901
Mean: 134
Station 3JH13: Sg Johor at Rantau Panjang
2000
P loads (kg/day)
1800
1600
1400
1200
1000
800
600
400
200
0
Jan97
Jan98
Jan99
Jan00
Jan01
Jan02
Jan03
Jan04
Jan05
Jan06
Jan07
Jan08
Tim e (year)
Figure D.1: DO concentrations and nutrient loads at Station 3JH13 (1997-2007)
120
Station 3JH03: Sungai Layang
Class I
Dissolved oxygen (mg/l)
8
7
Class II
6
5
Class III
4
3
Min:4.1
Max:8.8
Mean:6.7
2
1
Class IV
Class V
0
Jan-97 Jan-98 Jan-99 Jan-00 Jan-01 Jan-02 Jan-03 Jan-04 Jan-05 Jan-06 Jan-07 Jan-08
Tim e (year)
Min: 0.3
Max: 729.9
Mean: 75.4
Station 3JH03: Sungai Layang
800
N loads (kg/day)
700
600
500
400
300
200
100
0
Jan97
Jan98
Jan99
Jan00
Jan01
Jan02
Jan03
Jan04
Jan05
Jan06
Jan07
Jan08
Tim e (year)
Min: 0.1
Max: 190.1
Mean: 12.8
Station 3JH03: Sungai Layang
200
P loads (kg/day)
180
160
140
120
100
80
60
40
20
0
Jan97
Jan98
Jan99
Jan00
Jan01
Jan02
Jan03
Jan04
Jan05
Jan06
Jan07
Jan08
Tim e (year)
Figure D.2: DO concentrations and nutrient loads at Station 3JH03 (1997-2007)
121
Min: 0.3
Max: 8.1
Mean: 3.6
Station 3JH05: Sungai Serai
Dissolved oxygen (mg/l)
8
7
Class I
Class II
6
5
4
Class III
3
2
Class IV
1
Class V
0
Jan-97 Jan-98 Jan-99 Jan-00 Jan-01 Jan-02 Jan-03 Jan-04 Jan-05 Jan-06 Jan-07 Jan-08
Tim e (year)
4000
N loads (kg/day)
3500
Min: 2
Max: 3 478
Mean: 749
Station 3JH05: Sungai Serai
3000
2500
2000
1500
1000
500
0
Jan97
Jan98
Jan99
Jan00
Jan01
Jan02
Jan03
Jan04
Jan05
Jan06
Jan07
Jan08
Jan05
Jan06
Jan07
Jan08
Tim e (year)
500
P loads (kg/day)
450
400
Min: 0
Max: 461
Mean: 76
Station 3JH05: Sungai Serai
350
300
250
200
150
100
50
0
Jan97
Jan98
Jan99
Jan00
Jan01
Jan02
Jan03
Jan04
Tim e (year)
Figure D.3: DO concentrations and nutrient loads at Station 3JH05 (1997-2007)
122
Min: 1.3
Max: 6.0
Mean: 3.7
Dissolved oxygen (mg/l)
8
7
Station 3JH06: Sungai Tiram
Class I
Class II
6
5
Class III
4
3
2
Class IV
1
Class V
0
Jan-97 Jan-98 Jan-99 Jan-00 Jan-01 Jan-02 Jan-03 Jan-04 Jan-05 Jan-06 Jan-07 Jan-08
Tim e (year)
Min: 55
Max: 4 793
Mean: 887
Station 3JH06: Sungai Tiram
6000
N loads (kg/day)
5000
4000
3000
2000
1000
0
Jan97
Jan98
Jan99
Jan00
Jan01
Jan02
Jan03
Jan04
Jan05
Jan06
Jan07
Jan08
Tim e (year)
Min: 0
Max: 190
Mean: 24
Station 3JH06: Sungai Tiram
200
P loads (kg/day)
180
160
140
120
100
80
60
40
20
0
Jan97
Jan98
Jan99
Jan00
Jan01
Jan02
Jan03
Jan04
Jan05
Jan06
Jan07
Jan08
Tim e (year)
Figure D.4: DO concentrations and nutrient loads at Station 3JH06 (1997-2007)
123
Min:0.01
Max: 7.8
Mean: 4.8
Dissolved oxygen (mg/l)
8
7
Station 3JH18: Sungai Berangan
Class I
Class II
6
5
4
Class III
3
2
Class IV
1
Class V
0
Jan-97 Jan-98 Jan-99 Jan-00 Jan-01 Jan-02 Jan-03 Jan-04 Jan-05 Jan-06 Jan-07 Jan-08
Tim e (year)
Min:1
Max: 5 713
Mean: 459
Station 3JH18: Sungai Berangan
6000
N loads (kg/day)
5000
4000
3000
2000
1000
0
Jan97
Jan98
Jan99
Jan00
Jan01
Jan02
Jan03
Jan04
Jan05
Jan06
Jan07
Jan08
Tim e (year)
Min:0
Max: 479
Mean: 39
Station 3JH18: Sungai Berangan
600
P loads (kg/day)
500
400
300
200
100
0
Jan97
Jan98
Jan99
Jan00
Jan01
Jan02
Jan03
Jan04
Jan05
Jan06
Jan07
Jan08
Tim e (year)
Figure D.5: DO concentrations and nutrient loads at Station 3JH18 (1997-2007)
124
8
Dissolved oxygen (mg/l)
Station 3JH22: Sungai Tem on
Min:1.3
Max: 9.0
Mean: 5.9
9
Class I
7
Class II
6
5
Class III
4
3
Class IV
2
1
Class V
0
Jan-97 Jan-98 Jan-99 Jan-00 Jan-01 Jan-02 Jan-03 Jan-04 Jan-05 Jan-06 Jan-07 Jan-08
Tim e (year)
Min:1
Max: 2 860
Mean: 245
3500
N loads (kg/day)
3000
Station 3JH22: Sungai Tem on
2500
2000
1500
1000
500
0
Jan97
Jan98
Jan99
Jan00
Jan01
Jan02
Jan03
Jan04
Jan05
Jan06
Jan07
Jan08
Jan05
Jan06
Jan07
Jan08
Tim e (year)
250
Min:0
Max: 229
Mean: 18
Station 3JH22: Sungai Tem on
P loads (kg/day)
200
150
100
50
0
Jan97
Jan98
Jan99
Jan00
Jan01
Jan02
Jan03
Jan04
Tim e (year)
Figure D.6: DO concentrations and nutrient loads at Station 3JH22 (1997-2007)
125
Station 3JH25: Sungai Layau
Class I
Dissolved oxygen (mg/l)
8
7
Class II
6
5
4
Class III
3
Min:3.0
Max: 7.8
Mean: 6.7
2
1
Class IV
Class V
0
Jan-97 Jan-98 Jan-99 Jan-00 Jan-01 Jan-02 Jan-03 Jan-04 Jan-05 Jan-06 Jan-07 Jan-08
Tim e (year)
Min:3
Max: 4 645
Mean: 345
Station 3JH25: Sungai Layau
5000
N loads (kg/day)
4500
4000
3500
3000
2500
2000
1500
1000
500
0
Jan97
Jan98
Jan99
Jan00
Jan01
Jan02
Jan03
Jan04
Jan05
Jan06
Jan07
Jan08
Tim e (year)
Min:0
Max: 409
Mean: 16
Station 3JH25: Sungai Layau
450
400
P loads (kg/day)
350
300
250
200
150
100
50
0
Jan97
Jan98
Jan99
Jan00
Jan01
Jan02
Jan03
Jan04
Jan05
Jan06
Jan07
Jan08
Tim e (year)
Figure D.7: DO concentrations and nutrient loads at Station 3JH25 (1997-2007)
126
Dissolved oxygen (mg/l)
Station 3JH30: Sungai Lebam
Min: 1.9
Max: 6.9
Mean: 3.7
8
7
Class I
Class II
6
5
4
Class III
3
2
Class IV
1
Class V
0
Jan-97 Jan-98 Jan-99 Jan-00 Jan-01 Jan-02 Jan-03 Jan-04 Jan-05 Jan-06 Jan-07 Jan-08
Tim e (year)
Min:1
Max: 6 965
Mean: 265
Station 3JH30: Sungai Lebam
8000
N loads (kg/day)
7000
6000
5000
4000
3000
2000
1000
0
Jan97
Jan98
Jan99
Jan00
Jan01
Jan02
Jan03
Jan04
Jan05
Jan06
Jan07
Jan08
Tim e (year)
Min:0
Max: 190
Mean: 12
Station 3JH30: Sungai Lebam
200
P loads (kg/day)
180
160
140
120
100
80
60
40
20
0
Jan97
Jan98
Jan99
Jan00
Jan01
Jan02
Jan03
Jan04
Jan05
Jan06
Jan07
Jan08
Tim e (year)
Figure D.8: DO concentrations and nutrient loads at Station 3JH30 (1997-2007)
127
Dissolved oxygen (mg/l)
8
7
Min:2.1
Max: 6.2
Mean: 4.7
Station 3JH33: Sungai Santi
Class I
Class II
6
5
4
Class III
3
2
Class IV
1
Class V
0
Jan-97 Jan-98 Jan-99 Jan-00 Jan-01 Jan-02 Jan-03 Jan-04 Jan-05 Jan-06 Jan-07 Jan-08
Tim e (year)
Min:1
Max: 1 878
Mean: 159
Station 3JH33: Sungai Santi
2000
N loads (kg/day)
1800
1600
1400
1200
1000
800
600
400
200
0
Jan97
Jan98
Jan99
Jan00
Jan01
Jan02
Jan03
Jan04
Jan05
Jan06
Jan07
Jan08
Tim e (year)
Min:0
Max: 1 355
Mean: 34
Station 3JH33: Sungai Santi
1600
P loads (kg/day)
1400
1200
1000
800
600
400
200
0
Jan97
Jan98
Jan99
Jan00
Jan01
Jan02
Jan03
Jan04
Jan05
Jan06
Jan07
Jan08
Tim e (year)
Figure D.9: DO concentrations and nutrient loads at Station 3JH33 (1997-2007)