2. Research method

The study has focused on three shallow water bodies namely Bukit Merah reservoir, Durian Tunggal reservoir and Bera Lake (Figure 1). Bera Lake is a natural floodplain lake with a mean depth of about 2.8 m, while Durian Tunggal is a water supply reservoir with a mean depth about 6 m. Bera Lake (N3�7<sup>0</sup> 0000, E102�36<sup>0</sup> 0000) is a dendritic, alluvial peat and freshwater swamp system situated in Bera District, Pahang [9], while Durian Tunggal Lake is located in the Malacca State, Malaysia (N2�20<sup>0</sup> 0000, E102�18<sup>0</sup> 0000). Bukit Merah reservoir is a multifunction shallow reservoir (mean depth of about 2.5 m) created for irrigation and flood mitigation. The reservoir is located in Northern Perak State at N5�2<sup>0</sup> 0000, E100�40<sup>0</sup> 0000. Durian Tunggal reservoir is small (surface area about 5.8 km<sup>2</sup> ), while Bukit Merah reservoir is large (surface area of about 33 km<sup>2</sup> ). Both Bera Lake and Durian Tunggal Lake are mesotrophic, while Bukit Merah Lake is mesotrophic-eutrophic [10].

was limited to open lake areas due to extensive presence of macrophytes. The bathymetry and shoreline data were meshed into three-dimensional grids (Table 1). Smaller vertical grid (0.5 m) was selected for the Bukit Merah reservoir due to its shallowness (~4 m) while vertical grid of 1 m was chosen for Bera and Durian Tunggal lakes. Bukit Merah Lake has nonuniform mesh type with larger grid for open water and smaller grid in channel connecting the north and south of the lake. For Durian Tunggal Lake, focus was given to the larger part of the lake in the east with areas separated by road excluded in the analysis due to the lack of observation data. The bathymetry of the three lakes is shown in Figure 2. Flow data at nearest station within each lake catchment were obtained from the Drainage and Irrigation Department of

Wind speedð Þ <sup>m</sup>=<sup>s</sup> <sup>p</sup>

Lake Bera Durian Tunggal Bukit Merah

Vertical grid (m) 1.0 1.0 0.5

Maximum time step (s) 3 2 10

Surface drag coefficient <sup>0</sup>:<sup>0007</sup> <sup>þ</sup> <sup>0</sup>:<sup>0004</sup> ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi

Table 1. Hydrodynamic model set up and computational setting.

Vertical eddy viscosity and diffusivity Predicted by the turbulence closure model

Horizontal grid designs 100 m uniform mesh model 100 m uniform mesh model 200 m � 200 m; 5 m

Simulation period March-June 2015 January-March 2014 February-August

Computational area (m) 3180 � 3500 4300 � 5200 6500 � 9500

mesh for channel

2014

Assessing the Hydrodynamic Pattern in Different Lakes of Malaysia

http://dx.doi.org/10.5772/intechopen.73274

73

Figure 1. Map of study lakes in Peninsular Malaysia.

Meteorological and hydrological data were obtained from automatic weather monitoring stations deployed at each lake or the nearest weather or rainfall stations over the study period. Meteorological data, such as wind, water discharge, and morphological properties, were important elements for three-dimensional hydrodynamic model application and calibration of the model. Bathymetry data for Bukit Merah Reservoir were attained from hydrographic survey results provided by Kerian Drainage and Irrigation Department. Hydrographic surveys were carried out for Bera Lake and Durian Tunggal reservoir. Survey area for Bera Lake Assessing the Hydrodynamic Pattern in Different Lakes of Malaysia http://dx.doi.org/10.5772/intechopen.73274 73

Figure 1. Map of study lakes in Peninsular Malaysia.

observed basin-scale cyclonic gyre in Lake Maracaibo [2]. Bailey and Hamilton showed that the effects of changing wind direction were found to have a greater influence on the sediment concentration distribution compared to advection and diffusion in lakes having directionally varying strong winds such as Thomsons Lake in Australia [3]. Numerous earlier works have established the spatial variability of winds and temperature on lake dynamics including

Many recent studies have also used hydrodynamic simulation to assess the impacts of climate change on lakes [4, 5]. One-dimensional hydrodynamic model was used to identify potential drawbacks of climate change in Lake Ammersee, Germany. The study found that global warming of the atmosphere increased water temperatures, subsequently extending duration of stratification and increasing thermal stability on the lake ecosystem [6]. In [7], simulation results indicated global warming led to possible increase in water transparency in Lake Mangueira, Brazil due to nutrient limitation and submerged macrophyte growths, while increase in inflows and low water levels changed the lake hydrodynamic behavior leading to algal bloom [8].

Despite known significant advances in lake hydrodynamic information published in literature, understanding on the circulation patterns in different lakes, in particular in the trophic zone, remains inadequate due to the absence of long-term monitoring data and technical knowledge on numerical model simulation [4]. The aim of this study is to improve understanding of the hydrodynamic characteristics in different tropical lakes located in Malaysia. Understanding

The study has focused on three shallow water bodies namely Bukit Merah reservoir, Durian Tunggal reservoir and Bera Lake (Figure 1). Bera Lake is a natural floodplain lake with a mean depth of about 2.8 m, while Durian Tunggal is a water supply reservoir with a mean depth

swamp system situated in Bera District, Pahang [9], while Durian Tunggal Lake is located in

function shallow reservoir (mean depth of about 2.5 m) created for irrigation and flood

Meteorological and hydrological data were obtained from automatic weather monitoring stations deployed at each lake or the nearest weather or rainfall stations over the study period. Meteorological data, such as wind, water discharge, and morphological properties, were important elements for three-dimensional hydrodynamic model application and calibration of the model. Bathymetry data for Bukit Merah Reservoir were attained from hydrographic survey results provided by Kerian Drainage and Irrigation Department. Hydrographic surveys were carried out for Bera Lake and Durian Tunggal reservoir. Survey area for Bera Lake

0000, E102�18<sup>0</sup>

0000) is a dendritic, alluvial peat and freshwater

). Both Bera Lake and Durian Tunggal Lake are mesotrophic,

0000). Bukit Merah reservoir is a multi-

0000, E100�40<sup>0</sup>

), while Bukit Merah reservoir is large

0000. Durian

such hydrodynamic pattern is necessary to enable effective management purposes.

0000, E102�36<sup>0</sup>

mitigation. The reservoir is located in Northern Perak State at N5�2<sup>0</sup>

Tunggal reservoir is small (surface area about 5.8 km<sup>2</sup>

while Bukit Merah Lake is mesotrophic-eutrophic [10].

horizontal and vertical mixing and stratification pattern [4].

72 Applications in Water Systems Management and Modeling

2. Research method

about 6 m. Bera Lake (N3�7<sup>0</sup>

(surface area of about 33 km<sup>2</sup>

the Malacca State, Malaysia (N2�20<sup>0</sup>


Table 1. Hydrodynamic model set up and computational setting.

was limited to open lake areas due to extensive presence of macrophytes. The bathymetry and shoreline data were meshed into three-dimensional grids (Table 1). Smaller vertical grid (0.5 m) was selected for the Bukit Merah reservoir due to its shallowness (~4 m) while vertical grid of 1 m was chosen for Bera and Durian Tunggal lakes. Bukit Merah Lake has nonuniform mesh type with larger grid for open water and smaller grid in channel connecting the north and south of the lake. For Durian Tunggal Lake, focus was given to the larger part of the lake in the east with areas separated by road excluded in the analysis due to the lack of observation data. The bathymetry of the three lakes is shown in Figure 2. Flow data at nearest station within each lake catchment were obtained from the Drainage and Irrigation Department of

∂ς

∂ ∂t

�1

horizontal viscosity and diffusivity coefficients (cm�<sup>2</sup>

The main characteristics of runs are explained in Table 1.

�1

vertical velocity (cm�s

Coriolis parameter (s

�2

and diffusivity coefficients (cm�<sup>2</sup>

(= 980 cm�s

balance.

<sup>∂</sup><sup>t</sup> <sup>¼</sup> <sup>w</sup> � <sup>∇</sup> � <sup>M</sup>1, wk�<sup>1</sup> <sup>¼</sup> wk � <sup>∇</sup> � Mk <sup>2</sup>≦k≦K�<sup>1</sup> , (2)

� Kz

) which is described as <sup>f</sup> <sup>0</sup> <sup>¼</sup> <sup>2</sup><sup>ω</sup> sinφ<sup>0</sup> with the angular velocity <sup>ω</sup> (s�<sup>1</sup>

�s �1

�s �1

hk ð Þþ � <sup>θ</sup><sup>k</sup> <sup>∇</sup> � <sup>M</sup><sup>k</sup> ð Þþ � <sup>θ</sup><sup>k</sup> ð Þj <sup>w</sup><sup>θ</sup> �Hk�<sup>1</sup> � ð Þj <sup>w</sup><sup>θ</sup> �Hk

In these equations, hk (k = 1, 2, …, K) represents thickness (cm) of each k layer, H the still-water

of the earth rotation and the mean latitude φ<sup>0</sup> of the lake, g the gravitational acceleration

integrated into vertically in each layer, ∇ the horizontal gradient operator, AH and KH the

The model employs the turbulence closure scheme, the k–L model, that considers turbulence kinetic energy (k) and mixing length (L) for hydrodynamic simulation due to the importance of vertical turbulence transport processes. The buoyancy effect due to stratification was based on Munk-Anderson empirical relationship [13]. Knudsen's expression was adopted for calculating the water density from water temperature and salinity. The heat exchange calculation through the water surface is estimated based on the contribution from the sensible heat flux, and latent heat flux, short wave radiation and long wave radiation. Meteorological data, such as air temperature, wind speed, and solar radiation, provide the input for the surface heat

The main inputs of the model were the river inflow; meteorological parameter, such as wind, rainfall, and air temperature; and bathymetry data. Comparison between different wind speeds and under calm wind conditions was made to evaluate the circulation pattern and transport phenomena. The model assumes no physical effect of the flow field and no contribution from the groundwater. Despite the fact that a large quantity of macrophytes, such as submerged species Cabomba furcata in Bukit Merah Lake and emergent plants Pandanus helicopus in Bera Lake, may shape the hydrodynamic features, this effect was not considered in the study. The bathymetry data for Bera Lake only cover open surface areas where accessibility is not limited, while the presence of submerged species in Bukit Merah was highly variable during the study period. The maximal allowable time step for the models follows the Courant-Friedrichs-Lewy (CFL) condition stability criterion with values completed at 3, 10, and 2 s for Bera, Bukit Merah, and Durian Tunggal, respectively. In this respect, we discuss the impact of drought and water management on the hydrodynamic pattern of the water bodies.

), <sup>r</sup> the seawater density (g�cm�<sup>3</sup>

perature. T and S represent water temperature (�C) and salinity (ppt), respectively.

∂θ ∂z �Hk�<sup>1</sup>

¼ ½ �� ∇ � ð Þ hkKH∇ θ<sup>k</sup> þ Kz

), and Pa the atmospheric pressure (g�cm�<sup>2</sup>

�s �1

depth (cm), v = (u, v) the horizontal velocity components (cm�s

r ¼ rð Þ S; T (3)

Assessing the Hydrodynamic Pattern in Different Lakes of Malaysia

http://dx.doi.org/10.5772/intechopen.73274

(4)

75

)

∂θ ∂z �Hk

�1

), and θ is the conservative physical value such as tem-

) in the x, y direction, w the

), ζ the sea-surface level (cm), f<sup>0</sup> the

). M<sup>k</sup> = hk�v<sup>k</sup> is volume transport

), A<sup>Z</sup> and KZ are vertical viscosity

Figure 2. Bathymetry of study lakes and their major inflow.

Malaysia or published information in [11, 12]. Additional measurements of main river discharges were also carried out in Bukit Merah and Bera lakes. Current data were based on current measurements using Acoustic Doppler Current Profiler (ADCP).

This study used a three-dimensional rectilinear grid hydrodynamic model to simulate the hydrodynamic structures of the lakes. The numerical model comprises integrated momentum, continuity, and heat transfer equations together with the equation of state developed by National Institutes of Japan [9]. The physical processes are described by fluid motion (Eq. (1)), flow continuity (Eq. (2)), state equation (Eq. 3), and heat transfer conservation (Eq. (4)), and the numerical schemes are adapted in-line with the methodology described in [13, 14].

For each layer k, the depth integrated momentum equation is described as follows:

$$\begin{split} \frac{\partial \mathbf{M}\_k}{\partial t} + (\mathbf{v} \cdot \nabla) \mathbf{M}\_k + (\mathbf{v}w)|\_{-H\_{k-1}} - (\mathbf{v}w)|\_{-H\_k} - f\_0 \mathbf{k} \times \mathbf{M}\_k \\ = \frac{h\_k}{\rho\_k} \left( \mathbf{u} \mathbf{l}\_k - \frac{1}{2} g h\_k \nabla \rho\_k \right) + [\nabla \cdot (A\_H \nabla)] \mathbf{M}\_k \\ + \left. \left( A\_z \frac{\partial \mathbf{v}}{\partial z} \right) \right|\_{-H\_{k-1}} - \left. \left( A\_z \frac{\partial \mathbf{v}}{\partial z} \right) \right|\_{-H\_k} \end{split} \tag{1}$$

Assessing the Hydrodynamic Pattern in Different Lakes of Malaysia http://dx.doi.org/10.5772/intechopen.73274 75

$$\frac{\partial \underline{\varsigma}}{\partial t} = \underline{w} - \nabla \cdot M\_{1\prime} \, \underline{w}\_{k-1} = \underline{w}\_k - \nabla \cdot M\_k \quad \left(\underline{2\not\leq} \underline{k} \, \underline{\leq} K^{-1}\right), \tag{2}$$

$$
\rho = \rho(S, T) \tag{3}
$$

$$\begin{split} \frac{\partial}{\partial t} (h\_k \cdot \theta\_k) + \nabla \cdot (\mathbf{M}\_k \cdot \theta\_k) + (w\theta)|\_{-H\_{k-1}} - (w\theta)|\_{-H\_k} \\ \frac{\partial}{\partial t} = \left[ \nabla \cdot (h\_k K\_H \nabla) \right] \cdot \theta\_k + \left( K\_z \frac{\partial \theta}{\partial z} \right) \Big|\_{-H\_{k-1}} - \left( K\_z \frac{\partial \theta}{\partial z} \right) \Big|\_{-H\_k} \end{split} \tag{4}$$

In these equations, hk (k = 1, 2, …, K) represents thickness (cm) of each k layer, H the still-water depth (cm), v = (u, v) the horizontal velocity components (cm�s �1 ) in the x, y direction, w the vertical velocity (cm�s �1 ), <sup>r</sup> the seawater density (g�cm�<sup>3</sup> ), ζ the sea-surface level (cm), f<sup>0</sup> the Coriolis parameter (s �1 ) which is described as <sup>f</sup> <sup>0</sup> <sup>¼</sup> <sup>2</sup><sup>ω</sup> sinφ<sup>0</sup> with the angular velocity <sup>ω</sup> (s�<sup>1</sup> ) of the earth rotation and the mean latitude φ<sup>0</sup> of the lake, g the gravitational acceleration (= 980 cm�s �2 ), and Pa the atmospheric pressure (g�cm�<sup>2</sup> �s �1 ). M<sup>k</sup> = hk�v<sup>k</sup> is volume transport integrated into vertically in each layer, ∇ the horizontal gradient operator, AH and KH the horizontal viscosity and diffusivity coefficients (cm�<sup>2</sup> �s �1 ), A<sup>Z</sup> and KZ are vertical viscosity and diffusivity coefficients (cm�<sup>2</sup> �s �1 ), and θ is the conservative physical value such as temperature. T and S represent water temperature (�C) and salinity (ppt), respectively.

The model employs the turbulence closure scheme, the k–L model, that considers turbulence kinetic energy (k) and mixing length (L) for hydrodynamic simulation due to the importance of vertical turbulence transport processes. The buoyancy effect due to stratification was based on Munk-Anderson empirical relationship [13]. Knudsen's expression was adopted for calculating the water density from water temperature and salinity. The heat exchange calculation through the water surface is estimated based on the contribution from the sensible heat flux, and latent heat flux, short wave radiation and long wave radiation. Meteorological data, such as air temperature, wind speed, and solar radiation, provide the input for the surface heat balance.

Malaysia or published information in [11, 12]. Additional measurements of main river discharges were also carried out in Bukit Merah and Bera lakes. Current data were based on

This study used a three-dimensional rectilinear grid hydrodynamic model to simulate the hydrodynamic structures of the lakes. The numerical model comprises integrated momentum, continuity, and heat transfer equations together with the equation of state developed by National Institutes of Japan [9]. The physical processes are described by fluid motion (Eq. (1)), flow continuity (Eq. (2)), state equation (Eq. 3), and heat transfer conservation (Eq. (4)), and the

<sup>∂</sup><sup>t</sup> <sup>þ</sup> ð Þ <sup>v</sup> � <sup>∇</sup> <sup>M</sup><sup>k</sup> <sup>þ</sup> ð Þj <sup>v</sup><sup>w</sup> �Hk�<sup>1</sup> � ð Þj <sup>v</sup><sup>w</sup> �Hk � <sup>f</sup> <sup>0</sup><sup>k</sup> � <sup>M</sup><sup>k</sup>

� Az

∂v ∂z �Hk

þ ½ � ∇ � ð Þ AH∇ M<sup>k</sup>

(1)

numerical schemes are adapted in-line with the methodology described in [13, 14]. For each layer k, the depth integrated momentum equation is described as follows:

> <sup>Ψ</sup><sup>k</sup> � <sup>1</sup> 2 ghk∇r<sup>k</sup>

∂v ∂z �Hk�<sup>1</sup>

current measurements using Acoustic Doppler Current Profiler (ADCP).

∂M<sup>k</sup>

Figure 2. Bathymetry of study lakes and their major inflow.

74 Applications in Water Systems Management and Modeling

¼ hk rk

þ Az

The main inputs of the model were the river inflow; meteorological parameter, such as wind, rainfall, and air temperature; and bathymetry data. Comparison between different wind speeds and under calm wind conditions was made to evaluate the circulation pattern and transport phenomena. The model assumes no physical effect of the flow field and no contribution from the groundwater. Despite the fact that a large quantity of macrophytes, such as submerged species Cabomba furcata in Bukit Merah Lake and emergent plants Pandanus helicopus in Bera Lake, may shape the hydrodynamic features, this effect was not considered in the study. The bathymetry data for Bera Lake only cover open surface areas where accessibility is not limited, while the presence of submerged species in Bukit Merah was highly variable during the study period. The maximal allowable time step for the models follows the Courant-Friedrichs-Lewy (CFL) condition stability criterion with values completed at 3, 10, and 2 s for Bera, Bukit Merah, and Durian Tunggal, respectively. In this respect, we discuss the impact of drought and water management on the hydrodynamic pattern of the water bodies. The main characteristics of runs are explained in Table 1.
