67
Groundwater Modeling Involving PRBs
where
Q = volumetric flow rate (m 3 /s or ft 3 /s)
A = flow area perpendicular to L (m 2 or ft 2 )
K = hydraulic conductivity (m/s or ft/s)
l = flow path length (m or ft)
h = hydraulic head (m or ft)
Δ = denotes the change in h over the path L
The specific requirements of the simulation models for the PRB modeling are summarized in Gupta and Fox (1999). In view of possible underflow,
overflow, and interaction between adjacent aquifers, a three-dimensional
model is the most suitable option. Hanging barriers (Chapters 2 and 3) will
result in significant vertical flow gradients and it is critical that the temporal distribution of vertical flow velocities should be accurately generated.
Another requirement for the simulation model is that it should consider
the variability of hydraulic conductivity induced by the PRB. Generally, the
reactive material we use for the barrier has a higher K value than the surrounding matrix. This will result in a significant gradient in the hydraulic
conductivity of the aquifer. In the case of a funnel-and-gate type of PRB, the
funnel walls are very thick and highly impermeable, this is in contrast with
the high-permeability barrier. For further information on funnel-and-gate
type of PRBs, readers are referred to Chapters 2 and 3. The stability conditions of most numerical models will not allow us to solve the resulting highcontrast hydraulic conductivity distributions.
The finite difference-based three-dimensional groundwater flow model
MODFLOW (McDonald and Harbaugh 1988) is the most widely used model
for simulating groundwater flow. MT3DMS (Zheng and Wang, 1999) and
RT3D (Clement et al., 1997) are also common reactive transport models
and these models can be used in conjunction with MODFLOW. Hsieh and
Freckleton (1993) developed a computer program to simulate the horizontal
flow barriers using a finite difference model. In order to evaluate the capture zone of the PRBs, the model results should be compatible with the use
of particle tracking algorithms. MODPATH (Pollock, 1989) is a widely used
particle tracking code used in conjunction with MODFLOW. The finite element-based flow and transport simulation model, FEFLOW (Diersch, 2013)
can also be used to model PRB systems. Finite element-based models have
advantages over finite difference-based models in terms of the stability criteria, and they have the potential to address the complex nature of the conditions induced by subsurface heterogeneity.
Many sites across the world—such as Somersworth Landfill, Sunnyvale,
CA, USA—have used MODFLOW in the PRB modeling. FLOWPATH developed by Waterloo Hydrogeologic has been used in Belfast, Ireland, and a
DOE site in Kansas (USA) to evaluate the design of PRBs. Besides these,
FLONET (Guiguer et al., 1992), FRAC3DVS (Therrien and Sudicky, 1995) also
been used at some sites (Gupta and Fox, 1999).
Groundwater Modeling Involving PRBs
where
Q = volumetric flow rate (m 3 /s or ft 3 /s)
A = flow area perpendicular to L (m 2 or ft 2 )
K = hydraulic conductivity (m/s or ft/s)
l = flow path length (m or ft)
h = hydraulic head (m or ft)
Δ = denotes the change in h over the path L
The specific requirements of the simulation models for the PRB modeling are summarized in Gupta and Fox (1999). In view of possible underflow,
overflow, and interaction between adjacent aquifers, a three-dimensional
model is the most suitable option. Hanging barriers (Chapters 2 and 3) will
result in significant vertical flow gradients and it is critical that the temporal distribution of vertical flow velocities should be accurately generated.
Another requirement for the simulation model is that it should consider
the variability of hydraulic conductivity induced by the PRB. Generally, the
reactive material we use for the barrier has a higher K value than the surrounding matrix. This will result in a significant gradient in the hydraulic
conductivity of the aquifer. In the case of a funnel-and-gate type of PRB, the
funnel walls are very thick and highly impermeable, this is in contrast with
the high-permeability barrier. For further information on funnel-and-gate
type of PRBs, readers are referred to Chapters 2 and 3. The stability conditions of most numerical models will not allow us to solve the resulting highcontrast hydraulic conductivity distributions.
The finite difference-based three-dimensional groundwater flow model
MODFLOW (McDonald and Harbaugh 1988) is the most widely used model
for simulating groundwater flow. MT3DMS (Zheng and Wang, 1999) and
RT3D (Clement et al., 1997) are also common reactive transport models
and these models can be used in conjunction with MODFLOW. Hsieh and
Freckleton (1993) developed a computer program to simulate the horizontal
flow barriers using a finite difference model. In order to evaluate the capture zone of the PRBs, the model results should be compatible with the use
of particle tracking algorithms. MODPATH (Pollock, 1989) is a widely used
particle tracking code used in conjunction with MODFLOW. The finite element-based flow and transport simulation model, FEFLOW (Diersch, 2013)
can also be used to model PRB systems. Finite element-based models have
advantages over finite difference-based models in terms of the stability criteria, and they have the potential to address the complex nature of the conditions induced by subsurface heterogeneity.
Many sites across the world—such as Somersworth Landfill, Sunnyvale,
CA, USA—have used MODFLOW in the PRB modeling. FLOWPATH developed by Waterloo Hydrogeologic has been used in Belfast, Ireland, and a
DOE site in Kansas (USA) to evaluate the design of PRBs. Besides these,
FLONET (Guiguer et al., 1992), FRAC3DVS (Therrien and Sudicky, 1995) also
been used at some sites (Gupta and Fox, 1999).