THE DEPARTMENT OF DEFENSE

Groundwater Modeling System

SEEP2D PRIMER
SEEP2D Primer

Copyright © 1998 Brigham Young University - Engineering Computer

Graphics Laboratory

All Rights Reserved

Unauthorized duplication of the GMS software or documentation is strictly

prohibited.

THE BRIGHAM YOUNG UNIVERSITY ENGINEERING COMPUTER

GRAPHICS LABORATORY MAKES NO WARRANTIES EITHER
EXPRESS OR IMPLIED REGARDING THE PROGRAM GMS AND ITS
FITNESS FOR ANY PARTICULAR PURPOSE OR THE VALIDITY OF
THE INFORMATION CONTAINED IN THIS USER'S MANUAL

Last revision: February 17, 1998

TABLE OF CONTENTS

1 OVERVIEW OF SEEP2D MODELING ............................................................................................... 1-1

1.1 INTRODUCTION .................................................................................................................................... 1-1
1.2 APPLICATIONS ..................................................................................................................................... 1-1
1.3 GOVERNING EQUATION ...................................................................................................................... 1-2
1.4 MODELING PROCESS ........................................................................................................................... 1-2
1.4.1 Mesh Construction........................................................................................................................ 1-3
1.4.2 Boundary Conditions.................................................................................................................... 1-3
1.4.3 SEEP2D........................................................................................................................................ 1-3
1.4.4 Post-Processing With GMS .......................................................................................................... 1-3
1.5 PRIMER OVERVIEW ............................................................................................................................. 1-3
2 MODEL CONCEPTUALIZATION ...................................................................................................... 2-1
2.1 INTRODUCTION .................................................................................................................................... 2-1
2.2 DETERMINING APPROPRIATE BOUNDARY CONDITIONS ................................................................ 2-1
2.2.1 No Flow Boundary Conditions ..................................................................................................... 2-1
2.2.2 Constant Head Boundary Conditions........................................................................................... 2-2
2.2.3 Exit Face Boundary Conditions.................................................................................................... 2-2
2.2.4 Flow Rate Boundary Conditions .................................................................................................. 2-2
2.2.5 Flux Boundary Conditions............................................................................................................ 2-2
2.3 CONFINED FLOW PROBLEMS ............................................................................................................. 2-2
2.4 UNCONFINED FLOW PROBLEMS ........................................................................................................ 2-4
2.4.1 Deforming Mesh ........................................................................................................................... 2-4
2.4.2 Unsaturated Flow ......................................................................................................................... 2-5
2.5 IMPERMEABLE FLOW BARRIERS ....................................................................................................... 2-5
2.6 INTERNAL DRAINS .............................................................................................................................. 2-6
2.7 CORES OF DAMS .................................................................................................................................. 2-7
2.8 FLOW TO A WELL ............................................................................................................................... . 2-8
2.9 PLAN OR AERIAL VIEW MODELS ....................................................................................................... 2-9
3 MESH CONSTRUCTION GUIDELINES ............................................................................................ 3-1
3.1 INTRODUCTION .................................................................................................................................... 3-1
3.2 MESH CONSTRUCTION ........................................................................................................................ 3-1
3.2.1 Mesh Construction With GMS ...................................................................................................... 3-2
3.2.2 Basic Element Types ..................................................................................................................... 3-2
3.2.3 Material IDs ................................................................................................................................. 3-2
3.2.4 Renumbering................................................................................................................................. 3-2
3.3 MESH DENSITY .................................................................................................................................... 3-4
4 MODEL PARAMETERS ....................................................................................................................... 4-1
4.1 INTRODUCTION .................................................................................................................................... 4-1
4.2 ANALYSIS PARAMETERS .................................................................................................................... 4-1
4.2.1 Title............................................................................................................................................... 4-1
4.2.2 Datum ........................................................................................................................................... 4-1
4.2.3 Unit Weight of Water .................................................................................................................... 4-2
4.2.4 Type of Flow ................................................................................................................................. 4-2
4.2.5 Confined vs. Unconfined............................................................................................................... 4-2
ii SEEP2D Primer

4.2.6 Saturated/Unsaturated Flow Modeling ........................................................................................ 4-2

4.2.7 Flow Lines .................................................................................................................................... 4-5
4.3 MATERIAL PROPERTIES ...................................................................................................................... 4-5
4.3.1 Hydraulic Conductivity................................................................................................................. 4-6
4.3.2 Unsaturated Zone Parameters...................................................................................................... 4-6
5 RUNNING SEEP2D................................................................................................................................. 5-1
5.1 INTRODUCTION .................................................................................................................................... 5-1
5.2 FILES ..................................................................................................................................................... 5-1
5.2.1 Input Files..................................................................................................................................... 5-1
5.2.2 Output Files .................................................................................................................................. 5-1
5.3 RUNNING SEEP2D ............................................................................................................................... 5-2
5.4 TROUBLE SHOOTING ........................................................................................................................... 5-2
5.4.1 Floating Point Error..................................................................................................................... 5-2
5.4.2 Maximum Number of Nodes or Elements Exceeded ..................................................................... 5-2
5.4.3 Unable to Compute Flow Lines .................................................................................................... 5-2
5.4.4 Mesh is Deformed when Modeling a Confined Problem .............................................................. 5-3
5.4.5 Maximum Iterations Reached ....................................................................................................... 5-3
5.4.6 Problem Unexpectedly Aborts, or Takes an Unusual Amount of Time......................................... 5-3
1Overview Of SEEP2D Modeling

CHAPTER 1
Overview Of SEEP2D Modeling

1.1 Introduction
This document is a primer for the GMS/SEEP2D numerical modeling system.
It should be read completely before any modeling is attempted with SEEP2D.
The purpose of this primer is as follows:

1. To provide a general introduction to the SEEP2D program.

2. To provide the user with some guidelines for mesh generation and
selection of boundary conditions and model parameters to help
minimize improper application of SEEP2D.

The SEEP2D software was developed by the United States Army Engineer
Waterways Experiment Station to model a variety of problems involving
seepage. It is assumed that the reader will be using GMS in conjunction with
SEEP2D. GMS is a graphical pre- and post-processor for SEEP2D. GMS was
developed by the Brigham Young University Engineering Computer Graphics
Laboratory in cooperation with the Waterways Experiment Station.

1.2 Applications
The following conditions can be modeled using SEEP2D:

1. Isotropic and anisotropic soil properties.

2. Confined and unconfined flow for profile models.

1-2 SEEP2D Primer

3. Saturated/unsaturated flow for unconfined profile models.

4. Confined flow for plan (areal) models.

5. Flow simulation in the saturated and unsaturated zones.

6. Heterogeneous soil conditions.

7. Axisymmetric models such as flow from a well.

8. Drains.

The following conditions cannot be modeled using SEEP2D:

1. Transient or time varying problems

2. Unconfined plan models

1.3 Governing Equation

The governing equation used in the SEEP2D models is:

∇ • ( K • ∇h) = 0 ..................................................................................... 1.1

or

∂  ∂h ∂h  ∂  ∂h ∂h 
 Kxx + Kxy  +  Kyy + Kyx  = 0 ............................ 1.2
∂x  ∂x ∂y  ∂y  ∂y ∂x 

where:

h = total head (elevation head plus pressure head)

K = hydraulic conductivity

This equation is often referred to in the literature as the Laplace equation.

1.4 Modeling Process

In a typical modeling problem involving the SEEP2D software, a series of
tasks are performed in a specific sequence. Each of these steps is described
briefly.
 Overview of SEEP2D Modeling 1-3

1.4.1 Mesh Construction

First of all, a finite element mesh must be constructed which represents the
region being modeled. This mesh is typically constructed using GMS. A
variety of mesh generation and interactive editing tools are provided in GMS.
These tools are described in more detail in the GMS Tutorial and GMS
Reference Manual.

1.4.2 Boundary Conditions

Once a mesh has been constructed, boundary conditions are applied to the
mesh. Boundary conditions are typically entered as constant head at a node,
head equals elevation (exit face) at a node, or as an incoming flux at a node or
along an element edge. Hydraulic conductivities must also be entered for the
two principal directions for different regions (representing different soil types)
in the mesh. All of these parameters can be input interactively using GMS.
The mesh geometry and boundary conditions are saved by GMS to a SEEP2D
input file.

1.4.3 SEEP2D
Once the mesh is constructed, the SEEP2D program can be executed to
calculate the head, flow, discharge (Darcian) velocity, and pore pressure at
every node in the mesh. For problems encountered when running SEEP2D,
refer to section 5.4 of this primer.

1.4.4 Post-Processing With GMS

After running SEEP2D, results may be viewed in GMS. GMS displays
contours of equipotential total or pressure head, contours of pore pressures,
velocity vectors, and, for many classes of problems, flow lines. In addition,
the summed flow of all selected nodes can be viewed. Upon viewing the
solution, the user must ascertain if the results are reasonable. If necessary, the
mesh should be refined, boundary conditions modified, or input coefficients
altered, and a new solution computed.

1.5 Primer Overview

Each of the steps of the modeling process outlined above are described in
more detail in the remainder of this primer. Some general guidelines
concerning model conceptualization, particularly with regard to the use of
boundary conditions, are given in Chapter 2. Mesh construction guidelines are
outlined in Chapter 3. Chapter 4 describes SEEP2D analysis parameters other
than boundary conditions, such as the datum and material properties. Finally,
in Chapter 5, details concerning the execution of SEEP2D are given, including
a section on trouble shooting of commonly encountered problems.
2Model Conceptualization

CHAPTER 2
Model Conceptualization

2.1 Introduction
SEEP2D is used to model seepage conditions for actual physical problems.
Before a problem can be modeled, the subregion of the actual site to be
modeled must be determined and a set of appropriate boundary conditions
must be selected. This process is called model conceptualization. Model
conceptualization is perhaps the most important part in developing any
SEEP2D model. The accuracy of the model will be significantly influenced by
the choices made during the model conceptualization process.

2.2 Determining Appropriate Boundary Conditions

Deciding what type of boundary conditions to use is the most important part of
model conceptualization. The boundary conditions define the flow model.
Constant head boundary conditions are typically used to represent standing
bodies of water. However, velocity, exit face, and flux type boundaries are
also important for modeling certain situations.

2.2.1 No Flow Boundary Conditions

If a boundary condition has not been explicitly applied to a boundary node, the
node is assumed to be a "no-flow" boundary and the flow direction will be
computed parallel to the boundary. Thus, all boundary nodes are no flow
boundaries by default.
2-2 SEEP2D Primer

2.2.2 Constant Head Boundary Conditions

Constant head boundary conditions represent boundaries where the head is
known. They typically are found where water is ponding or at the boundary of
a region where the water table is known to remain constant. Since the head
along such boundaries cannot change, they represent regions of the model
where flow enters or exits the system (flow lines are always orthogonal to
constant head boundaries).

2.2.3 Exit Face Boundary Conditions

Exit face boundary conditions imply that the head is equal to the elevation
(assuming that the datum is 0). They are used when modeling unconfined
flow problems and should be placed along the face where the free surface is
likely to exit the model. This boundary condition must be used if the option
for deforming the mesh to the phreatic surface has been selected. It may also
be used with a saturated/unsaturated flow model. In this case, if the head at a
node on the boundary becomes greater than the node elevation during the
iteration process, the head at the node is fixed at the nodal elevation and the
node acts as a specified head boundary. Thus, water is allowed to exit the
boundary above the tailwater. If an exit face boundary is not used with a
saturated/unsaturated flow model, all of the flow will be forced through the
tailwater.

2.2.4 Flow Rate Boundary Conditions

Flow rate boundary conditions are used to specify nodes at which a certain
flow rate is known to exist. They are used primarily when modeling wells and
the flow specified represents the pumping rate. Negative values represent
extraction of fluid from the system whereas positive values represent injection.

2.2.5 Flux Boundary Conditions

Flux boundary conditions are used to specify a known flux rate [L/T] along a
sequence of element edges on the perimeter of the mesh. They are often used
to simulate infiltration. Flux into the system is positive and flux out of the
system is negative.

2.3 Confined Flow Problems

The governing equation defined in the previous chapter assumes that the
porous media is saturated. This is always the case for confined flow problems
where the upper mesh boundary defines the limit of the saturated zone. These
types of boundaries typically occur in areas where a standing body of water or
an impervious clay liner or material overlays the region of interest as
demonstrated in Figure 2.1 and Figure 2.2. Confined flow problems are
 Model Conceptualization 2-3

modeled using a combination of constant head and no flow boundary

conditions.

The vertical boundaries on the sides of the models shown in Figure 2.1 and
Figure 2.2 are modeled using no flow boundary conditions. They can also be
modeled using constant head boundary conditions. Both approaches will give
similar solutions provided the region is extended a suitable distance as
discussed below.

In addition to determining which boundary conditions are appropriate for the

ends of the models shown in Figure 2.1 and Figure 2.2, the user must also
determine how far to extend the model in either direction. In either case, most
of the flow is concentrated in the vicinity of the sheet piles. As the mesh is
extended in either direction, a point is reached where extending the mesh
further makes very little difference in the solution. A good approach to
determining how far to extend a mesh is to compute a series of solutions where
the mesh is extended for each subsequent solution until the extension is seen to
have little effect on the solution.

Constant Head Boundaries

Figure 2.1 Confined Flow With Standing Water on Either Side of a Flow
Barrier.

Constant Head Boundaries

Clay Blanket

Figure 2.2 Confined Flow Where a Clay Blanket Exists.

2-4 SEEP2D Primer

2.4 Unconfined Flow Problems

With unconfined problems, the position of the free surface and the point
downstream where the free surface exits is unknown. The Dupuit problem
shown in Figure 2.3 is a classic example of an unconfined flow problem. This
same type of condition exists when modeling flow through a dam as shown in
Figure 2.4. Unconfined flow problems can be modeled in SEEP2D by either
deforming the mesh to the phreatic surface so that flow only occurs in the
saturated zone or by simulating flow in both the saturated and unsaturated
zone

Exit Face Boundary

x
Constant Head Boundaries

Figure 2.3 Unconfined Flow Represented by the Dupuit Problem.

Free Surface
Exit Face Boundary

Constant Head Boundaries

Figure 2.4 Unconfined Flow Through a Dam

2.4.1 Deforming Mesh

If the deforming mesh option is selected, the boundary conditions are
implemented by placing constant head boundary conditions in the locations
where the head and tail water are known, and placing exit face boundary
conditions along the boundary where the phreatic surface is assumed to exit as
shown in Figure 2.3 and Figure 2.4. SEEP2D will automatically solve for the
free surface and deform the mesh to this boundary. The resulting set of
 Model Conceptualization 2-5

solution files will include a geometry file containing the nodes and elements in
the deformed mesh.

2.4.2 Unsaturated Flow

If the option to model saturated/unsaturated flow is selected, the mesh in not
deformed. Rather the flow in the entire problem domain, both in the saturated
and the unsaturated zone is modeled. In this case, a method must be selected
for defining the relative conductivity in the unsaturated zone and parameters
related to that method must be specified for each of the material zones. These
options are defined in more detail in sections 4.2.5 and 4.2.6. If an exit face
boundary is used, water is allowed to exit the model along the exit face.
Otherwise, all of the water is forced to exit at the tail water.

2.5 Impermeable Flow Barriers

Sheet piles, grout curtains and other structures used to create impermeable
flow barriers can be modeled in SEEP2D by creating a discontinuity or
"crack" in the mesh. By default the boundary of a finite element mesh is a "no
flow" boundary, therefore flow barriers such as the ones shown in Figure 2.5a
are modeled by making the mesh boundary extend in a crack like fashion
down to the depth of the barrier as shown in Figure 2.5b (the "crack" in the
mesh has been exaggerated for illustration purposes). Since gradients are
typically higher around the tip of such barriers, it is often useful to refine the
mesh more in these regions.
2-6 SEEP2D Primer

Impervious
Fill

Sheet Piles

(a)

Sheet Piles

(b)

Figure 2.5 a) Sheet Piles Creating an Impermeable Flow Barrier (b) A

Mesh Representing the Sheet Piles.

2.6 Internal Drains

Internal drains can be modeled in SEEP2D by placing a hole in the mesh and
by applying either constant head or exit face boundary conditions to the nodes
on the perimeter of the hole. If the drain is placed below a region where water
is ponding or in a region where it is anticipated that the soil will remain
saturated, constant head boundary conditions should be assigned to the drain.
For example, the model shown in Figure 2.6 contains a sheet pile with a drain
at the bottom of the sheet pile. The water is expected to pond on both sides of
the sheet pile.

Constant Head Boundaries

Constant Head Boundary

Figure 2.6 Confined Approach for Modeling an Internal Drain.

 Model Conceptualization 2-7

In some cases, there is not a sufficient supply of water for the water to pond
above a drain and the drain should be modeled as an unconfined problem using
the saturated/unsaturated option with the boundary conditions shown in Figure
2.7. Another approach is to use the deforming mesh option with exit face
boundary conditions as shown in Figure 2.8. In this example it is assumed that
water is ponded on the downstream side of the flow barrier but not on the
upstream side.

Constant Head Boundaries

Constant Head Boundary

Figure 2.7 Saturated/Unsaturated Approach for Modeling an Internal Drain.

Constant Head Boundaries

Free Surface

Exit Face Boundary

Figure 2.8 Deforming Mesh Approach for Modeling an Internal Drain.

2.7 Cores of Dams

When modeling an earth dam with a core, the core can be modeled by forcing
the element edges to conform to the boundary of the core and by assigning
separate material properties to the core and shell of the dam (Figure 2.9).
2-8 SEEP2D Primer

Core

Figure 2.9 Earth Dam with a Core.

If the deforming mesh options is selected and if the hydraulic conductivity of

the core material differs by more than two orders of magnitude from the
hydraulic conductivity of the shell material, then SEEP2D may have
difficulties computing the location of the phreatic surface. This is due to the
fact that the slope of the phreatic surface is very steep in the core but is very
flat in the shell. The transition from a steep slope to a flat slope may cause
SEEP2D to have difficulty converging on a solution. In such cases, the
portion of the model downstream from the core can be omitted as shown in
Figure 2.10 without significant effect on the solution since most of the
drawdown occurs in the core of the dam.

Core

Figure 2.10 Suggested Boundaries when Modeling a Core.

2.8 Flow to a Well

Flow to a well is a classic flow problem which can be solved using SEEP2D.
For such a condition the axisymmetric option of SEEP2D should be selected
(see section 4.2.4). The axis of symmetry for the model is the axis of the well.
In other words, the well should be established along the y axis. The well
should be placed on the left edge of the model and the origin for the coordinate
system should be placed at the lower left corner of the model as shown in
Figure 2.11. For a fully penetrating well, the left edge of the mesh should
begin at x=r where r is the well radius. In the case of a partially penetrating
well, a notch the thickness of the well radius should be placed in the mesh to
the depth of the well as shown in Figure 2.11

The boundary conditions assigned to a well problem depend on the type of

aquifer being modeled. In most cases, a constant head boundary condition
should be applied to the right end of the model. For a confined aquifer, a flow
rate boundary condition should also be placed at the bottom of the well. For
 Model Conceptualization 2-9

an unconfined aquifer, a flow rate boundary condition should be placed at the

bottom of the well and, if the deforming mesh option is being used, exit face
boundary conditions should be applied to the nodes on the well above the
bottom.

Well

(r,d)
d

(0,0) x

Figure 2.11 Boundary for a Partially Penetrating Well.

2.9 Plan or Aerial View Models

The governing equation used for solving profile seepage models of confined
aquifers is as follows:

∇ • (T • ∇h) = 0 ......................................................................................2.8

or

∂  ∂h ∂h  ∂  ∂h ∂h 
 Txx + Txy  +  Tyy + Tyx  = 0 ................................2.9
∂x  ∂x ∂y  ∂y  ∂y ∂x 

where:

h = total head (elevation head plus pressure head)

T = transmissivity.

This equation is the same as the equation used in SEEP2D except that the
hydraulic conductivity terms (k) have been replaced by transmissivity terms (T
= k X aquifer thickness). Therefore, profile seepage models of confined
aquifers can be solved using SEEP2D as long as transmissivities are entered
for the hydraulic conductivities.
2-10 SEEP2D Primer

Since the governing equation used for modeling unconfined aquifers is

different than the equation used by SEEP2D, unconfined plan view models
cannot be modeled using SEEP2D.

For plan view models, the values used for the constant head boundary
conditions may need to be shifted by a constant value. For example, the model
shown in Figure 2.12a represents a simple plan view model with constant head
boundary conditions on two sides and no flow boundary conditions on the
other two sides. Since the y dimension of the model is 500’, the y coordinates
of the nodes at the top of the model will be 500’. If a constant head boundary
condition of 10’ is assigned to the nodes at the top of the model, SEEP2D will
assume that the model is an unconfined profile model (since h=500’ > h=10’)
and it will attempt to draw down a phreatic surface. This problem can be
avoided by ensuring that all constant head boundary conditions are given a
value that is greater than the value of any of the y coordinates in the model as
shown in Figure 2.12b. The constant added to the heads can be subtracted
from the computed heads. The problem could also be solved by setting the
datum to 500’ and the respective heads to 10’ and 6’.

River River
Head = 10’ Head = 510’

500’

Head = 6’ Head = 506’

(a) (b)

Figure 2.12 Plan View Modeling (a) Actual Parameters (b) Model
Parameters
3Mesh Construction Guidelines

CHAPTER 3
Mesh Construction Guidelines

3.1 Introduction
A fundamental part of solving a seepage problem using SEEP2D is the
construction of a two-dimensional finite element mesh representing the cross
section or region being modeled. This mesh is typically constructed using
GMS. Some general guidelines concerning the construction of meshes for
input into SEEP2D are described in this chapter.

3.2 Mesh Construction

The finite element mesh used by SEEP2D is composed of nodes and elements.
A sample mesh is shown in Figure 3.1. A finite element mesh can be thought
of as a cross section representing the region to be modeled that is formed by
piecing together a large number of small triangular and quadrilateral patches
called "elements". The nodes are the xy points that define the geometry of the
mesh. The elements define the mesh topology and are formed by connecting a
set of nodes in the mesh with edges.
3-2 SEEP2D Primer

Figure 3.1 Sample Finite Element Mesh.

3.2.1 Mesh Construction With GMS

Finite element meshes can be constructed using GMS. There are a large
number of tools available in GMS to assist the user in the construction and
editing of meshes. Details concerning the mesh construction and editing tools
are provided in the GMS Tutorial and GMS Reference Manual.

3.2.2 Basic Element Types

Two types of elements, triangles and quadrilaterals are commonly used in the
construction of two-dimensional meshes (Figure 3.2). Quadratic elements
(elements with midside nodes) are not supported by SEEP2D.

(a) (b)

Figure 3.2 The Two Basic Elements: (a) Linear Triangles (b) Linear
Quadrilaterals

3.2.3 Material IDs

Each element in the mesh should have an associated material ID. The material
ID is an index to a list of material properties (hydraulic conductivities). The
material properties are described in more detail in the Chapter 4.

3.2.4 Renumbering
Each node and element in the mesh has an associated ID. The order in which
the nodes are numbered is very critical and should be well understood by the
 Mesh Construction Guidelines 3-3

user. The node and element numbering sequence can be altered using GMS.
The first step in altering the numbering sequence is to select a node string. A
node string is a sequence of nodes which is typically on the boundary of the
mesh. GMS can be used to automatically renumber the nodes and elements
using this node string. The nodes and elements are numbered by first
numbering the nodes and elements connected to the string and then numbering
the remainder of the mesh by progressing outward from the string in a
sweeping fashion (Figure 3.3).

6 12 18 24 21 22 23 24

5 11 17 23 17 18 19 20
Node String

4 10 16 22 13 14 15 16

3 9 15 21 9 10 11 12

2 8 14 20 5 6 7 8

1 7 13 19 1 2 3 4

Node String
Half Band Width = 8 Half Band Width = 6
(a) (b)

Figure 3.3 Sample Results of Renumbering Process. (a) Results With Node
String on End of Mesh. (b) Results With Node String on Side of
Mesh.

As seen in Figure 3.3, the location of the node string controls the node and
element numbering. The result of the numbering sequence is represented by
the maximum nodal half band width. This parameter is related to the
maximum difference in ID’s of the nodes defining an element. Since the
solution time and the memory requirements for SEEP2D are proportional to
the square of the maximum nodal half band width, different numbering
sequences on the same mesh can produce drastically different solution times.
Different node strings can be selected and tested to find the string resulting in
the smallest half band width. In many cases, the optimal location of the node
string is immediately obvious. If the mesh is shaped such that there are
distinct longitudinal (major axis) and lateral (minor axis) dimensions, the node
string should be placed on one end of the mesh such that the numbering
sequence progresses longitudinally along the mesh as shown in Figure 3.3b.
This tends to minimize the nodal band widths.
3-4 SEEP2D Primer

Since the numbering sequence has such a dramatic effect on the SEEP2D
solution time and memory requirements, the mesh should always be
renumbered after the mesh has been constructed or after the mesh has been
edited.

3.3 Mesh Density

In general, the higher the resolution of the mesh, the more accurate the
solution. While theoretically the size (number of nodes and elements) of a
SEEP2D model is unlimited, practically speaking there are some limitations.
SEEP2D is a FORTRAN program with constant length array sizes. By
default, SEEP2D allows meshes as large as 2000 nodes and 2000 elements. If
needed, the SEEP2D program can be recompiled with larger dimensions.

When increasing the density of a mesh to increase the accuracy, one approach
is to increase the density of the mesh globally (by subdividing each element).
A more efficient approach is to only refine the mesh in areas where there is
high flow or high gradient in head. For example, in constrictions around flow
barriers, at wells, and near drains the node density should be higher. Most of
the computational error is concentrated in such areas.
4Model Parameters

CHAPTER 4
Model Parameters

4.1 Introduction
Once a finite element mesh is created, and boundary conditions have been
applied, several model parameters must be set to complete the definition of the
model. These parameters include global analysis parameters and material
properties. All model parameters can be specified interactively using GMS.

4.2 Analysis Parameters

Several global options for controlling how SEEP2D solves the seepage
problem must be specified. These options define the units and the type of
problem.

4.2.1 Title
A title can be input to SEEP2D. This title is used in the header of the text
output file.

4.2.2 Datum
By default, the datum of the model is at zero, but it can be specified to any
convenient value, such as the value corresponding to the base or lowest y
coordinate of the model.
4-2 SEEP2D Primer

4.2.3 Unit Weight of Water

The unit weight of water must be entered. SEEP2D uses this value to compute
pore pressures. The weight and length units defined in this value should be
consistent with the units used elsewhere in the model.

4.2.4 Type of Flow

The type of flow must be specified either as plane flow or axisymmetric flow.
The axisymmetric option should be selected for models corresponding to flow
to a single well as described in section 2.8. All other models should use the
plane flow option.

4.2.5 Confined vs. Unconfined

The type of model should be specified as either confined or unconfined. For
confined models, the entire model domain is assumed to be saturated. No exit
face boundary conditions should be applied and the unsaturated zone material
properties are not required.

For unconfined models, two options are available for dealing with the
unsaturated zone: (1) deforming mesh and (2) saturated/unsaturated flow
modeling. For both types of problems, exit face boundary conditions should
be applied along the boundary of the mesh where the free surface is expected
to exit. With the deforming mesh option, SEEP2D iterates to find the location
of the phreatic surface and the mesh is deformed or truncated so that the upper
boundary of the mesh matches the phreatic surface. The solution files from
this type of simulation include a geometry file containing the deformed mesh.
With the saturated/unsaturated option, the mesh is not modified and the flow
in both the saturated and unsaturated zone is modeled. The hydraulic
conductivity in the unsaturated zone is modified (reduced) using either the
linear frontal method or the Van Genuchten method. The hydraulic
conductivity in the unsaturated zone is modified using the equations described
in the following section.

4.2.6 Saturated/Unsaturated Flow Modeling

For unconfined problems, SEEP2D can be used to simulate flow in the
unsaturated zone. Equation 1.1 on page 1-2 is the governing differential
equation which is solved with SEEP2D. The solution to the equation is a
function describing the total head, h, as a function of x and y. The following
equation is the general form for computing heads:

h = h p + h el − h d .................................................................................... 2.1

where:

h = total head.
 Model Parameters 4-3

hp = pressure head

hel = elevation head

hd = datum.

The hydraulic conductivity term, K, in equation 1.1 is typically thought of as

the saturated hydraulic conductivity. In other words, it is only valid when the
hp in equation 2.1 is positive. In order to model flow in unsaturated regions
(negative hp) the hydraulic conductivity can be expressed as the product of the
saturated hydraulic conductivity ks and the relative hydraulic conductivity kr as
follows:

k = k r k s ...................................................................................................2.2

As long as the porous medium is saturated (hp > 0), kr = 1, but as hp goes
negative, the value of kr decreases towards zero. By using equation 2.2 for
hydraulic conductivity, SEEP2D can be used to simulate flow in unsaturated
portions of a model. The parameter kr is specified on a material-by-material
basis. Two approaches can be used to define kr: a frontal function or the Van
Genuchten model.

Frontal Function
The frontal function method is used by numerous models for simulating flow
in unsaturated regions and is defined by specifying a kr0 and an h0 for each
material zone in the model. In the saturated zone (hp > 0) and kr = 1. In the
unsaturated zone where hp < h0, kr = kr0, and for values of hp between 0 and h0 kr
varies linearly from 1 to kr0. This is illustrated in Figure 4.1.

kro

Pressure Head

Figure 4.1 Frontal Function.

Notice that if h0=0 then the front becomes a step function as shown in Figure
4.2.
4-4 SEEP2D Primer

kro

Pressure Head

Figure 4.2 Frontal Function Degenerates to a Step Function for h0 = 0.

Equations 2.3-2.5 summarize how kr is computed for different values of hp.

kr = 1 hp > 0 ................................... 2.3

 hp 
k r = (k r 0 − 1)  + 1 ho < hp < 0............................ 2.4
 h0 

k r = k r0 hp < ho .................................. 2.5

Van Genuchten Model

Van Genuchten (1980) developed a model for computing the relative hydraulic
conductivity using a term for effective saturation, S defined by equation 2.6.

( ) 
−m
S = 1 + αh p
n
................................................................................. 2.6


where

S = effective saturation

α,n = Van Genuchten parameters

and

1
m = 1−
n

The relative hydraulic conductivity, kr, can then be defined by equation 2.7.

2
 
1 1 m

k r = S 1 − 1 − S   ...................................................................... 2.7
2 m
   
 Model Parameters 4-5

Table 2.1 provides some representative values for α and n for different soil
types.
Table 2.1
Representative Values of Van Genuchten
Parameters α and n
Soil Type α [cm-1] n
Clay** 0.008 1.09
Clay Loam 0.019 1.31
Loam 0.036 1.56
Loam Sand 0.124 2.28
Silt 0.106 1.37
Silt Loam 0.020 1.41
Silty Clay 0.005 1.09
Silty Clay Loam 0.010 1.23
Sand 0.145 2.68
Sandy Clay 0.027 1.23
Sandy Clay Loam 0.059 1.48
Sandy Loam 0.075 1.89
** Agricultural soil, less than 60% clay
Source: Carsel and Parrish (1988)

References:

Carsel, R.F., and R.S. Parrish, Developing joint probability distributions of

soil-water retention characteristics, Water Resources Research, Vol. 24, No. 5,
pp. 755-769, 1988.

Van Genuchten, M. Th., 1980, “A Closed-Formed Equation for Predicting the

Hydraulic Conductivity of Unsaturated Soils,” Soil Science Society of America
Journal, Vol. 44, No. 5.

4.2.7 Flow Lines

The primary result of a SEEP2D analysis is the total head at every node in the
model. By contouring these values in GMS, lines of equipotential head can be
displayed. Flow lines or stream function (orthogonal to the equipotential
lines) can be computed using an equation identical to the one used for solving
heads. SEEP2D computes flow lines by first computing the head values and
then redefining the boundary conditions and solving for pseudo-flow values
("flow potential" values) at the nodes. These flow values are contoured in
GMS to generate flow lines.

4.3 Material Properties

In order for a solution to be computed, material or soil properties for all
elements must be defined. The material properties include hydraulic
conductivity and unsaturated zone parameters.
4-6 SEEP2D Primer

4.3.1 Hydraulic Conductivity

SEEP2D allows for different hydraulic conductivities along the major and
minor axes (anisotropic conditions) to be defined. In addition, an orientation
angle can be entered which defines the angle between the x axis of the model
and the major axis of hydraulic conductivity as shown in Figure 4.3 below.

y
k2

k1

α
x

Figure 4.3 Material Orientation Angle.

Heterogeneous models can be created by specifying different values of

hydraulic conductivity for the elements representing the different layers or
regions.

4.3.2 Unsaturated Zone Parameters

If one of the saturated/unsaturated flow modeling options has been selected for
an unconfined model, a set of unsaturated zone parameters must also be
defined for each material. If the linear front option is selected, a minimum
pressure head (ho) and a minimum relative conductivity must be specified
(kro). If the Van Genuchten option is selected, the Van Genuchten α and n
parameters must be specified. These options are described in more detail in
Section 4.2.6.
5Running SEEP2D

CHAPTER 5
Running SEEP2D

5.1 Introduction
Once a finite element mesh has been constructed and boundary conditions and
material properties have been defined, SEEP2D can be used to compute the
head, flow, and Darcy velocity at each node in the mesh. The steps necessary
to run SEEP2D are described in this chapter.

5.2 Files
The files associated with SEEP2D analysis can be divided into two categories:
input files and output files:

5.2.1 Input Files

The input files to SEEP2D consist of two files: a super file and the SEEP2D
input file. The super file is a short text file that contains the names of the input
file and the output files. The SEEP2D input file contains the mesh, boundary
conditions, and model parameters.

5.2.2 Output Files

The output from SEEP2D consists of three files: a printed output file, a
geometry file, and a data set file. The printed output file is a text file
containing a summary of the input data and listing of the computed solution.
The geometry file contains the modified mesh and is only output if the
5-2 SEEP2D Primer

problem is unconfined and the deformed mesh option is selected. This file be
read into GMS prior to reading and viewing the solution file. The data set file
is a special file used by GMS for post-processing. It contains the total head,
the pressure head, the Darcy velocity, and flow potential values (for
computing flow lines).

5.3 Running SEEP2D

SEEP2D can be executed two ways: directly from the GMS menu or from the
command line. When executing from the command line, the user is prompted
to input the name of the super file. SEEP2D then computes a solution and
writes the solution to the output files specified in the super file.

5.4 Trouble Shooting

Some of the common problems which can occur while running SEEP2D, along
with possible remedies are given below.

5.4.1 Floating Point Error

If a hydraulic conductivity for one of the material properties referenced by an
element is zero then a floating point divide by zero can occur. Be sure to
check all the material properties being referenced to make sure that valid
hydraulic conductivities are given.

5.4.2 Maximum Number of Nodes or Elements Exceeded

While GMS will allow you to create large meshes, the node and element arrays
in SEEP2D have fixed dimensions. By default, the number of nodes and the
number of elements are both set to 2000. If your model has more nodes or
elements than this you will have to either redefine your mesh so that it has
2000 or less or resize the SEEP2D arrays. To resize the arrays you will need a
FORTRAN compiler. The dimensions can be changed by editing the
SEEP.INC file found in the source directory. Change the parameters to the
desired limits (be sure to change MXBNDW proportional to the amount
MXNODES is changed). Follow the directions provided by your compiler to
recreate a SEEP2D executable file. If you do not have a FORTRAN compiler
contact GMS technical support for information regarding the acquisition of a
large array sized executable.

5.4.3 Unable to Compute Flow Lines

This error occurs when SEEP2D is unable to reverse the boundary conditions
to compute flow lines. However, all other results are accurate and valid.
 Running SEEP2D 5-3

5.4.4 Mesh is Deformed when Modeling a Confined Problem

Check to be sure that you have accounted for the elevation of the node when
assigning the boundary condition. If the constant head boundary condition
assigned to a node is smaller than the elevation of the node, SEEP2D will
attempt to deform the mesh in that region.

5.4.5 Maximum Iterations Reached

When solving unconfined problems, SEEP2D deforms the boundary of the
mesh to match the phreatic surface. The boundary is moved in an iterative
fashion. The progress of the iteration and a convergence parameter are printed
to the screen. In some cases, the iteration does not converge after the
maximum number of iterations (30) has been reached. In such cases, you
should review the assigned boundary conditions to ensure that they are correct.
You may also wish to import the computed solution to GMS. Often, the
solution can give clues to where the mesh deformation process is having
difficulty. Sometimes the problem can often be fixed by increasing the density
of the mesh in the problem area.

5.4.6 Problem Unexpectedly Aborts, or Takes an Unusual Amount

of Time
Be sure to check the band width before saving the file. If the model has a high
band width try renumbering again, or try renumbering using a different node
string.
INDEX

hydraulic conductivities ..................................... 3-2

A
hydraulic conductivity................................. 2-7, 2-9
Analysis Parameters............................................4-1
I
anisotropic ..........................................................4-6
Applications........................................................1-1 Impermeable Flow Barriers................................ 2-5
axisymmetric.......................................................2-8 Internal Drains.................................................... 2-6
Axisymmetric flow .............................................4-2
L
B
Laplace equation ................................................ 1-2
Basic Element Types ..........................................3-2 linear front.......................................................... 4-3
Boundary Conditions...................................1-3, 2-1
M
Brigham Young University Engineering Computer
Graphics Laboratory.......................................1-1 material ID ......................................................... 3-2
Material Properties............................................. 4-5
C
Maximum Iterations Reached............................. 5-3
confined aquifers ................................................2-9 Maximum Number of Nodes or Elements Exceeded
Confined flow.....................................................4-2 ....................................................................... 5-2
Confined Flow Problems ....................................2-2 memory requirements......................................... 3-3
Confined vs. unconfined .....................................4-2 Mesh Construction ............................................. 1-3
constant head boundary conditions.....................2-3 Mesh Construction ............................................. 3-2
constant head boundary conditions.....................2-4 Mesh Construction Guidelines ........................... 3-1
Constant Head Boundary Conditions..................2-2 Mesh Density ..................................................... 3-4
Cores of Dams ....................................................2-7 Mesh is Deformed when Modeling a Confined
Problem.......................................................... 5-3
D
Model Conceptualization ................................... 2-1
Datum .................................................................4-1 Model Parameters .............................................. 4-1
deforming mesh ..................................................4-2 Modeling Process............................................... 1-2
Dupuit problem...................................................2-4
N
E
no flow boundary conditions.............................. 2-3
Exit Face Boundary Conditions..........................2-2 No Flow Boundary Conditions........................... 2-1
nodal half band width......................................... 3-3
F
P
Files ....................................................................5-1
Floating Point Error............................................5-2 phreatic surface .................................................. 2-7
Flow Lines ..........................................................4-5 Plan or Aerial View Models............................... 2-9
Flow Rate Boundary Conditions.........................2-2 Plane flow .......................................................... 4-2
Flow to a Well ....................................................2-8 Post-Processing .................................................. 1-3
Flux Boundary Conditions..................................2-2 Problem Unexpectedly Aborts, or Takes an
frontal function ...................................................4-3 Unusual Amount of Time .............................. 5-3
G Q
Governing Equation............................................1-2 Quadratic elements............................................. 3-2
grout curtains ......................................................2-5 quadrilaterals...................................................... 3-2
H R
Heterogeneous models........................................4-6 relative conductivity........................................... 4-3
i-2 SEEP2D Primer

Renumbering.......................................................3-2 unconfined aquifer.............................................. 2-8

Running SEEP2D ....................................... 5-1, 5-2 unconfined aquifers ............................................ 2-9
Unconfined flow................................................. 4-2
S
Unconfined Flow Problems................................ 2-4
Saturated/unsaturated flow..................................4-2 Unconfined vs. confined..................................... 4-2
SEEP2D..............................................................1-3 Unit Weight of Water ......................................... 4-2
sheet pile.............................................................2-6 United States Army Engineer Waterways
Sheet piles...........................................................2-5 Experiment Station......................................... 1-1
solution time .......................................................3-3 Unsaturated flow ................................................ 4-2
T V
Title ....................................................................4-1 Van Genuchten model ........................................ 4-4
triangles ..............................................................3-2
W
Trouble Shooting ................................................5-2
Type of Flow ......................................................4-2 Waterways Experiment Station .......................... 1-1
well..................................................................... 2-8
U
Unable to compute flow lines .............................5-2