ISBN 0-7727-7010-7

UNIVERSITY OF TORONTO PUBLICATION 80-06

Department of Civil Engineering

MINE TAILINGS DISPOSAL:

II. HYDROLOGIC EVALUATION OF DISPOSAL SITES

P. A. Gonzales

4
B. J . Adams

JUNE 1980
f&A
£ *.*><£
 ISBN 0-7727-7010-7
PUBLICATION 80-06

MINE TAILINGS DISPOSAL:

II. HYDROLOGIC EVALUATION OF DISPOSAL SITES

P. A. Gonzales
B. J. Adams

JUNE 1980
 MINE TAILINGS DISPOSAL:

I I . HYOROLOGIC EVALUATION OF DISPOSAL SITES

P. A. Gonzalez
B. J. Adams

Department of Civil Engineering

University of Toronto

Toronto, Canada

M5S 1A4

JUNE 1980
 MINE TAILINGS DISPOSAL:
II. HYDROLOGIC EVALUATION OF DISPOSAL SITES

P. A. Gonzalez and B. J . Adams

Department of Civil Engineering, University of Toronto
Toronto, Ontario, Canada, M5S 1A4

ABSTRACT

The hydrologic evaluation of mine tailings disposal sites after they are
abandoned is considered*in relation to their potential environmental
impact on a long term basis- There is a direct relation between the
amounts and types of water leaving a disposal site and the severity of
the potential damage to the environment. The evaluation of the relative
distribution of the precipitation reaching the ground into evaporation,
runoff and infiltration is obtained for a selected site and type of
tailings material whose characteristics and physical properties were
determined in the soils laboratory. A conceptual model of the hydrologic
processes involved and the corresponding mathematical model were developed
to simulate the physical system. A computer program was written to solve
the set of equations forming the mathematical model, considering the
physical properties of the tailings and the rainfall data selected. The
results indicate that the relative distribution of the precipitation depends
on the surface and upper layer of the tailings and that the position of the
groundwater table is governed by the flow through the bottom of the profile
considered. The slope of the surface of the mass of tailings was found to
be one of the principal factors affecting the relative distribution of
precipitation and, therefore, the potential pollution of the environment.
 MINE TAILINGS DISPOSAL:
II. HYOROLOGIC EVALUATION OF DISPOSAL SITES

P. A. Gonzalez and B. J. Adams

Department of Civil Engineering, University of Toronto
Toronto, Ontario, Canada, M5S 1A4

INTRODUCTION
With increasing population growth and increasing material requirements
of modern Tife, questions concerning the adequacy of the methods of
disposal of wastes, refuse and tailing materials have arisen. Mineral
tailing disposal sites, including landfill ing or tailing ponds are
usually selected and designed considering the abandonment procedure and
sometimes also the reclamation procedure. The evaluation of the long
term effects on the environment of the abandonment procedure can be
performed to a large extent by studying the behaviour of the whole
system of disposal in time and its response to hydrologic variables.
Much of the environmental impact of mine tailing dispos31 sites is related
to the amount and quality of water leaving the site. Water passing
through the tailings carries with it various dissolved and suspended
materials which are often deleterious and can have undesirable chemical
and biological characteristics. This water, which is commonly called
leachate, may become a source of contamination of surface and/or groundwaters.
The volumes of leachate leaving the site are directly related to the
severity of potential environmental pollution for a given type of tailings.

Present address: McElhanney Surveying and Engineering, Ltd.,

13160-88th Avenue, Surrey, British Columbia, Canada, V3W 3K3.

- 2-
The distribution of the total amount of precipitation over the area of
the site into runoff, infiltration, and evapotranspiration has a marked
effect on the volumes of leachate and seepage leaving the site. It is,
therefore, important to analyze and quantify the modes and amounts of
water leaving the site.

The objectives of this paper are, consequently, to evaluate the long term
hydrologic effects of mine tailing disposal sites. In order to evaluate
the quantities involved in each of these hydrologic processes, it is
necessary to analyze the effect of different factors such as the slope of
the surface of the site, the type of cover, the physical properties of
the tailings material and to analyze how they are affected by a sequence
of precipitation and dry periods during an appropriate period of time.
Furthermore, there exist interrelations between hydrologic processes such
as runoff, evaporation and infiltration with some of the properties of
the soil such as permeability, moisture content and hydraulic conductivity.
These interrelations are time dependent and therefore, for the evaluation
of the process as a whole, it is necessary to simulate the hydrologic
phenomena over a period of time. The problem can be established as the
study of the vertical movement of water in a porous media subjected to
certain boundary conditions.

- 3-
 MATHEMATICAL MODEL
In hydrology, as in other fields of applied physics, one of the most
powerful methods of investigating complex phenomena is conceptual
modelling. The hydrologic processes of infiltration, evaporation, flow
of water through the soil and its subsequent movements in the mass of
soil are interrelated. The variables governing these processes are in
general interrelated and time-dependent; therefore, the evaluation of
the processes, including the quantification of the variables, can be done
by modelling the combined hydrologic processes taking place in the system.
The characteristics of the processes, with variables distributed and
interrelated, lead to the concept of a non-linear distributed-parameter
system model. The conceptual model considers the physical continuity of
the hydrologic system from the infiltration-evaporation process taking
place in the surface of the soil, to the final movement of water in the
saturated region, passing through the unsaturated region.

The conceptual model can be established in general by the study of the

vertical movement of water in a porous medium subjected to certain boundary
conditions in the surface and bottom of the profile. These boundary con-
ditions express the different processes such as infiltration, evaporation,
ponded water in the surface of the soil and the upward or downward movement
of water in the basal boundary.

The mathematical model results from the application of physical principles

to the hydrologic processes involved in the conceptual model. The flow of
water through both saturated and unsaturated regions is a continuous system
governed by the same general principle of motion.

- 4-
This principle of motion may be applied to the flow system combined with the
principle of mass conservation and the thermodynamic equilibrium to obtain
the mathematical equations required by the model. The continuous system
of flow indicates the existence of a variable which governs the flow in
both regions. This variable is the total potential, which is defined as the
energy per unit of mass of water. This energy of the water may be divided
into potential energy and kinetic energy. The potential energy, neglecting
the temperature gradient, corresponds to the gravitational potential energy
plus the pressure head. The kinetic energy is negligible for the small
velocities of the flow of water through the soil, especially in this case
where the velocities are in the range of a few meters per year.

According to infiltration theory, the equation governing the verticle move-

ment of water in an homogeneous porous medium for saturated and unsaturated
unsteady conditions, is
m c<f>!£.&
In this equation, the specific moisture capacity, COn), and the hydraulic
conductivity, K(V), are single-valued functions of pressure head, f, which
varies through depth, z, and time, t. This equation leads to a simpler
expression for the saturated flow case where the specific moisture capacity
is zero and the hydraulic conductivity is constant for homogeneous soil and
equal to the saturated permeability, K s a t , and may be written

Hence

which is known as the Laplace equation and is valid in the saturated region
of the system.

- 5-
Equations [1] and [2] are the basic equations that constitute the
mathematical model, representing the flow of water through the soil in
both unsaturated and saturated regions, respectively. The mathematical
model represents the vertical movement of water in a soil profile
selected in the mine tailings site, assuming that tha flow takes place
through a vertical column of soil with unit cross-section area. This
column is divided into 19 sections by 20 nodes separated 10 cm each.
The nodes are numbered from the bottom to the surface of the soil. The
bottom node or bottom boundary is located in the saturated region of the
soil. The upper node is located coincident with the surface of the soil
and corresponds to the surface boundary of the model. The vertical
co-ordinate, z, is oriented positive downwards. A schematic diagram of
the model is shown in Figure 1.
BOUNDARY CONDITIONS
The mathematical model requires the relations governing the flow through
the soil and the mathematical relations representing the processes that
take place at both the upper and lowar boundaries. In order to obtain
these relations, it is necessary to determine the physical phenomena
governing the flow of water in and out of the profile, through the
corresponding boundaries.
Surface Boundary
The physical processes occurring in the surface and upper layer of the
soil include precipitation, evapotranspiration, infiltration, runoff or
overland flow and snow or freezing conditions. All of these processes
affect directly or indirectly the flow of water through the soil. Both
the input and the output of water to the model can be described mainly by
the processes of infiltration and evaporation from soil, respectively.

- 6 -
The first includes the effects of precipitation, evaporation from water,
overland flow and ponded water, and the second includes the effects of
evapotranspiration. Both processes, infiltration and evaporation from
soil are closely related to soil surface characteristics, soil properties
and moisture content of the soil.

The infiltration process is one aspect of the general theory of fluid

movement in a porous medium. The mathematical relation governing
infiltration may be expressed in terms of the same variables affecting the
flow through the soil with the addition of restrictions arising from the
application of boundary conditions.

The infiltration theory has been developed from Darcy's Law and a
continuity statement which have lead to theoretical equations relating
the infiltration rate to soil parameters such as hydraulic conductivity,
soil sorptivity, the initial moisture content of the profile, the saturated
moisture content, and a series of factors which are functions of these
parameters. The empirical equations, derived from field observations and
laboratory research, give values of the infiltration rate close to the
observed values and they may be used to evaluate infiltration with the same
degree of approximation as the other variables; furthermore, the theoretical
equations lead to expressions similar to those of the empirical equations,
when higher degree terms of the theoretical solutions are neglected
(Philip 1957).

In order to describe the process of infiltration, two of the most common

empirical equations are considered:
Horton's (1940) equation
[4] f = f c + (f Q - f.) e" k t

- 7-
and Hoi tan's (1961) equation
[5] f = a (S-F) n + f c

where f is the infiltration rate at time t, f is the final constant

infiltration rate, f is the initial infiltration capacity, S is the
o
storage potential of the soil expressed as a volumetric difference
between saturation and wilting point, F is the. accumulated infiltration
volume and k, a, and n are constants for a particular soil in a given
condition. These two equations are based on the reduction of the
infiltration rate as a result of the increasing moisture content of a
control layer in the upper part of the soil. In the Holtan eauation [5],
the above idea is expressed in the difference between the storage potential,
S, as a limit value of the infiltration, and the accumulated infiltration,
F. In the Horton equation [4], the idea of proportionality is implicit.
The infiltration rate difference ( f - O decreases with time. •

The evaluation of infiltration for different relative values of rainfall

intensity may be performed by applying both equations. Figure 2(a) shows
the shape of tie infiltration curve and the rainfall intensities, i, for
different possibilities, which are, in the first time interval i<f, in the
second time interval i>f and, in the third time interval, the infiltration
capacity is greater than the rainfall intensity at the beginning of the
interval (Z-M:*) but smaller at the end (t*->-3). This situation and the
assumption of the Horton equation that is valid when there is available
water, lead to the combined use of the Horton and Holtan equations. The
first is used directly in the cases where i>f and is complemented by the
Holtan equation in the cases where i<f and when moisture variations in
the upper layer of the soil modify the initial value, f , of the Horton
equation.

- 8-
Using the concept of storage potential or volumetric difference between
the moisture content of the soil at saturation and at the wilting point
(15 bar), it is possible to determine, for a given soil, the effects of
increasing mass infiltration, which reduces the infiltration capacity,
and the recovery of the infiltration capacity when the moisture content
of the upper layer of the soil is reduced by the flow of water toward
deeper layers or by evaporation from soil. Both processes have similar
effects on the recovery of the infiltration capacity.

Figure 2(b) shows the effects of wet and dry periods. The infiltration
capacity, originally at point A, which corresponds to a dry condition of
soil, decreases with increasing moisture content in the soil to reach the
point B where the rainfall stops and the moisture content of the soil
decreases due to the combined effects of gravity flow toward deeper layers
and evaporation to the atmosphere. This decrease in moisture content of
the soil increases the infiltration capacity to point C where a new period
of rainfall starts, decreasing the infiltration capacity to point 0, where
the rainfall stops. Again, the infiltration capacity increases at a
decreasing rate. In order to evaluate the recovery of the infiltration
capacity during the dry period between two consecutive rainfall events,
the model determines the variation of moisture content in the upper layer
of the soil, and, using the Holtan equation, evaluates the recovery of
infiltration capacity.

To determine the variation of moisture content of the upper layer of the

soil, evaporation, infiltration and percolation are considered. The
evaporation is evaluated from atmospheric and hydrometeorologic conditions
as potential evapotranspiration, which is in general higher than the
hydraulic conductivity values of the fine tailing materials in normal

- 9-
abandonment conditions; therefore, the evaporation process can be considered
as a constant value for the total period of simulation. This value represents
a potential evaporation. The actual evaporation is calculated separately
for ponded conditions and for evaporation from soil. The percolation is
obtained from the application of the model to the complete system of flow.

To complete the evaluation of the processes occurring in the surface of the

soil, the model must consider the effects of the slope, roughness and size
of the surface of the tailing disposal site. This can be accomplished by
analyzing the overland flow process and calculating the average depth of
flow corresponding to a given rainfall intensity, slope, roughness
characteristic and size of the tailing surface. The time period involved
in these calculations can be selected considering the availability of
rainfall intensity data, the general accuracy of the model and the
sensitivity of the variables with time. An appropriate time step for this
surface process is judged to be one hour, which is a relatively short time
step for xhe deeper region processes, but gives acceptable values for the
more rapid variations of the surface processes.

The equation relating the velocity of flow on a surface (sheet flow) to the
slope, roughness, and depth of flow is the Manning equation. ' Considering
the size of an average mine tailing surface and the selected time step,
the average depth of flow can be evaluated with the slope as a parameter,
for a given roughness coefficient. In this work, the slopes considered
range between 0.1% and 6% corresponding to two different approaches in mine
tailings disposal site design: a traditional approach, with slopes near 0.1%
after abandonment and a more recent approach with final slopes of about 6%
(Robinsky 1975}. Both values were considered as lower and upper limits
for the range of slope variation.
- 10 -
A balance of water can be applied to the surface of the soil for each
time step considering the rainfall as an input and the excess, being
overland flow, evaporation and i n f i l t r a t i o n , as outputs. Assuming that
there is no evaporation from the soil surface during the rainfall event,
the average depth of precipitation during the time step is available
for infiltration and overland flow. During the dry periods, the ponded
water is available for evaporation and infiltration. The amount of
evaporation and infiltration depends on the potential evaporation and
infiltration capacity.

The infiltration is calculated from its final value of the previous time
step and the rainfall intensity value of the time step being analyzed.
The i n i t i a l value of the infiltration capacity corresponds to the final
value of the previous step or the value given by the Hoi tan equation,
which considers the variation in moisture content in the soil. In general,
there are three cases of relative values for rainfall intensity and
infiltration rate (see Figure 2(a)):
(A) i<f (Time Q * 1)
(B) 1>f (Time 1 * 2)
(C) f varies from f>1 to f<i (Time 2 - 3 )
The mass infiltration in these three cases can be evaluated as follows:
Case (A), f > i , therefore the total infiltrated water is
E6] F = Ct
o-i h i• V
Casa (B), f<i, assuming that the portion of the f-curve in
the interval is calculated from the previous
condition, then
[7] F,_, - ( 2f(t)dt
'*1

- 11 -
[8]
•ft + Cf -* ) e" k t ]
° c
dt

f -f. -let, -kt.

[9]

Case (C)» f varies from f>i to f < i , passing through a

value f = i at a time t*. Assuming the
validity of the Horton equation in the
interval ( t 2 - t 3 ) , determine t * as follows:

[10]

hence
[11]
0 C,

Therefore

[12] F
2-t* - t-

and
[13] F
-kt* - e"kt3
t*-3 " f c
hence
[14] F = F + F
2-3 2-t* t*-3
or
[15] r
2-3 • ,

The mass infiltration calculated and the percolation during the time step
considered are used to evaluate the i n i t i a l infiltration capacity for the
next time step, by applying the Holtan equation. The relation between
infiltration rates for each hour and the i n i t i a l rate are obtained from
typical infiltration curves.
The mathematical relation that considers the effects of infiltration and
evaporation may be obtained by applying the definition of specific
moisture capacity.
[161 C (¥) = ||

where 9 is the moisture content of the soil, to the upper layer of soil
where the infiltration or evaporation takes place. Then from equation [1]:

11
fl 3t

Considering the depth Az of the upper layer and for a time step equal to
unity

[18] KW [£•+ 1) = Q

Therefore

or

This equation relates the two consecutive values of pressure head in one
layer (JJ) and in the layer above (JJ-1) to the respective depths, and
the evaporation or infiltration Q (Q>0 represents infiltration and
Q<0 represents evaporation).
Bottom Boundary
According to the model, the bottom boundary is located in the saturated
region where the flow is governed by the Laplace equation. The following
mathematical relation representing the bottom boundary condition can be
obtained from equation [2] in which the hydraulic conductivity is a
constant equal to the saturated permeability
 K
sat T
Therefore

where R is the flow through the bottom boundary. The flow may be from the
model system downward, R>0, or from the groundwater system upward entering
the model system through the bottom boundary, R<0. These flows can be
considered as recharge cr discharge of the groundwater system (Freeze, 1967).
METHOD OF SOLUTION
The mathematical model described in the previous section includes two
differential equations of flow and two equations of boundary conditions.
The partial differential equation governing the unsaturated flow is of the
non-linear type (Equation [1]) in which the coefficients C(¥) and K(f) are
dependent on the variable. The solution of equations of this type can be
performed by numerical methods using a high-speed digital computer.

The method of solution applied is the finite difference method, in which

the partial differential equations and the corresponding boundary condition
equations are replaced with equivalent algebraic equations. This replace-
ment is made of discrete points in the continuous domain substituting the
derivatives with finite difference expressions connecting the values of the
variables considered in adjacent points. These finite difference equations
can be established in each internal point considered leading to a set of
equations relating the values of the variables through the complete profile.
This can be represented in a two dimensional plane (Z,T) in which the
ordinate is the vertical distance from the bottom of the soil profile to
the surface, and the abscissa is the time. Ffgure 3 shows these co-ordinates
and the nomenclature adopted.

- 14 -
The differential equation governing the flow in the unsaturated region
(Equation [1]) is replaced v;ith a finite difference equation as
follows:

[23] CI

in which the values of tne indexes of ±h indicates the points half way
between the nodes, therefore the point n+H is in between the nodes n and
n+1. Rearranging the above equation

un+l
AT J

This difference equation includes the values of the functions C(¥) and
at intermediate points of the grid, such as

The values of the variable in the intermediate points are determined by

extrapolation from previously known values as follows:

[25] _ 3.
" 4

[261

and

- 15 -
[27] VJJ** = 2 ^ - £ (VJ + TJJ" 1 }

The differential equation governing the flow in the saturated region

(Equation £2]) is replaced with a finite difference equation as follows:

[28] - < £ • Z^ - <£ - 0

and the boundary conditions (Equation; [20] and [22]) are replaced with
the corresponding finite difference equations:

[29] f*1 - l"**

r = S£~. CSurface Boundary)
JJ-l
and
[30] I'll - ll', = 4-5- (Bottom Boundary)
L l
Sat
The value of the pressure head ^ J J _ ^ is obtained by extrapolation.

The system of simultaneous equations formed by the finite difference

equations may be written as follows for each internal node:

C33] D
J

C
J

and
- 16 -
[35] Bj = Aj

In the saturated region, the finite difference equation can be written

as follows:

[36] - vJ+J + 2 v j + 1 - Vn+_\ = 0

The above equation represents a linear variation of pressure head with depth
in the saturated zone.

The finite difference equations for the boundary conditions are Equations
[29] and [30]. The solution of the system of simultaneous equations may be
expressed as ¥(Z,T), in which the variables Z and T are discrete. The
system of simultaneous equations constitute a tridiagonal system and can
be solved by elimination with an appropriate algorithm using a digital
computer (Conte 1972).
EXPERIMENTAL DATA
In order to solve the equations governing the flow through the profile,
it is necessary to evaluate the hydraulic conductivity values and tne
specific moisture capacity values for each node and intermediate points
required by the finite difference equations, and the pressure versus
moisture content curve, which is used to express the pressure head solution
to a moisture content solution of the system.

These functional relations between moisture content and hydraulic conduc-

tivity versus pressure head can be obtained only experimentally. The
specific moisture capacity curve can be derived from-the moisture content
versus pressure curve by geometric or numerical methods. Other properties
of the soil that are required directly or indirectly to solve the problem
are the saturated permeability, K s a t ; the air pressure entry of the soil,
which is the required pressure head or suction to allow air entry the soil
surface, and the initial moisture conditions of the profile (Gonzalez
and Adams 1980).

The soil surface characteristics such as roughness coefficient, slope, and

amount and type of vegetation are necessary to determine the surface
processes which affect the boundary conditions. Rainfall data and
potential evaporation are also required to solve the system. The
precipitation data used is rainfall data from the Mai ton Weather
Station of the Atmospheric Environmental Service of Environment Canada,
which was recorded on an hourly basis and measured in hundredths of an
inch (0.254 mm).

- 18 -
 COMPUTER PROGRAM
In order to apply the mathematical model to simulate the hydro!ogic
processes taking place in the mine tailing disposal site, the set of
mathematical equations defining the model must be solved with the use
of a high speed digital computer for which a computer program is
required. The computer program was written in Fortran IV language for
the IBM 370/165 computer of the University of Toronto Computer - Centre.

The schematic flow chart presented in Figure 4 indicates the general

features of the main program. Beginning with the reading of all the
rainfall data, functional relations for both wetting and drying processes,
the parameters defining the surface characteristics, the initial values
of the variables, and the parameters defining the profile, the program
determines the depth of depression storage, above which the surface
water is considered to run off. The initial moisture content of the
soil profile is then calculated.

The developed mathematical model considers the hydraulic conductivity,

the specific moisture capacity and the moisture content as single-value
functions of pressure head which is true only for separated processes of
wetting or drying due to the hysteresis involved; therefore, in order to
use the correct curve corresponding to the process that is taking place in
a particular node, the program determines the type of process occurring
and selects the proper functional relation for the unsaturated case of
flow. Thus, the program determines the type of flow (saturated or
unsaturated) and the type of process (wetting or drying) for all nodes of
the profile in each time step.

- 19 -
At the surface boundary, the program analyzes the processes of rainfall,
potential evaporation, infiltration and runoff on an hourly basis
determining the total available water on the surface, the evaporation
from water, the evaporation from soil, the runoff and the infiltration.
The water balance on the surface is performed on an hourly basis and the
values of che variables are determined for 6 hour periods. The
coefficfents of the surface boundary condition equation are calculated
from the above balance for time steps of 6 hours. After the matrix of
coefficients for the mathematical equations is completed, the program
solves the simultaneous equations by means of a subroutine.

The main program uses a subroutine named TRID, which uses the method of
elimination for tridiagonal systems. The solution of the system is a
vector whose elements are the pressure head in each node of the profile.
The subroutine INTERP is used to determine the values of the functional
relations for the values of pressure head required and to interpolate the
values of moisture content. The subroutine INFIL analyzes the rate of
infiltration corresponding to the surface conditions and the moisture
content of the upper layer of the soil integrating the infiltration curve
with the limitations of availability of water and moisture content of the
soil, determining the amount of infiltration, ponded water or evaporation.
T
he subroutine READY gives the hourly rainfall data from the original set
of data which was read as a set of non zero values with the corresponding
location.

APPLICATION
The model was applied with the data collected and the functional relations
determined in the laboratory to simulate the hydrologic processes in a
tailing disposal site with a 2500 ft (762 m) radius and a surface roughness
coefficient (Manning n) of 0.025 which represents average conditions.
For a simulation period of 37.5 days, calculations were performed on an
hourly basis for the surface and upper soil layer processes and on a
6 hour basis for the rest of the soil profile. This somewhat arbitrary
separation of the time increment was made because the physical process
of infiltration which takes place on tlie surface and upper layer of ths
soil is more sensitive to time variation and rainfall intensity variations
than the lower parts of the profile, and because there is a substantial
reduction in the computer time required.

The functional relations were entered into the computer as tables of

values for both wetting and drying processes. The evaporation was con-
sidered to be the yearly average for the Southern Ontario region as
25 in/yr (635 mm/yr). This value is well above the maximum rate at which
a soil of the type of the tailings can sustain even under initially
saturated conditions; therefore, it was unnecessary to provide the program
with a function representing variable evaporation. The flow through the
bottom of the profile considered was selected to be in a range with a maximum
value smaller than the saturated permeability, which is in direct relation-
ship with the groundwater flow.

Mention is made regarding the slope of the tailings surface because of the
two different approaches in the disposal site design; the traditional
approach, with slopes near 0.1%, and a newer approach with final slopes
of about 6%, after abandonment. Both values were considered as the lower
and upper limits for the range of slope variations.

Some remarks are also made concerning limitations of the program. According
to the theory of linear differential equations, the general solution of a
J th order differential equation will contain J arbitrary constants. These
constants can be evaluated from the N boundary conditions in terms of the
dependent variables and their derivatives at certain points or boundaries.
The numerical solution of the boundary value problems can be accomplished
either by iterative methods or can be reduced to the solution of a set of
N simultaneous linear equations in N unknowns. In this case, the computer
program was written to solve a system of N simultaneous equations in which
the number of equations depends on the size of the depth stsp. The
selection of the size of the step is a compromise between the accuracy of
the solution and the computer time used. This balance can be supplemented
by including the limitations to the accuracy of the solution introduced by
the several assumptions made in the development of the mathematical model.
The linear interpolation of values of the functional relations and the
laboratory determination of them, make it appropriate for this case to
select the size of the depth steps of about 20 cm and time steps of 1 hour
for the surface processes and 6 hours for the rest of the profile.
RESULTS
The mathematical model was run to simulate the hydrologic processes
occurring in a mine tailing disposal site under a given pattern of hydro-
meteorological conditions and to evaluate the effects of various parameters
on the system of flow through both saturated and unsaturated conditions,
and on the relative distribution of precipitation into runoff, infiltration
and evapotranspiration. Several preliminary runs were made to determine
the appropriate depth of the profile to analyze the corresponding depth
steps, the time increment, the length of simulation and the sensitivity
of the model to various parameters. The initial conditions for the sim-
ulation corresponds to a saturated soil, which is the situation most likely
to occur when the mine tailing site is abandoned. The preliminary runs

- 22 -
demonstrated that simulation periods of slightly over a month yielded
consistent results. Longer simulation periods did not materially change
the relations extracted from the results.

The relation between infiltration and the slope of the surface of the
site is illustrated in Figure 5. Infiltration increases with decreasing
slope, from its minimum for the 6% slope to its maximum at the 0.1% slope.
The flow through the bottom of the profile, Q, has a negligible effect on
this relation. The infiltration process is governed by surface and upper
layer conditions.

The relation between runoff.and slope of the surface is presented in

Figure 5. It was found that the runoff increases with increasing slope
from the zero value at a slope near 0.1% to the maximum at the 6% slope,
which is the upper limit of the slope range considered. The values of the
flow through the bottom have a negligible effect on this relation. The
effects of slope on both infiltration and runoff are presented together
in order to visualize that these processes are strongly related.

The relation between the evaporation from ponded water, which takes place
when there is available water on the surface of the soil, and the slope, is
shown in Figure 6. The evaporation from ponded water was found to increase
with decreasing slope, due to the increasing evaporation opportunity when
slope decreases (the combined effects of infiltration, which increases,
and runoff, which decreases, with decreasing slope).

The relation between the evaporation from soil, which takes place when
there is no ponded water on the surface of the soil, and the slope is shown,
for four different values of flow through the bottom, in Figure 6. The
values of Q are those indicated. It is clear that the slope of
- 23 -
the- soil surface has very ltttle effect on the evaporation from soil which
decreases with increasing flow through the bottom. The flow through the
bottom, as it will be shown later, governs the position of the groundwater
table. The variations in the position of the groundwater table affect the
hydraulic conductivity of the soil in the unsaturated region and in the
surface of the soil and correspondingly influence the upward movement of
water towards the evaporating region.

The relation between moisture variation, whic,h is the decrease in water

content in the soil profile, and the slope is shown in Figure 7 for the
values of Q indicated previously. It was found that the moisture variation
increases with the slope. This effect is small compared to the effect of
the flow through the bottom on the moisture variation.

The relation between the position of the groundwater table and the flow
through the bottom is shown in Figure 8 for different values of the slope
of the surface. It was found that the depth of the groundwater table
increases with increasing flow through the bottom for all the considered
slope values. In other words, the position of the groundwater table is
governed by the flow through the bottom. The curves for different slopes
are markedly parallel and close to each other.

Based on the results of the hydrological evaluation, a long-term water

balance for a disposal site can be made for different values of the
surface slope. Taking the general pattern of rainfall as the average
conditions for a period of time, for example 10 years, the amounts of
infiltration and runoff may be evaluated for that period considering their
relative distribution to be constant.

The relative distribution of the infiltration and runoff is shown in

Table 1 for various values of slope.
 SUMMARY AND CONCLUSIONS
The environmental impact of mine tailing disposal sites after they are
abandoned is closely related to the amounts of water leaving the site
and to their relative distribution. According to the results obtained
from the above simulation model, the relative distribution of the water
leaving the site depends on the distribution of precipitation reaching
the ground, which can be considered the main source of water to the system.
The following are some general but quantitative observations.

The precipitation reaching the ground may be divided into runoff,

infiltration and evaporation. The relative distribution of each of these
depends on the characteristics of the surface of the site and on the
properties of the tailings, especially the upper layer, thus, for surface
slopes greater than 1%, the amount of total rainfall which infiltrates is
approximately 50% or less. For a surface slope of 6%, the amount of
infiltrated water is only around 25% of the total precipitation. The
slope of the surface also affects the evaporation from water. Increasing
the slope decreases the amount of evaporation from water, thus, for a
surface slope of 0.1% the evaporation from ponded water on the soil is of
the order of 1.37 cm for the period of time considered, and for a surface
slope of 6% the resulting evaporation from ponded water is 0.37 cm, for
the same period of time.

The slope of the surface has a small influence on the evaporation from
soil. The values obtained are 4.9 cm of evaporation from soil for a
slope surface of 0.1%,. and 5.1 cm of evaporation from soil for a surface
slope of 6%. The position of the groundwater table has a small effect on

- 25 -
this relative distribution. This effect decreases with deeper levels of
the groundwater table. The upward movement of water from the saturated
region to the aeration zone becomes extremely slow, limiting the evaporation
rates. After dry periods, in which the groundwater table is likely to
descend, the evaporation from soil is small or negligible. Runoff and
infiltration are the most important factors affecting the potential
pollution of the environment, the latter being most difficult to analyze.

The general pattern of flow toward the groundwater region and the limitations
of the upward movement of water and subsequent evaporation, make it more
important, from the point of view of environmental protection, to control
the infiltration by increasing the runoff, which is easier to collect and
treat, if necessary, before delivery to receiving water bodies.

According to the results of the simulations, the slope of the surface of

the site is one of the most important factors governing the relative
distribution of the precipitation reaching the ground. Therefore, the
infiltration may be controlled to an extent by increasing the slope of the
surface of the site. In this case, the runoff increases from a value of around
zero for a surface slope of 0.1% to 3.4 cm for a surface slope of 6%, and
the infiltration decreases from a value of 3.6 cm for a surface slope of
0.1% to 1.2 cm of infiltrated water for a surface slope of 6%. These two
combined effects result in a general net effect of diminishing the potential
pollution of the environment. Thus, increasing the slope of the site
surface from 0.1% to 6% results in an approximate two-thirds reduction of
the potential pollution of the groundwater system.

- 26 -
The relations obtained were final values for simulations of the behaviour
of the flow system through the profile of soil selected and for the complete
period of time analyzed. Longer simulation runs were found not to vary
substantially the relations between the variables.

- 27 -
 ACKNOWLEDGEMENTS
The work upon which this paper is based was supported in part by funds
provided by the Natural Sciences and Engineering Research Council of
Canada and the University of Toronto. The authors wish to express their
thanks to Dr. E. I. Robinsky of the Department of Civil Engineering,
University of Toronto, for his valuable comments on the manuscript.

- 28 -
 REFERENCES
Conte, S. D., and de Boor, C. 1972. Elementary numerical analysis,
McGraw-Hill Book Co., New York, N.Y.

Freeze, R. A. 1967. The continuity between groundwater flow systems

and flow in the unsaturated zone. Proc. Hydrology Symposium No. 6,
National Research Council of Canada.

Gonzalez, P. A., and Adams, B. J. 1980. Mine tailings disposal:

I. Laboratory characterization of tailings. Paper submitted for
publication to the Canadian Journal of Civil Engineering.

Holtan, H. N. 1961. A concept of infiltration estimates in watershed

engineering. U.S. Dept. of Agriculture - ARS, 41-51, 25 pp.

Horto.n, R. E. 1940. An approach toward a physical interpretation of

infiltration capacity. Proc. Soil Science Society of America,
5, pp. 399-417.

Philip, J. R. 1957. The theory of infiltration: 1. The infiltration

equation and its solution. Soil Science, 83, pp. 345-357.

Robinsky, E. I. 1975. The thickened discharge: a new approach to

tailing disposal. Canadian Mining and Metallurgical Bulletin,
December.

- 29 -
Table 1. Relative distribution of infiltration and runoff as a
percentage of total rainfall.

Slope Infiltration Runoff

(X) (X) (X)

0.1 72 0
1.0 54 25
3.0 33 55
6.0 24 68

- 30 -
Precipitation Evaporation

Evaporation from water

Surface runoff
i Depression storage

Nodes-"''
11 Infiltration
Infiltration
' t Moisture UNSATURATED ZONE
• . content

Percolation
K.C.S are functions of

Capilari ty
Ground table position
v < > 5 «Ground table
? " ' n
SATURATED ZONE

Saturated flow
K,6 are constants)
C=0 i
Lower boundry

1 I
Recharge Discharge

Figure 1. Schematic Diagram of the Model

- 31 -
 INFILTRATION CAPACITY INFILTRATION CAPACITY f
AND RAINFALL INTENSITY
o

1
o

1
CO U3

Infilt
-i

V
Vs o'
It -jj
3
V
\\
to
o
(D r t
l/i Ct» \XY^X<NNS>W: TJ
C rt
3.O
CU 3
o \\ \XW
^X/_.
\
\ 33 s;
o
O
m o o

CM cu -a
CU CU
\> II
3 CU
2 ON
O- O
m 51 \\ o+
—*»
(D CU
1 3
a. x> N^lxl 3
<-*
o*

v SxlxW
CU TO 1
«< CU
(0
CU »^- o
\\
(D -•• 3
O-
o o> a>
33 i
CU ^-
3
3 \ f •
rt
N
%
\
tn
3
!-»•
X] /
O JO
n>
\ 1
—« CU rt
\ |a>-
N1
fD rt *^.
0>
in
N
^ J+1
UJ
O
Z

TIME T
QC
111

Z J-1-
o
-AT- Nodes
_1

J-2-
N-2 N-1 N N+1

Figure 3. Two Dimensional Net Representing the

Space and Time Coordinates (ij/j represents
the Pressure Head at Node J in the
Time Step N).

- 33 -
 V

'aat -a i i r .- S ia C :-ar. S , ;
a! ici:.on aaracetars
l

Jacaraine I2aff : ;isp» or :ouncarv

:s i:,ra5a

\i cs aaranecsrs rala^ad :o i n i t i a l ,
:
sona ;i anj i
•
i

a actual avaaoration
:«:a; •si na :Sn,3r3,i,a ••acar nour!/ jas-3;

3ecsmine avaaora;'3n r-a

Oetsraine runof"

Jar;na evsa of srocass (•«•£-i eg ar srying) ; Jecermine inr;icracian

Js-ar^i-e 'aiuas ar •.sr-asoons

r\nc: ions' -slacians Oecirmine values for 5 :-,r. aer'oa

1 !
3acsrni.-i9 :a53 i ava i aols »azar
i^rraes' :'l*our! i •lalculass caa/^isiant3 of iur-aca soursary
conai:ion aquation ?ar 5 i i * . zer'za
r
vapora^on
Solve siiTiul'aneous «cuaci:ns i :r.
ci^e itaoi

MO

Firgure 4. Schematic Flow Chart of the Computer Program. i 'as

\ 3

/
\ /
V

- 34 -
 s •— Runoff

5 • —

Ui
a.
O
03 3 -

) 0.5 1.0 T.5 2.0 2.5 3.0 3.5 4.0

INFILTRATION OR RUNOFF (cm)

Figure 5. Infiltration vs. Slope and Runoff vs. Slope

- 35 -
 EVAPORATION FROM WATER cm
0,2 0.4 0,S 0,8 1.0 1.2 1.4 1.6
I
Evaporation from soil vs.slope
y
s -

38 4
Q =6.Ox 10~°cm/min
Ui Q z6.0x 10~5cm/min
Q.
Q3 r1.5 x 10" 4 cm/min
o
-1 3 Q4 r2.0x10~ 4 cm/min
CO
Evaporation
from water

2 3 4 5 6
EVAPORATION FROM SOIL cm

Figure 6. Evaporation from Water vs. Slope and Evaporation from

Soil vs. Slope

- 36 -
0 2 4 6 8 10 12 14
MOSTURE VARIATION (cm)

Figure 7. Moisture Variation vs. Slope.

- 37 -
 FLOW THROUGH BOTTOM cm/min x 1 0 ~ 6
40 80 120_ 160 200 240 280

100 -

200

Eo
X 300
t-
Q.
HI
o
400

500

60C

Figure 8. Depth of Ground Water Table vs. Flow through Bottom

Boundary.

- 38 -