Project 1.: A Mathematical Model of A Groundwater Contaminant Source
Project 1.: A Mathematical Model of A Groundwater Contaminant Source
Project 1.: A Mathematical Model of A Groundwater Contaminant Source
Contaminant plume
Groundwater flow
velocity vd
cin = 0
cout = c(t)
Source mass
m(t)
Cross-sectional area As
1
R. W. Falta, Rao, P. S.,and
Parameters and variables
N. Basu, “Assessing
relevantthetoImpacts of Partiala Mass
formulating Depletion in model
mathematical DNAPL of
Source Zones:
contaminant
I. Analytical Modeling of Source Strength Functions and Plume Response,” Journal of Contaminant Hydrology 78,
discharge from the source region are defined as follows:
4 (2005), pp. 259-280.
cross-sectional area of the source region
Darcy groundwater flow velocity10
total DNAPL mass in source region
In this project, we study a first order differential equation that describes the time-
dependent rate of dissolved contaminant discharge leaving the source zone and
entering the plume2 .
Parameters and variables relevant to formulating a mathematical model of conta-
minant discharge from the source region are defined as follows:
The equation describing the rate of DNAPL mass discharge from the source region
is
dm
= −As vd cs (t) (1)
dt
whereas an algebraic relationship between cs (t) and m(t) is postulated in the form
of a power law, γ
cs (t) m(t)
= , (2)
c0 m0
in which γ > 0 is empirically determined. Combining Eqs. (1) and (2) (Problem
1) yields a first order differential equation
dm
= −αmγ (3)
dt
that models the dissolution of DNAPL into the groundwater flowing through the
source region.
2
The output of this model can then be used as input into another mathematical model that, in turn, describes
the processes of advection, adsorption, dispersion, and degradation of contaminant within the plume.
3
In porous media flow, the Darcy flow velocity vd is defined by vd = Q/A, where A is a cross-sectional area
available for flow and Q is the volumetric flow rate (volume/time) through A.
(15
dy
Separating the variables in Eq. (15) leads to
Projects 113
PROBLEMS
Project 2 PROBLEMS dy (16
1. Derive Eq. (3) from Eqs. (1) and (2) and show that 5. Effects of Partial Source Remediation.
𝛾
𝛼 = 𝑣d As c0 ∕m0 . We solve Eq. (16) by integrating
(a) both sides
Assume thattoaarrive
sourceatremediation
the implicitly defined
process solution
results in a
2. Additional processes due to biotic and abiotic degradation 90% reduction in the initial amount of DNAPL mass in the
contributing to source decay can be accounted for by adding source region. Repeat=Problem
ln ln 4 with m0 and c0 in Eq. (2) re- (17
a decay term to (3) that is proportional to m(t), placed by m1 = 0.1 m0 and c1 = (0.1)𝛾 c0 , respectively. Com-
m′ (t) = −𝛼mThe𝛾 solution of Eq. (12)(i)is then
− λm, obtained
pare by of
the graphs cs (t) in this case withinto
resubstituting Eq. (17):
the graphs obtained
in Problem 4.
where λ is the associated decay rate constant. Find solutions (b) Assume that the 90% efficient source remediation pro-
of Eq. (i) using the initial condition m(0) = m0 for the follow- ln ln
cess is not applied until t1 = 10 years have elapsed following
(18
ing cases: (i) 𝛾 = 1, (ii) 𝛾 ≠ 1 and λ = 0, (iii) 𝛾 ≠ 1 and λ ≠ the initial deposition of the contaminant. Under this sce-
0. Then find expressions for cs (t) using Eq. (2). nario, plot the graphs of cs (t) using the parameters and initial
Hint: Eq. (i) is a type of nonlinear equation known as a conditions of Problem 4. In this case, use Eq. (2) to compute
Bernoulli equation. A method for Remark.
solvingLooking back,
Bernoulli had weconcentration
equa- used the substitution
for 0 ≤ t < t1 . Following remediation, use the
tions is discussed in Section 2.7. m(t1 ) = m1 =
initial condition dy du0.1m(t1 − 0) = 0.1 limt↑t1 m(t)
3. Show that when 𝛾 ≥ 1, the source has an infinite lifetime, for Eq. ux (i) and use the following modification of
dx dx
but if 0 < 𝛾 < 1, the source has a finite lifetime. In the latter Eq. (2):
case, find the time that the DNAPL the algebra in Eq.
source mass (14) the
attains would have been slightly worse in that simplifying the left side woul
[ ]𝛾
value zero. have entailed multiplying two binomials, whereas cs (t) we only
m(t)had to multiply a monomial times
= , t > t1 , (ii)
in Eq. (14) when using c1 m1
4. Assume the following values for the parameters:
m0 = 1,620 kg, c0 = 100 mg/L, As = 30 m2 , 𝑣d = 20 where c1 = (0.1)𝛾 c(t1 − 0) = (0.1)𝛾 limt↑t1 c(t) to compute
m/year, λ = 0. Use the solutions obtained in Problem 2
Bernoulli Differential Equations
to plot graphs of cs (t) for each of the following cases: (i)
concentrations for times t > t1 . Compare the graphs of cs (t)
in this case with the graphs obtained in Problems 4 and 5(a).
𝛾 = 0.5 for 0 ≤ t ≤ tf , where csA ) = 0,order
(tf first and (ii) = 2 for equation
di 𝛾erential Can yourelated
draw to
anylinear di erential
conclusions equations
about the possibleiseffective-
the so-calle
0 ≤ t ≤ 100 years. Bernoulli equation, named after nessJacob Bernoulli
of source (1654–1705)
remediation? andaresolved
If so, what they? first by Leibnit
in 1696. Such an equation has the following form.