Solutions to the AdvectionDispersion Equation - PowerPoint PPT Presentation

1 / 19
About This Presentation
Title:

Solutions to the AdvectionDispersion Equation

Description:

For a given boundary condition and time t*, the solution is unique and independent of R ... Only include region of x 0 in domain ... – PowerPoint PPT presentation

Number of Views:25
Avg rating:3.0/5.0
Slides: 20
Provided by: breO
Category:

less

Transcript and Presenter's Notes

Title: Solutions to the AdvectionDispersion Equation


1
Solutions to the Advection-Dispersion Equation
2
Road Map to Solutions
  • We will discuss the following solutions
  • Instantaneous injection in infinite and
    semi-infinite 1-dimensional columns
  • Continuous injection into semi-infinite 1-D
    column
  • Instantaneous point source solution in
    two-dimensions (line source in 3-D)
  • Instantaneous point source in 3-dimensions
  • Keep an eye on
  • the initial assumptions
  • symmetry in space, asymmetry in time

3
Recall the Governing Equation
  • What have we assumed thus far?
  • Dispersion can be expressed as a Fickian process
  • Diffusion and diffusion can be folded into a
    single hydrodynamic dispersion
  • First order decay
  • What do we need next?
  • More Assumptions!

4
Adding Sorption
  • Thus far we have addressed only the solute
    behavior in the liquid state.
  • We now add sorption using a linear isotherm
  • Recall the linear isotherm relationshipwhere
    cl and cs are in mass per volume of water and
    mass per mass of solid respectively
  • The total concentration is then

5
Retardation factor
  • We have
  • Which may be written as
  • where we have defined the retardation factor R to
    be

6
Putting this all together
  • The ADE with 1st order decay linear isotherm
  • What do we need now? More assumptions!
  • ? constant in space (pull from derivatives
    cancel)
  • D constant in space (slide it out of derivative)
  • R constant in time (slide it out of derivative)
  • Use the chain rulethe divergence of
  • a scalar gradientdivergence of a constant is
    zero

7
Applying the previous assumptions
  • The divergence operators turn into gradient
    operators since they are applied to scalar
    quantities.
  • What does this give us? The new ADE

8
Looking at 1-D case for a moment
  • To see how this retardation factor works, take t
    t/R, and ? 0. With a little algebra,
  • The punch line
  • the spatial distribution of solutes is the same
    in the case of non-adsorbed vs. adsorbed
    compounds!
  • For a given boundary condition and time t, the
    solution is unique and independent of R

9
1-D infinite column Instantaneous Point Injection
  • Column goes to ??and -?
  • Area A
  • steady velocity u
  • mass M injected at x o and t0(boundary
    condition)
  • initially uncontaminated column. i.e. c(x,0) 0
    (initial condition)
  • linear sorption (retardation R)
  • first order decay (?)

velocity u
x 0
10
  • Features of solution
  • Gaussian, symmetric in space, ?2 Dt/R
  • Exponential decay of pulse
  • Except for decay, R only shows up as t/R

Upstream solutes
Peak at 1.23 hr
Center of Mass 2 hr
Spatial Distribution
Temporal Distribution
11
1-D semi-infinite Instantaneous Point Injection
  • How do we handle a surface application?
  • Use the linearity of the simplified ADE
  • Can add any two solutions, and still a solution
  • By uniqueness, any solution which satisfies
    initial and boundary conditions is THE solution
  • Boundary and initial conditions

12
Semi-infinite solution
  • Upward pulse and downward.
  • Only include region of x gt 0 in domain
  • Solution symmetric about x 0, therefore slope
    of dc(x0,t)/dx 0 for all t, as required
  • Compared to infinite column, c starts twice as
    high, but in time goes to same solution

13
Continuous injection, 1-D
  • Since the simplified ADE is linear, we use
    superposition. Basically get a continuous
    injection solution by adding infinitely many
    infinitely small Gaussian plumes.
  • Use the complementary error function erfc

14
Plot of erfc(x)
15
Solution for continuous injection, 1-D
16
Plot of solution

? 0.1, R 1, ?? 0.02, u 1.0, and m 1
17
1-D, Cont., simplified
  • With no sorption or degradation this reduces to

18
2-D and 3-D instantaneous solutions
Note - Same Gaussian form as 1-D - Note
separation of longitudinal and transverse
dispersion
19
Review of Assumptions
  • Assumption Effects if Violated
  • ? constant in space -R higher where ??lower
  • -Velocity varies inversely with ?
  • D constant in space -Increased overall
    dispersion due
  • to heterogeneity
  • D independent of scale - Plume will grow more
    slowly at
  • first, then faster.
  • Reversible Sorption - Increase plume spreading
    and
  • overall region of contamination
  • Equilibrium Sorption - Increased tailing and
    spreading
  • Linear Sorption - Higher peak C and faster
    travel
  • Anisotropic media - Stretching smearing along
    beds
  • Heterogeneous Media - Greater scale effects of D
    and ALL
  • EFFECTS DISCUSSED ABOVE
Write a Comment
User Comments (0)
About PowerShow.com