AdePy contains analytical solutions for the advection-dispersion equation (ADE) describing solute transport in groundwater, written in Python.
Currently, all solutions shown in Wexler (1992) are provided as separate Python functions. These simulate 1D, 2D or 3D solute transport in uniform background flow for a variety of boundary conditions and source geometries. The solute may be subjected to linear sorption and first-order decay.
Since all equations are linear, superposition in time and space can be applied to create complex source geometries with time-varying source concentrations. Gauss-Legendre quadrature is used to solve the integrals which require numerical integration.
Download or git clone
the GitHub repository locally. Then install using:
pip install -e <path/to/local/clone>
AdePy depends on NumPy, SciPy and Numba.
Coming soon.
Module | Function | Dimensionality | Source geometry | Boundary type | Aquifer geometry | Reference |
---|---|---|---|---|---|---|
uniform |
finite1() |
1D | Inlet | Dirichlet | Finite | Wexler (1992) |
finite3() |
1D | Inlet | Cauchy | Finite | Wexler (1992) | |
seminf1() |
1D | Inlet | Dirichlet | Semi-infinite | Wexler (1992) | |
seminf3() |
1D | Inlet | Cauchy | Semi-infinite | Wexler (1992) | |
point2() |
2D | Point | Cauchy | Infinite | Wexler (1992) | |
stripf() |
2D | Finite Y at X=0 | Dirichlet | Finite Y | Wexler (1992) | |
stripi() |
2D | Finite Y at X=0 | Dirichlet | Semi-infinite | Wexler (1992) | |
gauss() |
2D | Gaussian along Y at X=0 | Dirichlet | Semi-infinite | Wexler (1992) | |
point3() |
3D | Point | Cauchy | Infinite | Wexler (1992) | |
patchf() |
3D | Finite Y and Z at X=0 | Dirichlet | Finite Y and Z | Wexler (1992) | |
patchi() |
3D | Finite Y and Z at X=0 | Dirichlet | Semi-infinite | Wexler (1992) |
The fate of a contaminant plume generated by continuous injection of a point source in an aquifer with uniform background flow is simulated. The source generates a plume which extends in three dimensions and migrates due to advection and mechanical dispersion. Molecular diffusion, linear sorption and first-order decay are neglected in this example.
import numpy as np
import matplotlib.pyplot as plt
# 3D ADE solution of a continuous point source in uniform background flow
from adepy.uniform import point3
# Source parameters ----
xc = 0 # x-coordinate of point source, m
yc = 0 # y-coordinate of point source, m
zc = 0 # z-coordinate of point source, m
c0 = 100 # injection concentration, mg/L
Q = 1 # injection rate, m^3/d
# Aquifer parameters ----
v = 0.05 # uniform groundwater flow velocity in x-direction, m/d
al = 5 # longitudinal dispersivity, m
ah = 1 # horizontal transverse dispersivity, m
av = 0.1 # vertical transverse dispersivity, m
n = 0.2 # porosity, -
# Calculate and plot the concentration field after 1 year at z = 0 ----
t = 365 # output time, d
x, y = np.meshgrid(np.linspace(-10, 20, 100),
np.linspace(-7.5, 7.5, 100)) # output grid x-y coordinates, m
z = 0 # output grid z-coordinate, m
c = point3(c0, x, y, z, t, v, n, al, ah, av, Q, xc, yc, zc) # simulated concentration, mg/L
plt.contour(x, y, c, levels=np.arange(100, 2501, 100))
plt.xlabel('x (m)')
plt.ylabel('y (m)')
plt.gca().set_aspect("equal")
plt.grid()
# Calculate and plot the concentration time series for 5 years at a location downstream ----
obs = (10, 0, 0) # x-y-z coordinates of observation point, m
t = np.linspace(1, 5 * 365, 100) # output times, d
cobs = point3(c0, obs[0], obs[1], obs[2], t, v, n, al, ah, av, Q, xc, yc, zc)
plt.plot(t, cobs)
plt.xlabel('Time (d)')
plt.ylabel('Concentration (mg/L)')