Walkthrough animations

This example script shows how to make animations of concentration distribution with mibitrans

The two lines below allow animations to be displayed.

In [1]:

import matplotlib.pyplot as plt
import numpy as np
import mibitrans as mbt

import matplotlib.pyplot as plt import numpy as np import mibitrans as mbt

Define model parameters

In [2]:

hydro = mbt.HydrologicalParameters(
    velocity=0.08,  # Groundwater flow velocity, in [m/day]
    porosity=0.25,  # Effective soil porosity [-]
    alpha_x=5,  # Longitudinal dispersivity, in [m]
    alpha_y=0.2,  # Transverse horizontal dispersivity, in [m]
    alpha_z=0.005,  # Transverse vertical dispersivity, in [m]
)

hydro = mbt.HydrologicalParameters( velocity=0.08, # Groundwater flow velocity, in [m/day] porosity=0.25, # Effective soil porosity [-] alpha_x=5, # Longitudinal dispersivity, in [m] alpha_y=0.2, # Transverse horizontal dispersivity, in [m] alpha_z=0.005, # Transverse vertical dispersivity, in [m] )

In [3]:

att = mbt.AttenuationParameters(
    bulk_density=1.7,  # Soil bulk density in [g/m^3]
    partition_coefficient=53,  # Partition coefficient of contaminant to soil organic matter, in [m^3/g]
    fraction_organic_carbon=1e-3,  # Fraction of organic material in the soil [-]
    half_life=0,  # Contaminant half life, in [days]
)

att = mbt.AttenuationParameters( bulk_density=1.7, # Soil bulk density in [g/m^3] partition_coefficient=53, # Partition coefficient of contaminant to soil organic matter, in [m^3/g] fraction_organic_carbon=1e-3, # Fraction of organic material in the soil [-] half_life=0, # Contaminant half life, in [days] )

In [4]:

source = mbt.SourceParameters(
    source_zone_boundary=np.array([2, 11, 20]),  # Source zone extent, in [m]
    source_zone_concentration=np.array([15, 4, 1]),  # Source zone concentrations, in [g/m^3]
    depth=3,  # Source depth extent, in [m]
    total_mass=5e6,  # Source mass, in [g] or as np.inf / "infinite"
)

source = mbt.SourceParameters( source_zone_boundary=np.array([2, 11, 20]), # Source zone extent, in [m] source_zone_concentration=np.array([15, 4, 1]), # Source zone concentrations, in [g/m^3] depth=3, # Source depth extent, in [m] total_mass=5e6, # Source mass, in [g] or as np.inf / "infinite" )

In [5]:

model = mbt.ModelParameters(
    model_length=600,  # Model extent in the longitudinal (x) direction in [m].
    model_width=60,  # Model extent in the transverse horizontal (y) direction in [m].
    model_time=20 * 365,  # Model duration in [days].
    dx=5,  # Model grid discretization step size in the longitudinal (x) direction, in [m].
    dy=0.5,  # Model grid discretization step size in the transverse horizontal (y) direction, in [m].
    dt=182,  # Model time discretization step size, in [days]
)

model = mbt.ModelParameters( model_length=600, # Model extent in the longitudinal (x) direction in [m]. model_width=60, # Model extent in the transverse horizontal (y) direction in [m]. model_time=20 * 365, # Model duration in [days]. dx=5, # Model grid discretization step size in the longitudinal (x) direction, in [m]. dy=0.5, # Model grid discretization step size in the transverse horizontal (y) direction, in [m]. dt=182, # Model time discretization step size, in [days] )

Make and run some models, everything shown here also works for anatrans and bioscreen models

In [6]:

mbt_model = mbt.Mibitrans(hydro, att, source, model)
mbt_results_nodecay = mbt_model.run()
mbt_model.attenuation_parameters.half_life = 10 * 365
mbt_results_lindecay = mbt_model.run()
mbt_model.instant_reaction(electron_acceptors=[9, 6, 8, 5, 4], utilization_factor=[3.14, 4.9, 21.8, 4.7, 0.78])
mbt_results_instant = mbt_model.run()

mbt_model = mbt.Mibitrans(hydro, att, source, model) mbt_results_nodecay = mbt_model.run() mbt_model.attenuation_parameters.half_life = 10 * 365 mbt_results_lindecay = mbt_model.run() mbt_model.instant_reaction(electron_acceptors=[9, 6, 8, 5, 4], utilization_factor=[3.14, 4.9, 21.8, 4.7, 0.78]) mbt_results_instant = mbt_model.run()

To produce an animation, set animate to True. Also assign the animation object to a variable. To view the animation save it to a .gif file, with the fps as an optional argument.

In [7]:

anim = mbt_results_nodecay.centerline(animate=True)
anim.save("plume_nd_animation.gif", fps=10)
plt.clf()

anim = mbt_results_nodecay.centerline(animate=True) anim.save("plume_nd_animation.gif", fps=10) plt.clf()

MovieWriter ffmpeg unavailable; using Pillow instead.

<Figure size 640x480 with 0 Axes>

Combine multiple models by using list input

In [8]:

anim2 = mbt.centerline(
    model=[mbt_results_nodecay, mbt_results_lindecay, mbt_results_instant],
    legend_names=["no decay", "linear decay", "instant reaction"],
    animate=True,
)
anim2.save("plume_center_animation.gif", fps=10)
plt.clf()

anim2 = mbt.centerline( model=[mbt_results_nodecay, mbt_results_lindecay, mbt_results_instant], legend_names=["no decay", "linear decay", "instant reaction"], animate=True, ) anim2.save("plume_center_animation.gif", fps=10) plt.clf()

MovieWriter ffmpeg unavailable; using Pillow instead.

<Figure size 640x480 with 0 Axes>

Also works for transverse plots

In [9]:

anim3 = mbt.transverse(
    model=[mbt_results_nodecay, mbt_results_lindecay, mbt_results_instant],
    x_position=150,
    legend_names=["no decay", "linear decay", "instant reaction"],
    animate=True,
)
anim3.save("plume_trans_animation.gif", fps=10)
plt.clf()

anim3 = mbt.transverse( model=[mbt_results_nodecay, mbt_results_lindecay, mbt_results_instant], x_position=150, legend_names=["no decay", "linear decay", "instant reaction"], animate=True, ) anim3.save("plume_trans_animation.gif", fps=10) plt.clf()

MovieWriter ffmpeg unavailable; using Pillow instead.

<Figure size 640x480 with 0 Axes>

For breakthrough curve, animation draws the next point for each time step.

In [10]:

anim4 = mbt.breakthrough(
    model=[mbt_results_nodecay, mbt_results_lindecay, mbt_results_instant],
    x_position=150,
    legend_names=["no decay", "linear decay", "instant reaction"],
    animate=True,
)
anim4.save("plume_btc_animation.gif", fps=10)
plt.clf()

anim4 = mbt.breakthrough( model=[mbt_results_nodecay, mbt_results_lindecay, mbt_results_instant], x_position=150, legend_names=["no decay", "linear decay", "instant reaction"], animate=True, ) anim4.save("plume_btc_animation.gif", fps=10) plt.clf()

MovieWriter ffmpeg unavailable; using Pillow instead.

<Figure size 640x480 with 0 Axes>

Animate 2-dimensional and 3-dimensional plots as well

In [11]:

anim5 = mbt_results_instant.plume_2d(animate=True, cmap="viridis")
anim5.save("plume_2d_animation.gif", fps=10)
plt.clf()

anim5 = mbt_results_instant.plume_2d(animate=True, cmap="viridis") anim5.save("plume_2d_animation.gif", fps=10) plt.clf()

MovieWriter ffmpeg unavailable; using Pillow instead.

<Figure size 640x480 with 0 Axes>

In [12]:

anim6 = mbt_results_nodecay.plume_3d(animate=True, cmap="viridis")
anim6.save("plume_3d_animation.gif", fps=10)
plt.close()

anim6 = mbt_results_nodecay.plume_3d(animate=True, cmap="viridis") anim6.save("plume_3d_animation.gif", fps=10) plt.close()

MovieWriter ffmpeg unavailable; using Pillow instead.

Currently it is not possible to adjust individual line colors/styles when animating multiple models in a single plot. A first version of this functionality is found in mibitrans.visualize.animation.animate_1d. However, this is still under development and needs improvement to become a core feature of the mibitrans package.

To display animations inside a python IDE, use the import statement below. Then after assignment of the animation object to a variable, use plt.show() to display the animation. This however might require additional setup with TkAgg.

import matplotlib

matplotlib.use("TkAgg")