# PySDM
[](https://www.python.org/)
[](https://numba.pydata.org)
[](https://pypi.org/project/ThrustRTC/)
[](https://en.wikipedia.org/wiki/Linux)
[](https://en.wikipedia.org/wiki/macOS)
[](https://en.wikipedia.org/wiki/Windows)
[](https://jupyter.org/)
[](https://github.com/open-atmos/PySDM/graphs/commit-activity)
[](https://www.openhub.net/p/atmos-cloud-sim-uj-PySDM)
[](https://joss.theoj.org/papers/62cad07440b941f73f57d187df1aa6e9)
[](https://zenodo.org/badge/latestdoi/199064632)
[](https://www.fnp.org.pl/en/)
[](https://www.ncn.gov.pl/?language=en)
[](https://asr.science.energy.gov/)
[](https://www.gnu.org/licenses/gpl-3.0.html)
[](https://github.com/open-atmos/PySDM/actions)
[](https://ci.appveyor.com/project/slayoo/pysdm/branch/main)
[](https://app.codecov.io/gh/open-atmos/PySDM)
[](https://pypi.org/project/PySDM)
[](https://open-atmos.github.io/PySDM/)
PySDM is a package for simulating the dynamics of population of particles.
It is intended to serve as a building block for simulation systems modelling
fluid flows involving a dispersed phase,
with PySDM being responsible for representation of the dispersed phase.
Currently, the development is focused on atmospheric cloud physics
applications, in particular on modelling the dynamics of particles immersed in moist air
using the particle-based (a.k.a. super-droplet) approach
to represent aerosol/cloud/rain microphysics.
The package features a Pythonic high-performance implementation of the
Super-Droplet Method (SDM) Monte-Carlo algorithm for representing collisional growth
([Shima et al. 2009](https://rmets.onlinelibrary.wiley.com/doi/abs/10.1002/qj.441)), hence the name.
There is a growing set of example Jupyter notebooks exemplifying how to perform
various types of calculations and simulations using PySDM.
Most of the example notebooks reproduce results and plot from literature, see below for
a list of examples and links to the notebooks (which can be either executed or viewed
"in the cloud").
There are also a growing set of [tutorials](https://github.com/open-atmos/PySDM/tree/main/tutorials), also in the form of Jupyter notebooks.
These tutorials are intended for teaching purposes and include short explanations of cloud microphysical
concepts paired with widgets for running interactive simulations using PySDM.
Each tutorial also comes with a set of questions at the end that can be used as homework problems.
Like the examples, these tutorials can be executed or viewed "in the cloud" making it an especially
easy way for students to get started.
PySDM has two alternative parallel number-crunching backends
available: multi-threaded CPU backend based on [Numba](http://numba.pydata.org/)
and GPU-resident backend built on top of [ThrustRTC](https://pypi.org/project/ThrustRTC/).
The [`Numba`](https://open-atmos.github.io/PySDM/PySDM/backends/numba/numba.html) backend (aliased ``CPU``) features multi-threaded parallelism for
multi-core CPUs, it uses the just-in-time compilation technique based on the LLVM infrastructure.
The [`ThrustRTC`](https://open-atmos.github.io/PySDM/PySDM/backends/thrustRTC/thrustRTC.html) backend (aliased ``GPU``) offers GPU-resident operation of PySDM
leveraging the [SIMT](https://en.wikipedia.org/wiki/Single_instruction,_multiple_threads)
parallelisation model.
Using the ``GPU`` backend requires nVidia hardware and [CUDA driver](https://developer.nvidia.com/cuda-downloads).
For an overview of PySDM features (and the preferred way to cite PySDM in papers), please refer to our JOSS papers:
- [Bartman et al. 2022](https://doi.org/10.21105/joss.03219) (PySDM v1).
- [de Jong, Singer et al. 2023](https://doi.org/10.21105/joss.04968) (PySDM v2).
PySDM includes an extension of the SDM scheme to represent collisional breakup described in [de Jong, Mackay et al. 2023](10.5194/gmd-16-4193-2023).
For a list of talks and other materials on PySDM, see the [project wiki](https://github.com/open-atmos/PySDM/wiki).
A [pdoc-generated](https://pdoc3.github.io/pdoc) documentation of PySDM public API is maintained at: [https://open-atmos.github.io/PySDM](https://open-atmos.github.io/PySDM)
## Example Jupyter notebooks (reproducing results from literature):
See [PySDM-examples README](https://github.com/open-atmos/PySDM/blob/main/examples/README.md).
## Dependencies and Installation
PySDM dependencies are: [Numpy](https://numpy.org/), [Numba](http://numba.pydata.org/), [SciPy](https://scipy.org/),
[Pint](https://pint.readthedocs.io/), [chempy](https://pypi.org/project/chempy/),
[pyevtk](https://pypi.org/project/pyevtk/),
[ThrustRTC](https://fynv.github.io/ThrustRTC/) and [CURandRTC](https://github.com/fynv/CURandRTC).
To install PySDM using ``pip``, use: ``pip install PySDM``
(or ``pip install git+https://github.com/open-atmos/PySDM.git`` to get updates
beyond the latest release).
Conda users may use ``pip`` as well, see the [Installing non-conda packages](https://docs.conda.io/projects/conda/en/latest/user-guide/tasks/manage-pkgs.html#installing-non-conda-packages) section in the conda docs. Dependencies of PySDM are available at the following conda channels:
- numba: [numba](https://anaconda.org/numba/numba)
- conda-forge: [pyevtk](https://anaconda.org/conda-forge/pyevtk), [pint](https://anaconda.org/conda-forge/pint) and []()
- fyplus: [ThrustRTC](https://anaconda.org/fyplus/thrustrtc), [CURandRTC](https://anaconda.org/fyplus/curandrtc)
- bjodah: [chempy](https://anaconda.org/bjodah/chempy)
- nvidia: [cudatoolkit](https://anaconda.org/nvidia/cudatoolkit)
For development purposes, we suggest cloning the repository and installing it using ``pip -e``.
Test-time dependencies can be installed with ``pip -e .[tests]``.
PySDM examples constitute the [``PySDM-examples``](https://github.com/open-atmos/PySDM-examples) package.
The examples have additional dependencies listed in [``PySDM_examples`` package ``setup.py``](https://github.com/open-atmos/PySDM/blob/main/examples/setup.py) file.
Running the example Jupyter notebooks requires the ``PySDM_examples`` package to be installed.
The suggested install and launch steps are:
```
git clone https://github.com/open-atmos/PySDM.git
pip install -e PySDM
pip install -e PySDM/examples
jupyter-notebook PySDM/examples/PySDM_examples
```
Alternatively, one can also install the examples package from pypi.org by
using ``pip install PySDM-examples`` (note that this does not apply to notebooks itself,
only the supporting .py files).
## Submodule organization
```mermaid
mindmap
root((PySDM))
Builder
Formulae
Particulator
((attributes))
(physics)
DryVolume: ExtensiveAttribute
Kappa: DerivedAttribute
...
(chemistry)
Acidity
...
(...)
((backends))
CPU
GPU
((dynamics))
AqueousChemistry
Collision
Condensation
...
((environments))
Box
Parcel
Kinematic2D
...
((initialisation))
(spectra)
Lognormal
Exponential
...
(sampling)
(spectral_sampling)
ConstantMultiplicity
UniformRandom
Logarithmic
...
(...)
(...)
((physics))
(hygroscopicity)
KappaKoehler
...
(condensation_coordinate)
Volume
VolumeLogarithm
(...)
((products))
(size_spectral)
EffectiveRadius
WaterMixingRatio
...
(ambient_thermodynamics)
AmbientRelativeHumidity
...
(...)
```
## Hello-world coalescence example in Python, Julia and Matlab
In order to depict the PySDM API with a practical example, the following
listings provide sample code roughly reproducing the
Figure 2 from [Shima et al. 2009 paper](http://doi.org/10.1002/qj.441)
using PySDM from Python, Julia and Matlab.
It is a [`Coalescence`](https://open-atmos.github.io/PySDM/PySDM/dynamics/coalescence.html)-only set-up in which the initial particle size
spectrum is [`Exponential`](https://open-atmos.github.io/PySDM/PySDM/initialisation/spectra.html#PySDM.initialisation.spectra.Exponential) and is deterministically sampled to match
the condition of each super-droplet having equal initial multiplicity:
<details>
<summary>Julia (click to expand)</summary>
```Julia
using Pkg
Pkg.add("PyCall")
Pkg.add("Plots")
Pkg.add("PlotlyJS")
using PyCall
si = pyimport("PySDM.physics").si
ConstantMultiplicity = pyimport("PySDM.initialisation.sampling.spectral_sampling").ConstantMultiplicity
Exponential = pyimport("PySDM.initialisation.spectra").Exponential
n_sd = 2^15
initial_spectrum = Exponential(norm_factor=8.39e12, scale=1.19e5 * si.um^3)
attributes = Dict()
attributes["volume"], attributes["multiplicity"] = ConstantMultiplicity(spectrum=initial_spectrum).sample(n_sd)
```
</details>
<details>
<summary>Matlab (click to expand)</summary>
```Matlab
si = py.importlib.import_module('PySDM.physics').si;
ConstantMultiplicity = py.importlib.import_module('PySDM.initialisation.sampling.spectral_sampling').ConstantMultiplicity;
Exponential = py.importlib.import_module('PySDM.initialisation.spectra').Exponential;
n_sd = 2^15;
initial_spectrum = Exponential(pyargs(...
'norm_factor', 8.39e12, ...
'scale', 1.19e5 * si.um ^ 3 ...
));
tmp = ConstantMultiplicity(initial_spectrum).sample(int32(n_sd));
attributes = py.dict(pyargs('volume', tmp{1}, 'multiplicity', tmp{2}));
```
</details>
<details open>
<summary>Python (click to expand)</summary>
```Python
from PySDM.physics import si
from PySDM.initialisation.sampling.spectral_sampling import ConstantMultiplicity
from PySDM.initialisation.spectra.exponential import Exponential
n_sd = 2 ** 15
initial_spectrum = Exponential(norm_factor=8.39e12, scale=1.19e5 * si.um ** 3)
attributes = {}
attributes['volume'], attributes['multiplicity'] = ConstantMultiplicity(initial_spectrum).sample(n_sd)
```
</details>
The key element of the PySDM interface is the [``Particulator``](https://open-atmos.github.io/PySDM/PySDM/particulator.html)
class instances of which are used to manage the system state and control the simulation.
Instantiation of the [``Particulator``](https://open-atmos.github.io/PySDM/PySDM/particulator.html) class is handled by the [``Builder``](https://open-atmos.github.io/PySDM/PySDM/builder.html)
as exemplified below:
<details>
<summary>Julia (click to expand)</summary>
```Julia
Builder = pyimport("PySDM").Builder
Box = pyimport("PySDM.environments").Box
Coalescence = pyimport("PySDM.dynamics").Coalescence
Golovin = pyimport("PySDM.dynamics.collisions.collision_kernels").Golovin
CPU = pyimport("PySDM.backends").CPU
ParticleVolumeVersusRadiusLogarithmSpectrum = pyimport("PySDM.products").ParticleVolumeVersusRadiusLogarithmSpectrum
radius_bins_edges = 10 .^ range(log10(10*si.um), log10(5e3*si.um), length=32)
env = Box(dt=1 * si.s, dv=1e6 * si.m^3)
builder = Builder(n_sd=n_sd, backend=CPU(), environment=env)
builder.add_dynamic(Coalescence(collision_kernel=Golovin(b=1.5e3 / si.s)))
products = [ParticleVolumeVersusRadiusLogarithmSpectrum(radius_bins_edges=radius_bins_edges, name="dv/dlnr")]
particulator = builder.build(attributes, products)
```
</details>
<details>
<summary>Matlab (click to expand)</summary>
```Matlab
Builder = py.importlib.import_module('PySDM').Builder;
Box = py.importlib.import_module('PySDM.environments').Box;
Coalescence = py.importlib.import_module('PySDM.dynamics').Coalescence;
Golovin = py.importlib.import_module('PySDM.dynamics.collisions.collision_kernels').Golovin;
CPU = py.importlib.import_module('PySDM.backends').CPU;
ParticleVolumeVersusRadiusLogarithmSpectrum = py.importlib.import_module('PySDM.products').ParticleVolumeVersusRadiusLogarithmSpectrum;
radius_bins_edges = logspace(log10(10 * si.um), log10(5e3 * si.um), 32);
env = Box(pyargs('dt', 1 * si.s, 'dv', 1e6 * si.m ^ 3));
builder = Builder(pyargs('n_sd', int32(n_sd), 'backend', CPU(), 'environment', env));
builder.add_dynamic(Coalescence(pyargs('collision_kernel', Golovin(1.5e3 / si.s))));
products = py.list({ ParticleVolumeVersusRadiusLogarithmSpectrum(pyargs( ...
'radius_bins_edges', py.numpy.array(radius_bins_edges), ...
'name', 'dv/dlnr' ...
)) });
particulator = builder.build(attributes, products);
```
</details>
<details open>
<summary>Python (click to expand)</summary>
```Python
import numpy as np
from PySDM import Builder
from PySDM.environments import Box
from PySDM.dynamics import Coalescence
from PySDM.dynamics.collisions.collision_kernels import Golovin
from PySDM.backends import CPU
from PySDM.products import ParticleVolumeVersusRadiusLogarithmSpectrum
radius_bins_edges = np.logspace(np.log10(10 * si.um), np.log10(5e3 * si.um), num=32)
env = Box(dt=1 * si.s, dv=1e6 * si.m ** 3)
builder = Builder(n_sd=n_sd, backend=CPU(), environment=env)
builder.add_dynamic(Coalescence(collision_kernel=Golovin(b=1.5e3 / si.s)))
products = [ParticleVolumeVersusRadiusLogarithmSpectrum(radius_bins_edges=radius_bins_edges, name='dv/dlnr')]
particulator = builder.build(attributes, products)
```
</details>
The ``backend`` argument may be set to ``CPU`` or ``GPU``
what translates to choosing the multi-threaded backend or the
GPU-resident computation mode, respectively.
The employed [`Box`](https://open-atmos.github.io/PySDM/PySDM/environments/box.html) environment corresponds to a zero-dimensional framework
(particle positions are not considered).
The vectors of particle multiplicities ``n`` and particle volumes ``v`` are
used to initialise super-droplet attributes.
The [`Coalescence`](https://open-atmos.github.io/PySDM/PySDM/dynamics/coalescence.html)
Monte-Carlo algorithm (Super Droplet Method) is registered as the only
dynamic in the system.
Finally, the [`build()`](https://open-atmos.github.io/PySDM/PySDM/builder.html#PySDM.builder.Builder.build) method is used to obtain an instance
of [`Particulator`](https://open-atmos.github.io/PySDM/PySDM/particulator.html#PySDM.particulator.Particulator) which can then be used to control time-stepping and
access simulation state.
The [`run(nt)`](https://open-atmos.github.io/PySDM/PySDM/particulator.html#PySDM.particuparticulatorr.Particulator.run) method advances the simulation by ``nt`` timesteps.
In the listing below, its usage is interleaved with plotting logic
which displays a histogram of particle mass distribution
at selected timesteps:
<details>
<summary>Julia (click to expand)</summary>
```Julia
using Plots; plotlyjs()
for step = 0:1200:3600
particulator.run(step - particulator.n_steps)
plot!(
radius_bins_edges[1:end-1] / si.um,
particulator.formulae.particle_shape_and_density.volume_to_mass(
particulator.products["dv/dlnr"].get()[:]
)/ si.g,
linetype=:steppost,
xaxis=:log,
xlabel="particle radius [µm]",
ylabel="dm/dlnr [g/m^3/(unit dr/r)]",
label="t = $step s"
)
end
savefig("plot.svg")
```
</details>
<details>
<summary>Matlab (click to expand)</summary>
```Matlab
for step = 0:1200:3600
particulator.run(int32(step - particulator.n_steps));
x = radius_bins_edges / si.um;
y = particulator.formulae.particle_shape_and_density.volume_to_mass( ...
particulator.products{"dv/dlnr"}.get() ...
) / si.g;
stairs(...
x(1:end-1), ...
double(py.array.array('d',py.numpy.nditer(y))), ...
'DisplayName', sprintf("t = %d s", step) ...
);
hold on
end
hold off
set(gca,'XScale','log');
xlabel('particle radius [µm]')
ylabel("dm/dlnr [g/m^3/(unit dr/r)]")
legend()
```
</details>
<details open>
<summary>Python (click to expand)</summary>
```Python
from matplotlib import pyplot
for step in [0, 1200, 2400, 3600]:
particulator.run(step - particulator.n_steps)
pyplot.step(
x=radius_bins_edges[:-1] / si.um,
y=particulator.formulae.particle_shape_and_density.volume_to_mass(
particulator.products['dv/dlnr'].get()[0]
) / si.g,
where='post', label=f"t = {step}s"
)
pyplot.xscale('log')
pyplot.xlabel('particle radius [µm]')
pyplot.ylabel("dm/dlnr [g/m$^3$/(unit dr/r)]")
pyplot.legend()
pyplot.savefig('readme.png')
```
</details>
The resultant plot (generated with the Python code) looks as follows:
The component submodules used to create this simulation are visualized below:
```mermaid
graph
COAL[":Coalescence"] --->|passed as arg to| BUILDER_ADD_DYN(["Builder.add_dynamic()"])
BUILDER_INSTANCE["builder :Builder"] -...-|has a method| BUILDER_BUILD(["Builder.build()"])
ATTRIBUTES[attributes: dict] -->|passed as arg to| BUILDER_BUILD
N_SD["n_sd :int"] ---->|passed as arg to| BUILDER_INIT
BUILDER_INIT(["Builder.__init__()"]) --->|instantiates| BUILDER_INSTANCE
BUILDER_INSTANCE -..-|has a method| BUILDER_ADD_DYN(["Builder.add_dynamic()"])
ENV_INIT(["Box.__init__()"]) -->|instantiates| ENV
DT[dt :float] -->|passed as arg to| ENV_INIT
DV[dv :float] -->|passed as arg to| ENV_INIT
ENV[":Box"] -->|passed as arg to| BUILDER_INIT
B["b: float"] --->|passed as arg to| KERNEL_INIT(["Golovin.__init__()"])
KERNEL_INIT -->|instantiates| KERNEL
KERNEL[collision_kernel: Golovin] -->|passed as arg to| COAL_INIT(["Coalesncence.__init__()"])
COAL_INIT -->|instantiates| COAL
PRODUCTS[products: list] ----->|passed as arg to| BUILDER_BUILD
NORM_FACTOR[norm_factor: float]-->|passed as arg to| EXP_INIT
SCALE[scale: float]-->|passed as arg to| EXP_INIT
EXP_INIT(["Exponential.__init__()"]) -->|instantiates| IS
IS["initial_spectrum :Exponential"] -->|passed as arg to| CM_INIT
CM_INIT(["ConstantMultiplicity.__init__()"]) -->|instantiates| CM_INSTANCE
CM_INSTANCE[":ConstantMultiplicity"] -.-|has a method| SAMPLE
SAMPLE(["ConstantMultiplicity.sample()"]) -->|returns| n
SAMPLE -->|returns| volume
n -->|added as element of| ATTRIBUTES
PARTICULATOR_INSTANCE -.-|has a method| PARTICULATOR_RUN(["Particulator.run()"])
volume -->|added as element of| ATTRIBUTES
BUILDER_BUILD -->|returns| PARTICULATOR_INSTANCE["particulator :Particulator"]
PARTICULATOR_INSTANCE -.-|has a field| PARTICULATOR_PROD(["Particulator.products:dict"])
BACKEND_INSTANCE["backend :CPU"] ---->|passed as arg to| BUILDER_INIT
PRODUCTS -.-|accessible via| PARTICULATOR_PROD
NP_LOGSPACE(["np.logspace()"]) -->|returns| EDGES
EDGES[radius_bins_edges: np.ndarray] -->|passed as arg to| SPECTRUM_INIT
SPECTRUM_INIT["ParticleVolumeVersusRadiusLogarithmSpectrum.__init__()"] -->|instantiates| SPECTRUM
SPECTRUM[":ParticleVolumeVersusRadiusLogarithmSpectrum"] -->|added as element of| PRODUCTS
click COAL "https://open-atmos.github.io/PySDM/PySDM/dynamics/collisions/collision.html"
click BUILDER_INSTANCE "https://open-atmos.github.io/PySDM/PySDM/builder.html"
click BUILDER_INIT "https://open-atmos.github.io/PySDM/PySDM/builder.html"
click BUILDER_ADD_DYN "https://open-atmos.github.io/PySDM/PySDM/builder.html"
click ENV_INIT "https://open-atmos.github.io/PySDM/PySDM/environments/index.html"
click ENV "https://open-atmos.github.io/PySDM/PySDM/environments/index.html"
click KERNEL_INIT "https://open-atmos.github.io/PySDM/PySDM/dynamics/collisions/collision_kernels/index.html"
click KERNEL "https://open-atmos.github.io/PySDM/PySDM/dynamics/collisions/collision_kernels/index.html"
click EXP_INIT "https://open-atmos.github.io/PySDM/PySDM/initialisation/spectra/index.html"
click IS "https://open-atmos.github.io/PySDM/PySDM/initialisation/spectra/index.html"
click CM_INIT "https://open-atmos.github.io/PySDM/PySDM/initialisation/sampling/spectral_sampling.html"
click CM_INSTANCE "https://open-atmos.github.io/PySDM/PySDM/initialisation/sampling/spectral_sampling.html"
click SAMPLE "https://open-atmos.github.io/PySDM/PySDM/initialisation/sampling/spectral_sampling.html"
click PARTICULATOR_INSTANCE "https://open-atmos.github.io/PySDM/PySDM/particulator.html"
click BACKEND_INSTANCE "https://open-atmos.github.io/PySDM/PySDM/backends/numba.html"
click BUILDER_BUILD "https://open-atmos.github.io/PySDM/PySDM/builder.html"
click NP_LOGSPACE "https://numpy.org/doc/stable/reference/generated/numpy.logspace.html"
click SPECTRUM_INIT "https://open-atmos.github.io/PySDM/PySDM/products/size_spectral/particle_volume_versus_radius_logarithm_spectrum.html"
click SPECTRUM "https://open-atmos.github.io/PySDM/PySDM/products/size_spectral/particle_volume_versus_radius_logarithm_spectrum.html"
```
## Hello-world condensation example in Python, Julia and Matlab
In the following example, a condensation-only setup is used with the adiabatic
[`Parcel`](https://open-atmos.github.io/PySDM/PySDM/environments/parcel.html) environment.
An initial [`Lognormal`](https://open-atmos.github.io/PySDM/PySDM/initialisation/spectra.html#PySDM.initialisation.spectra.Lognormal)
spectrum of dry aerosol particles is first initialised to equilibrium wet size for the given
initial humidity.
Subsequent particle growth due to [`Condensation`](https://open-atmos.github.io/PySDM/PySDM/dynamics/condensation.html) of water vapour (coupled with the release of latent heat)
causes a subset of particles to activate into cloud droplets.
Results of the simulation are plotted against vertical
[`ParcelDisplacement`](https://open-atmos.github.io/PySDM/PySDM/products/housekeeping/parcel_displacement.html)
and depict the evolution of
[`PeakSupersaturation`](https://open-atmos.github.io/PySDM/PySDM/products/condensation/peak_supersaturation.html),
[`EffectiveRadius`](https://open-atmos.github.io/PySDM/PySDM/products/size_spectral/effective_radius.html),
[`ParticleConcentration`](https://open-atmos.github.io/PySDM/PySDM/products/size_spectral/particle_concentration.html#PySDM.products.particles_concentration.ParticleConcentration)
and the
[`WaterMixingRatio `](https://open-atmos.github.io/PySDM/PySDM/products/size_spectral/water_mixing_ratio.html).
<details>
<summary>Julia (click to expand)</summary>
```Julia
using PyCall
using Plots; plotlyjs()
si = pyimport("PySDM.physics").si
spectral_sampling = pyimport("PySDM.initialisation.sampling").spectral_sampling
discretise_multiplicities = pyimport("PySDM.initialisation").discretise_multiplicities
Lognormal = pyimport("PySDM.initialisation.spectra").Lognormal
equilibrate_wet_radii = pyimport("PySDM.initialisation").equilibrate_wet_radii
CPU = pyimport("PySDM.backends").CPU
AmbientThermodynamics = pyimport("PySDM.dynamics").AmbientThermodynamics
Condensation = pyimport("PySDM.dynamics").Condensation
Parcel = pyimport("PySDM.environments").Parcel
Builder = pyimport("PySDM").Builder
Formulae = pyimport("PySDM").Formulae
products = pyimport("PySDM.products")
env = Parcel(
dt=.25 * si.s,
mass_of_dry_air=1e3 * si.kg,
p0=1122 * si.hPa,
initial_water_vapour_mixing_ratio=20 * si.g / si.kg,
T0=300 * si.K,
w= 2.5 * si.m / si.s
)
spectrum = Lognormal(norm_factor=1e4/si.mg, m_mode=50*si.nm, s_geom=1.4)
kappa = .5 * si.dimensionless
cloud_range = (.5 * si.um, 25 * si.um)
output_interval = 4
output_points = 40
n_sd = 256
formulae = Formulae()
builder = Builder(backend=CPU(formulae), n_sd=n_sd, environment=env)
builder.add_dynamic(AmbientThermodynamics())
builder.add_dynamic(Condensation())
r_dry, specific_concentration = spectral_sampling.Logarithmic(spectrum).sample(n_sd)
v_dry = formulae.trivia.volume(radius=r_dry)
r_wet = equilibrate_wet_radii(r_dry=r_dry, environment=env, kappa_times_dry_volume=kappa * v_dry)
attributes = Dict()
attributes["multiplicity"] = discretise_multiplicities(specific_concentration * env.mass_of_dry_air)
attributes["dry volume"] = v_dry
attributes["kappa times dry volume"] = kappa * v_dry
attributes["volume"] = formulae.trivia.volume(radius=r_wet)
particulator = builder.build(attributes, products=[
products.PeakSupersaturation(name="S_max", unit="%"),
products.EffectiveRadius(name="r_eff", unit="um", radius_range=cloud_range),
products.ParticleConcentration(name="n_c_cm3", unit="cm^-3", radius_range=cloud_range),
products.WaterMixingRatio(name="liquid water mixing ratio", unit="g/kg", radius_range=cloud_range),
products.ParcelDisplacement(name="z")
])
cell_id=1
output = Dict()
for (_, product) in particulator.products
output[product.name] = Array{Float32}(undef, output_points+1)
output[product.name][1] = product.get()[cell_id]
end
for step = 2:output_points+1
particulator.run(steps=output_interval)
for (_, product) in particulator.products
output[product.name][step] = product.get()[cell_id]
end
end
plots = []
ylbl = particulator.products["z"].unit
for (_, product) in particulator.products
if product.name != "z"
append!(plots, [plot(output[product.name], output["z"], ylabel=ylbl, xlabel=product.unit, title=product.name)])
end
global ylbl = ""
end
plot(plots..., layout=(1, length(output)-1))
savefig("parcel.svg")
```
</details>
<details>
<summary>Matlab (click to expand)</summary>
```Matlab
si = py.importlib.import_module('PySDM.physics').si;
spectral_sampling = py.importlib.import_module('PySDM.initialisation.sampling').spectral_sampling;
discretise_multiplicities = py.importlib.import_module('PySDM.initialisation').discretise_multiplicities;
Lognormal = py.importlib.import_module('PySDM.initialisation.spectra').Lognormal;
equilibrate_wet_radii = py.importlib.import_module('PySDM.initialisation').equilibrate_wet_radii;
CPU = py.importlib.import_module('PySDM.backends').CPU;
AmbientThermodynamics = py.importlib.import_module('PySDM.dynamics').AmbientThermodynamics;
Condensation = py.importlib.import_module('PySDM.dynamics').Condensation;
Parcel = py.importlib.import_module('PySDM.environments').Parcel;
Builder = py.importlib.import_module('PySDM').Builder;
Formulae = py.importlib.import_module('PySDM').Formulae;
products = py.importlib.import_module('PySDM.products');
env = Parcel(pyargs( ...
'dt', .25 * si.s, ...
'mass_of_dry_air', 1e3 * si.kg, ...
'p0', 1122 * si.hPa, ...
'initial_water_vapour_mixing_ratio', 20 * si.g / si.kg, ...
'T0', 300 * si.K, ...
'w', 2.5 * si.m / si.s ...
));
spectrum = Lognormal(pyargs('norm_factor', 1e4/si.mg, 'm_mode', 50 * si.nm, 's_geom', 1.4));
kappa = .5;
cloud_range = py.tuple({.5 * si.um, 25 * si.um});
output_interval = 4;
output_points = 40;
n_sd = 256;
formulae = Formulae();
builder = Builder(pyargs('backend', CPU(formulae), 'n_sd', int32(n_sd), 'environment', env));
builder.add_dynamic(AmbientThermodynamics());
builder.add_dynamic(Condensation());
tmp = spectral_sampling.Logarithmic(spectrum).sample(int32(n_sd));
r_dry = tmp{1};
v_dry = formulae.trivia.volume(pyargs('radius', r_dry));
specific_concentration = tmp{2};
r_wet = equilibrate_wet_radii(pyargs(...
'r_dry', r_dry, ...
'environment', env, ...
'kappa_times_dry_volume', kappa * v_dry...
));
attributes = py.dict(pyargs( ...
'multiplicity', discretise_multiplicities(specific_concentration * env.mass_of_dry_air), ...
'dry volume', v_dry, ...
'kappa times dry volume', kappa * v_dry, ...
'volume', formulae.trivia.volume(pyargs('radius', r_wet)) ...
));
particulator = builder.build(attributes, py.list({ ...
products.PeakSupersaturation(pyargs('name', 'S_max', 'unit', '%')), ...
products.EffectiveRadius(pyargs('name', 'r_eff', 'unit', 'um', 'radius_range', cloud_range)), ...
products.ParticleConcentration(pyargs('name', 'n_c_cm3', 'unit', 'cm^-3', 'radius_range', cloud_range)), ...
products.WaterMixingRatio(pyargs('name', 'liquid water mixing ratio', 'unit', 'g/kg', 'radius_range', cloud_range)) ...
products.ParcelDisplacement(pyargs('name', 'z')) ...
}));
cell_id = int32(0);
output_size = [output_points+1, length(py.list(particulator.products.keys()))];
output_types = repelem({'double'}, output_size(2));
output_names = [cellfun(@string, cell(py.list(particulator.products.keys())))];
output = table(...
'Size', output_size, ...
'VariableTypes', output_types, ...
'VariableNames', output_names ...
);
for pykey = py.list(keys(particulator.products))
get = py.getattr(particulator.products{pykey{1}}.get(), '__getitem__');
key = string(pykey{1});
output{1, key} = get(cell_id);
end
for i=2:output_points+1
particulator.run(pyargs('steps', int32(output_interval)));
for pykey = py.list(keys(particulator.products))
get = py.getattr(particulator.products{pykey{1}}.get(), '__getitem__');
key = string(pykey{1});
output{i, key} = get(cell_id);
end
end
i=1;
for pykey = py.list(keys(particulator.products))
product = particulator.products{pykey{1}};
if string(product.name) ~= "z"
subplot(1, width(output)-1, i);
plot(output{:, string(pykey{1})}, output.z, '-o');
title(string(product.name), 'Interpreter', 'none');
xlabel(string(product.unit));
end
if i == 1
ylabel(string(particulator.products{"z"}.unit));
end
i=i+1;
end
saveas(gcf, "parcel.png");
```
</details>
<details open>
<summary>Python (click to expand)</summary>
```Python
from matplotlib import pyplot
from PySDM.physics import si
from PySDM.initialisation import discretise_multiplicities, equilibrate_wet_radii
from PySDM.initialisation.spectra import Lognormal
from PySDM.initialisation.sampling import spectral_sampling
from PySDM.backends import CPU
from PySDM.dynamics import AmbientThermodynamics, Condensation
from PySDM.environments import Parcel
from PySDM import Builder, Formulae, products
env = Parcel(
dt=.25 * si.s,
mass_of_dry_air=1e3 * si.kg,
p0=1122 * si.hPa,
initial_water_vapour_mixing_ratio=20 * si.g / si.kg,
T0=300 * si.K,
w=2.5 * si.m / si.s
)
spectrum = Lognormal(norm_factor=1e4 / si.mg, m_mode=50 * si.nm, s_geom=1.5)
kappa = .5 * si.dimensionless
cloud_range = (.5 * si.um, 25 * si.um)
output_interval = 4
output_points = 40
n_sd = 256
formulae = Formulae()
builder = Builder(backend=CPU(formulae), n_sd=n_sd, environment=env)
builder.add_dynamic(AmbientThermodynamics())
builder.add_dynamic(Condensation())
r_dry, specific_concentration = spectral_sampling.Logarithmic(spectrum).sample(n_sd)
v_dry = formulae.trivia.volume(radius=r_dry)
r_wet = equilibrate_wet_radii(r_dry=r_dry, environment=env, kappa_times_dry_volume=kappa * v_dry)
attributes = {
'multiplicity': discretise_multiplicities(specific_concentration * env.mass_of_dry_air),
'dry volume': v_dry,
'kappa times dry volume': kappa * v_dry,
'volume': formulae.trivia.volume(radius=r_wet)
}
particulator = builder.build(attributes, products=[
products.PeakSupersaturation(name='S_max', unit='%'),
products.EffectiveRadius(name='r_eff', unit='um', radius_range=cloud_range),
products.ParticleConcentration(name='n_c_cm3', unit='cm^-3', radius_range=cloud_range),
products.WaterMixingRatio(name='liquid water mixing ratio', unit='g/kg', radius_range=cloud_range),
products.ParcelDisplacement(name='z')
])
cell_id = 0
output = {product.name: [product.get()[cell_id]] for product in particulator.products.values()}
for step in range(output_points):
particulator.run(steps=output_interval)
for product in particulator.products.values():
output[product.name].append(product.get()[cell_id])
fig, axs = pyplot.subplots(1, len(particulator.products) - 1, sharey="all")
for i, (key, product) in enumerate(particulator.products.items()):
if key != 'z':
axs[i].plot(output[key], output['z'], marker='.')
axs[i].set_title(product.name)
axs[i].set_xlabel(product.unit)
axs[i].grid()
axs[0].set_ylabel(particulator.products['z'].unit)
pyplot.savefig('parcel.svg')
```
</details>
The resultant plot (generated with the Matlab code) looks as follows:
## Contributing, reporting issues, seeking support
#### Our technologicial stack:
[](https://www.python.org/)
[](https://numba.pydata.org)
[](https://llvm.org)
[](https://pypi.org/project/ThrustRTC/)
[](https://numpy.org/)
[](https://pytest.org/)
[](https://colab.research.google.com/)
[](https://codecov.io/)
[](https://pypi.org/)
[](https://github.com/features/actions)
[](https://jupyter.org/)
[](https:///)
Submitting new code to the project, please preferably use [GitHub pull requests](https://github.com/open-atmos/PySDM/pulls) - it helps to keep record of code authorship,
track and archive the code review workflow and allows to benefit
from the continuous integration setup which automates execution of tests
with the newly added code.
Code contributions are assumed to imply transfer of copyright.
Should there be a need to make an exception, please indicate it when creating
a pull request or contributing code in any other way. In any case,
the license of the contributed code must be compatible with GPL v3.
Developing the code, we follow [The Way of Python](https://www.python.org/dev/peps/pep-0020/) and
the [KISS principle](https://en.wikipedia.org/wiki/KISS_principle).
The codebase has greatly benefited from [PyCharm code inspections](https://www.jetbrains.com/help/pycharm/code-inspection.html)
and [Pylint](https://pylint.org), [Black](https://black.readthedocs.io/en/stable/) and [isort](https://pycqa.github.io/isort/)
code analysis (which are all part of the CI workflows).
We also use [pre-commit hooks](https://pre-commit.com).
In our case, the hooks modify files and re-format them.
The pre-commit hooks can be run locally, and then the resultant changes need to be staged before committing.
To set up the hooks locally, install pre-commit via `pip install pre-commit` and
set up the git hooks via `pre-commit install` (this needs to be done every time you clone the project).
To run all pre-commit hooks, run `pre-commit run --all-files`.
The `.pre-commit-config.yaml` file can be modified in case new hooks are to be added or
existing ones need to be altered.
Further hints addressed at PySDM developers are maintained in the [open-atmos/python-dev-hints Wiki](https://github.com/open-atmos/python-dev-hints/wiki).
Issues regarding any incorrect, unintuitive or undocumented bahaviour of
PySDM are best to be reported on the [GitHub issue tracker](https://github.com/open-atmos/PySDM/issues/new).
Feature requests are recorded in the "Ideas..." [PySDM wiki page](https://github.com/open-atmos/PySDM/wiki/Ideas-for-new-features-and-examples).
We encourage to use the [GitHub Discussions](https://github.com/open-atmos/PySDM/discussions) feature
(rather than the issue tracker) for seeking support in understanding, using and extending PySDM code.
We look forward to your contributions and feedback.
## Credits:
The development and maintenance of PySDM is led by [Sylwester Arabas](https://github.com/slayoo/).
[Piotr Bartman](https://github.com/piotrbartman/) had been the architect and main developer
of technological solutions in PySDM.
PySDM includes contributions from researchers
from [Jagiellonian University](https://en.uj.edu.pl/en) departments of computer science, physics and chemistry;
from [Caltech's Climate Modelling Alliance](https://clima.caltech.edu/),
from [University of Warsaw](https://en.uw.edu.pl/) (dept. physics), and
from [AGH University of Krakow](https://agh.edu.pl/en) (dept. physics \& applied computer science) where release maintenance takes place currently.
Development of PySDM had been initially supported by the EU through a grant of the
[Foundation for Polish Science](https://www.fnp.org.pl/)) (grant no. POIR.04.04.00-00-5E1C/18)
realised at the [Jagiellonian University](https://en.uj.edu.pl/en).
The immersion freezing support in PySDM was developed with support from the
US Department of Energy [Atmospheric System Research](https://asr.science.energy.gov/) programme
through a grant (no. DE-SC0021034) realised at the
[University of Illinois at Urbana-Champaign](https://illinois.edu/).
Development of isotopic fractionation representation and mixed-phase support is carried out with support from
the [Polish National Science Centre](https://ncn.gov.pl/en) (grant no. 2020/39/D/ST10/01220).
copyright: [Jagiellonian University](https://en.uj.edu.pl/en) (2019-2023) & [AGH University of Krakow](https://agh.edu.pl/en) (2023-...)
licence: [GPL v3](https://www.gnu.org/licenses/gpl-3.0.html)
## Related resources and open-source projects
### SDM patents (some expired, some withdrawn):
- https://patents.google.com/patent/US7756693B2
- https://patents.google.com/patent/EP1847939A3
- https://patents.google.com/patent/JP4742387B2
- https://patents.google.com/patent/CN101059821B
### Other SDM implementations:
- SCALE-SDM (Fortran):
https://github.com/Shima-Lab/SCALE-SDM_BOMEX_Sato2018/blob/master/contrib/SDM/sdm_coalescence.f90
- Pencil Code (Fortran):
https://github.com/pencil-code/pencil-code/blob/master/src/particles_coagulation.f90
- PALM LES (Fortran):
https://palm.muk.uni-hannover.de/trac/browser/palm/trunk/SOURCE/lagrangian_particle_model_mod.f90
- libcloudph++ (C++):
https://github.com/igfuw/libcloudphxx/blob/master/src/impl/particles_impl_coal.ipp
- LCM1D (Python)
https://github.com/SimonUnterstrasser/ColumnModel
- superdroplet (Cython/Numba/C++11/Fortran 2008/Julia)
https://github.com/darothen/superdroplet
- NTLP (FORTRAN)
https://github.com/Folca/NTLP/blob/SuperDroplet/les.F
- CLEO (C++)
https://yoctoyotta1024.github.io/CLEO/
### non-SDM probabilistic particle-based coagulation solvers
- PartMC (Fortran):
https://github.com/compdyn/partmc
### Python models with discrete-particle (moving-sectional) representation of particle size spectrum
- pyrcel: https://github.com/darothen/pyrcel
- PyBox: https://github.com/loftytopping/PyBox
- py-cloud-parcel-model: https://github.com/emmasimp/py-cloud-parcel-model
Raw data
{
"_id": null,
"home_page": "https://github.com/open-atmos/PySDM",
"name": "PySDM",
"maintainer": null,
"docs_url": null,
"requires_python": null,
"maintainer_email": null,
"keywords": "physics-simulation, monte-carlo-simulation, gpu-computing, atmospheric-modelling, particle-system, numba, thrust, nvrtc, pint, atmospheric-physics",
"author": "https://github.com/open-atmos/PySDM/graphs/contributors",
"author_email": "sylwester.arabas@agh.edu.pl",
"download_url": "https://files.pythonhosted.org/packages/72/3b/18f2304825f395f4aaad6d6414fab3a702384fc29d9fb499d9bd86f8641b/PySDM-2.52.tar.gz",
"platform": null,
"description": "# PySDM\n\n[](https://www.python.org/)\n[](https://numba.pydata.org)\n[](https://pypi.org/project/ThrustRTC/)\n[](https://en.wikipedia.org/wiki/Linux)\n[](https://en.wikipedia.org/wiki/macOS)\n[](https://en.wikipedia.org/wiki/Windows)\n[](https://jupyter.org/)\n[](https://github.com/open-atmos/PySDM/graphs/commit-activity)\n[](https://www.openhub.net/p/atmos-cloud-sim-uj-PySDM)\n[](https://joss.theoj.org/papers/62cad07440b941f73f57d187df1aa6e9)\n[](https://zenodo.org/badge/latestdoi/199064632) \n[](https://www.fnp.org.pl/en/)\n[](https://www.ncn.gov.pl/?language=en)\n[](https://asr.science.energy.gov/)\n\n[](https://www.gnu.org/licenses/gpl-3.0.html)\n\n[](https://github.com/open-atmos/PySDM/actions)\n[](https://ci.appveyor.com/project/slayoo/pysdm/branch/main)\n[](https://app.codecov.io/gh/open-atmos/PySDM) \n[](https://pypi.org/project/PySDM)\n[](https://open-atmos.github.io/PySDM/)\n\nPySDM is a package for simulating the dynamics of population of particles. \nIt is intended to serve as a building block for simulation systems modelling\n fluid flows involving a dispersed phase,\n with PySDM being responsible for representation of the dispersed phase.\nCurrently, the development is focused on atmospheric cloud physics\n applications, in particular on modelling the dynamics of particles immersed in moist air \n using the particle-based (a.k.a. super-droplet) approach \n to represent aerosol/cloud/rain microphysics.\nThe package features a Pythonic high-performance implementation of the \n Super-Droplet Method (SDM) Monte-Carlo algorithm for representing collisional growth \n ([Shima et al. 2009](https://rmets.onlinelibrary.wiley.com/doi/abs/10.1002/qj.441)), hence the name. \n\nThere is a growing set of example Jupyter notebooks exemplifying how to perform \n various types of calculations and simulations using PySDM.\nMost of the example notebooks reproduce results and plot from literature, see below for \n a list of examples and links to the notebooks (which can be either executed or viewed \n \"in the cloud\").\n\nThere are also a growing set of [tutorials](https://github.com/open-atmos/PySDM/tree/main/tutorials), also in the form of Jupyter notebooks.\nThese tutorials are intended for teaching purposes and include short explanations of cloud microphysical \n concepts paired with widgets for running interactive simulations using PySDM.\nEach tutorial also comes with a set of questions at the end that can be used as homework problems.\nLike the examples, these tutorials can be executed or viewed \"in the cloud\" making it an especially \n easy way for students to get started.\n\nPySDM has two alternative parallel number-crunching backends \n available: multi-threaded CPU backend based on [Numba](http://numba.pydata.org/) \n and GPU-resident backend built on top of [ThrustRTC](https://pypi.org/project/ThrustRTC/).\nThe [`Numba`](https://open-atmos.github.io/PySDM/PySDM/backends/numba/numba.html) backend (aliased ``CPU``) features multi-threaded parallelism for \n multi-core CPUs, it uses the just-in-time compilation technique based on the LLVM infrastructure.\nThe [`ThrustRTC`](https://open-atmos.github.io/PySDM/PySDM/backends/thrustRTC/thrustRTC.html) backend (aliased ``GPU``) offers GPU-resident operation of PySDM\n leveraging the [SIMT](https://en.wikipedia.org/wiki/Single_instruction,_multiple_threads) \n parallelisation model. \nUsing the ``GPU`` backend requires nVidia hardware and [CUDA driver](https://developer.nvidia.com/cuda-downloads).\n\nFor an overview of PySDM features (and the preferred way to cite PySDM in papers), please refer to our JOSS papers:\n- [Bartman et al. 2022](https://doi.org/10.21105/joss.03219) (PySDM v1).\n- [de Jong, Singer et al. 2023](https://doi.org/10.21105/joss.04968) (PySDM v2).\n\nPySDM includes an extension of the SDM scheme to represent collisional breakup described in [de Jong, Mackay et al. 2023](10.5194/gmd-16-4193-2023). \nFor a list of talks and other materials on PySDM, see the [project wiki](https://github.com/open-atmos/PySDM/wiki).\n\nA [pdoc-generated](https://pdoc3.github.io/pdoc) documentation of PySDM public API is maintained at: [https://open-atmos.github.io/PySDM](https://open-atmos.github.io/PySDM) \n\n## Example Jupyter notebooks (reproducing results from literature):\n\nSee [PySDM-examples README](https://github.com/open-atmos/PySDM/blob/main/examples/README.md).\n\n\n\n## Dependencies and Installation\n\nPySDM dependencies are: [Numpy](https://numpy.org/), [Numba](http://numba.pydata.org/), [SciPy](https://scipy.org/), \n[Pint](https://pint.readthedocs.io/), [chempy](https://pypi.org/project/chempy/), \n[pyevtk](https://pypi.org/project/pyevtk/),\n[ThrustRTC](https://fynv.github.io/ThrustRTC/) and [CURandRTC](https://github.com/fynv/CURandRTC).\n\nTo install PySDM using ``pip``, use: ``pip install PySDM`` \n(or ``pip install git+https://github.com/open-atmos/PySDM.git`` to get updates\nbeyond the latest release).\n\nConda users may use ``pip`` as well, see the [Installing non-conda packages](https://docs.conda.io/projects/conda/en/latest/user-guide/tasks/manage-pkgs.html#installing-non-conda-packages) section in the conda docs. Dependencies of PySDM are available at the following conda channels:\n- numba: [numba](https://anaconda.org/numba/numba)\n- conda-forge: [pyevtk](https://anaconda.org/conda-forge/pyevtk), [pint](https://anaconda.org/conda-forge/pint) and []()\n- fyplus: [ThrustRTC](https://anaconda.org/fyplus/thrustrtc), [CURandRTC](https://anaconda.org/fyplus/curandrtc)\n- bjodah: [chempy](https://anaconda.org/bjodah/chempy)\n- nvidia: [cudatoolkit](https://anaconda.org/nvidia/cudatoolkit)\n\nFor development purposes, we suggest cloning the repository and installing it using ``pip -e``.\nTest-time dependencies can be installed with ``pip -e .[tests]``.\n\nPySDM examples constitute the [``PySDM-examples``](https://github.com/open-atmos/PySDM-examples) package.\nThe examples have additional dependencies listed in [``PySDM_examples`` package ``setup.py``](https://github.com/open-atmos/PySDM/blob/main/examples/setup.py) file.\nRunning the example Jupyter notebooks requires the ``PySDM_examples`` package to be installed.\nThe suggested install and launch steps are:\n```\ngit clone https://github.com/open-atmos/PySDM.git\npip install -e PySDM\npip install -e PySDM/examples\njupyter-notebook PySDM/examples/PySDM_examples\n```\nAlternatively, one can also install the examples package from pypi.org by \nusing ``pip install PySDM-examples`` (note that this does not apply to notebooks itself,\nonly the supporting .py files).\n\n## Submodule organization\n```mermaid\nmindmap\n root((PySDM))\n Builder\n Formulae\n Particulator\n ((attributes))\n (physics)\n DryVolume: ExtensiveAttribute\n Kappa: DerivedAttribute\n ...\n (chemistry)\n Acidity\n ...\n (...)\n ((backends))\n CPU\n GPU\n ((dynamics))\n AqueousChemistry\n Collision\n Condensation\n ...\n ((environments))\n Box\n Parcel\n Kinematic2D\n ...\n ((initialisation))\n (spectra)\n Lognormal\n Exponential\n ...\n (sampling)\n (spectral_sampling)\n ConstantMultiplicity\n UniformRandom\n Logarithmic\n ...\n (...) \n (...)\n ((physics))\n (hygroscopicity)\n KappaKoehler\n ...\n (condensation_coordinate)\n Volume\n VolumeLogarithm\n (...)\n ((products))\n (size_spectral)\n EffectiveRadius\n WaterMixingRatio\n ...\n (ambient_thermodynamics)\n AmbientRelativeHumidity\n ...\n (...)\n```\n\n## Hello-world coalescence example in Python, Julia and Matlab\n\nIn order to depict the PySDM API with a practical example, the following\n listings provide sample code roughly reproducing the \n Figure 2 from [Shima et al. 2009 paper](http://doi.org/10.1002/qj.441)\n using PySDM from Python, Julia and Matlab.\nIt is a [`Coalescence`](https://open-atmos.github.io/PySDM/PySDM/dynamics/coalescence.html)-only set-up in which the initial particle size \n spectrum is [`Exponential`](https://open-atmos.github.io/PySDM/PySDM/initialisation/spectra.html#PySDM.initialisation.spectra.Exponential) and is deterministically sampled to match\n the condition of each super-droplet having equal initial multiplicity:\n<details>\n<summary>Julia (click to expand)</summary>\n\n```Julia\nusing Pkg\nPkg.add(\"PyCall\")\nPkg.add(\"Plots\")\nPkg.add(\"PlotlyJS\")\n\nusing PyCall\nsi = pyimport(\"PySDM.physics\").si\nConstantMultiplicity = pyimport(\"PySDM.initialisation.sampling.spectral_sampling\").ConstantMultiplicity\nExponential = pyimport(\"PySDM.initialisation.spectra\").Exponential\n\nn_sd = 2^15\ninitial_spectrum = Exponential(norm_factor=8.39e12, scale=1.19e5 * si.um^3)\nattributes = Dict()\nattributes[\"volume\"], attributes[\"multiplicity\"] = ConstantMultiplicity(spectrum=initial_spectrum).sample(n_sd)\n```\n</details>\n<details>\n<summary>Matlab (click to expand)</summary>\n\n```Matlab\nsi = py.importlib.import_module('PySDM.physics').si;\nConstantMultiplicity = py.importlib.import_module('PySDM.initialisation.sampling.spectral_sampling').ConstantMultiplicity;\nExponential = py.importlib.import_module('PySDM.initialisation.spectra').Exponential;\n\nn_sd = 2^15;\ninitial_spectrum = Exponential(pyargs(...\n 'norm_factor', 8.39e12, ...\n 'scale', 1.19e5 * si.um ^ 3 ...\n));\ntmp = ConstantMultiplicity(initial_spectrum).sample(int32(n_sd));\nattributes = py.dict(pyargs('volume', tmp{1}, 'multiplicity', tmp{2}));\n```\n</details>\n<details open>\n<summary>Python (click to expand)</summary>\n\n```Python\nfrom PySDM.physics import si\nfrom PySDM.initialisation.sampling.spectral_sampling import ConstantMultiplicity\nfrom PySDM.initialisation.spectra.exponential import Exponential\n\nn_sd = 2 ** 15\ninitial_spectrum = Exponential(norm_factor=8.39e12, scale=1.19e5 * si.um ** 3)\nattributes = {}\nattributes['volume'], attributes['multiplicity'] = ConstantMultiplicity(initial_spectrum).sample(n_sd)\n```\n</details>\n\nThe key element of the PySDM interface is the [``Particulator``](https://open-atmos.github.io/PySDM/PySDM/particulator.html) \n class instances of which are used to manage the system state and control the simulation.\nInstantiation of the [``Particulator``](https://open-atmos.github.io/PySDM/PySDM/particulator.html) class is handled by the [``Builder``](https://open-atmos.github.io/PySDM/PySDM/builder.html)\n as exemplified below:\n<details>\n<summary>Julia (click to expand)</summary>\n\n```Julia\nBuilder = pyimport(\"PySDM\").Builder\nBox = pyimport(\"PySDM.environments\").Box\nCoalescence = pyimport(\"PySDM.dynamics\").Coalescence\nGolovin = pyimport(\"PySDM.dynamics.collisions.collision_kernels\").Golovin\nCPU = pyimport(\"PySDM.backends\").CPU\nParticleVolumeVersusRadiusLogarithmSpectrum = pyimport(\"PySDM.products\").ParticleVolumeVersusRadiusLogarithmSpectrum\n\nradius_bins_edges = 10 .^ range(log10(10*si.um), log10(5e3*si.um), length=32) \n\nenv = Box(dt=1 * si.s, dv=1e6 * si.m^3)\nbuilder = Builder(n_sd=n_sd, backend=CPU(), environment=env)\nbuilder.add_dynamic(Coalescence(collision_kernel=Golovin(b=1.5e3 / si.s)))\nproducts = [ParticleVolumeVersusRadiusLogarithmSpectrum(radius_bins_edges=radius_bins_edges, name=\"dv/dlnr\")] \nparticulator = builder.build(attributes, products)\n```\n</details>\n<details>\n<summary>Matlab (click to expand)</summary>\n\n```Matlab\nBuilder = py.importlib.import_module('PySDM').Builder;\nBox = py.importlib.import_module('PySDM.environments').Box;\nCoalescence = py.importlib.import_module('PySDM.dynamics').Coalescence;\nGolovin = py.importlib.import_module('PySDM.dynamics.collisions.collision_kernels').Golovin;\nCPU = py.importlib.import_module('PySDM.backends').CPU;\nParticleVolumeVersusRadiusLogarithmSpectrum = py.importlib.import_module('PySDM.products').ParticleVolumeVersusRadiusLogarithmSpectrum;\n\nradius_bins_edges = logspace(log10(10 * si.um), log10(5e3 * si.um), 32);\n\nenv = Box(pyargs('dt', 1 * si.s, 'dv', 1e6 * si.m ^ 3));\nbuilder = Builder(pyargs('n_sd', int32(n_sd), 'backend', CPU(), 'environment', env));\nbuilder.add_dynamic(Coalescence(pyargs('collision_kernel', Golovin(1.5e3 / si.s))));\nproducts = py.list({ ParticleVolumeVersusRadiusLogarithmSpectrum(pyargs( ...\n 'radius_bins_edges', py.numpy.array(radius_bins_edges), ...\n 'name', 'dv/dlnr' ...\n)) });\nparticulator = builder.build(attributes, products);\n```\n</details>\n<details open>\n<summary>Python (click to expand)</summary>\n\n```Python\nimport numpy as np\nfrom PySDM import Builder\nfrom PySDM.environments import Box\nfrom PySDM.dynamics import Coalescence\nfrom PySDM.dynamics.collisions.collision_kernels import Golovin\nfrom PySDM.backends import CPU\nfrom PySDM.products import ParticleVolumeVersusRadiusLogarithmSpectrum\n\nradius_bins_edges = np.logspace(np.log10(10 * si.um), np.log10(5e3 * si.um), num=32)\n\nenv = Box(dt=1 * si.s, dv=1e6 * si.m ** 3)\nbuilder = Builder(n_sd=n_sd, backend=CPU(), environment=env)\nbuilder.add_dynamic(Coalescence(collision_kernel=Golovin(b=1.5e3 / si.s)))\nproducts = [ParticleVolumeVersusRadiusLogarithmSpectrum(radius_bins_edges=radius_bins_edges, name='dv/dlnr')]\nparticulator = builder.build(attributes, products)\n```\n</details>\n\nThe ``backend`` argument may be set to ``CPU`` or ``GPU``\n what translates to choosing the multi-threaded backend or the \n GPU-resident computation mode, respectively.\nThe employed [`Box`](https://open-atmos.github.io/PySDM/PySDM/environments/box.html) environment corresponds to a zero-dimensional framework\n (particle positions are not considered).\nThe vectors of particle multiplicities ``n`` and particle volumes ``v`` are\n used to initialise super-droplet attributes.\nThe [`Coalescence`](https://open-atmos.github.io/PySDM/PySDM/dynamics/coalescence.html)\n Monte-Carlo algorithm (Super Droplet Method) is registered as the only\n dynamic in the system.\nFinally, the [`build()`](https://open-atmos.github.io/PySDM/PySDM/builder.html#PySDM.builder.Builder.build) method is used to obtain an instance\n of [`Particulator`](https://open-atmos.github.io/PySDM/PySDM/particulator.html#PySDM.particulator.Particulator) which can then be used to control time-stepping and\n access simulation state.\n\nThe [`run(nt)`](https://open-atmos.github.io/PySDM/PySDM/particulator.html#PySDM.particuparticulatorr.Particulator.run) method advances the simulation by ``nt`` timesteps.\nIn the listing below, its usage is interleaved with plotting logic\n which displays a histogram of particle mass distribution \n at selected timesteps:\n<details>\n<summary>Julia (click to expand)</summary>\n\n```Julia\nusing Plots; plotlyjs()\n\nfor step = 0:1200:3600\n particulator.run(step - particulator.n_steps)\n plot!(\n radius_bins_edges[1:end-1] / si.um,\n particulator.formulae.particle_shape_and_density.volume_to_mass(\n particulator.products[\"dv/dlnr\"].get()[:]\n )/ si.g,\n linetype=:steppost,\n xaxis=:log,\n xlabel=\"particle radius [\u00b5m]\",\n ylabel=\"dm/dlnr [g/m^3/(unit dr/r)]\",\n label=\"t = $step s\"\n ) \nend\nsavefig(\"plot.svg\")\n```\n</details>\n<details>\n<summary>Matlab (click to expand)</summary>\n\n```Matlab\nfor step = 0:1200:3600\n particulator.run(int32(step - particulator.n_steps));\n x = radius_bins_edges / si.um;\n y = particulator.formulae.particle_shape_and_density.volume_to_mass( ...\n particulator.products{\"dv/dlnr\"}.get() ...\n ) / si.g;\n stairs(...\n x(1:end-1), ... \n double(py.array.array('d',py.numpy.nditer(y))), ...\n 'DisplayName', sprintf(\"t = %d s\", step) ...\n );\n hold on\nend\nhold off\nset(gca,'XScale','log');\nxlabel('particle radius [\u00b5m]')\nylabel(\"dm/dlnr [g/m^3/(unit dr/r)]\")\nlegend()\n```\n</details>\n<details open>\n<summary>Python (click to expand)</summary>\n\n```Python\nfrom matplotlib import pyplot\n\nfor step in [0, 1200, 2400, 3600]:\n particulator.run(step - particulator.n_steps)\n pyplot.step(\n x=radius_bins_edges[:-1] / si.um,\n y=particulator.formulae.particle_shape_and_density.volume_to_mass(\n particulator.products['dv/dlnr'].get()[0]\n ) / si.g,\n where='post', label=f\"t = {step}s\"\n )\n\npyplot.xscale('log')\npyplot.xlabel('particle radius [\u00b5m]')\npyplot.ylabel(\"dm/dlnr [g/m$^3$/(unit dr/r)]\")\npyplot.legend()\npyplot.savefig('readme.png')\n```\n</details>\n\nThe resultant plot (generated with the Python code) looks as follows:\n\n\n\nThe component submodules used to create this simulation are visualized below:\n```mermaid\n graph\n COAL[\":Coalescence\"] --->|passed as arg to| BUILDER_ADD_DYN([\"Builder.add_dynamic()\"])\n BUILDER_INSTANCE[\"builder :Builder\"] -...-|has a method| BUILDER_BUILD([\"Builder.build()\"])\n ATTRIBUTES[attributes: dict] -->|passed as arg to| BUILDER_BUILD\n N_SD[\"n_sd :int\"] ---->|passed as arg to| BUILDER_INIT\n BUILDER_INIT([\"Builder.__init__()\"]) --->|instantiates| BUILDER_INSTANCE\n BUILDER_INSTANCE -..-|has a method| BUILDER_ADD_DYN([\"Builder.add_dynamic()\"])\n ENV_INIT([\"Box.__init__()\"]) -->|instantiates| ENV\n DT[dt :float] -->|passed as arg to| ENV_INIT\n DV[dv :float] -->|passed as arg to| ENV_INIT\n ENV[\":Box\"] -->|passed as arg to| BUILDER_INIT\n B[\"b: float\"] --->|passed as arg to| KERNEL_INIT([\"Golovin.__init__()\"])\n KERNEL_INIT -->|instantiates| KERNEL\n KERNEL[collision_kernel: Golovin] -->|passed as arg to| COAL_INIT([\"Coalesncence.__init__()\"])\n COAL_INIT -->|instantiates| COAL\n PRODUCTS[products: list] ----->|passed as arg to| BUILDER_BUILD\n NORM_FACTOR[norm_factor: float]-->|passed as arg to| EXP_INIT\n SCALE[scale: float]-->|passed as arg to| EXP_INIT\n EXP_INIT([\"Exponential.__init__()\"]) -->|instantiates| IS\n IS[\"initial_spectrum :Exponential\"] -->|passed as arg to| CM_INIT\n CM_INIT([\"ConstantMultiplicity.__init__()\"]) -->|instantiates| CM_INSTANCE\n CM_INSTANCE[\":ConstantMultiplicity\"] -.-|has a method| SAMPLE\n SAMPLE([\"ConstantMultiplicity.sample()\"]) -->|returns| n\n SAMPLE -->|returns| volume\n n -->|added as element of| ATTRIBUTES\n PARTICULATOR_INSTANCE -.-|has a method| PARTICULATOR_RUN([\"Particulator.run()\"])\n volume -->|added as element of| ATTRIBUTES\n BUILDER_BUILD -->|returns| PARTICULATOR_INSTANCE[\"particulator :Particulator\"]\n PARTICULATOR_INSTANCE -.-|has a field| PARTICULATOR_PROD([\"Particulator.products:dict\"])\n BACKEND_INSTANCE[\"backend :CPU\"] ---->|passed as arg to| BUILDER_INIT\n PRODUCTS -.-|accessible via| PARTICULATOR_PROD\n NP_LOGSPACE([\"np.logspace()\"]) -->|returns| EDGES \n EDGES[radius_bins_edges: np.ndarray] -->|passed as arg to| SPECTRUM_INIT\n SPECTRUM_INIT[\"ParticleVolumeVersusRadiusLogarithmSpectrum.__init__()\"] -->|instantiates| SPECTRUM\n SPECTRUM[\":ParticleVolumeVersusRadiusLogarithmSpectrum\"] -->|added as element of| PRODUCTS\n\n click COAL \"https://open-atmos.github.io/PySDM/PySDM/dynamics/collisions/collision.html\"\n click BUILDER_INSTANCE \"https://open-atmos.github.io/PySDM/PySDM/builder.html\"\n click BUILDER_INIT \"https://open-atmos.github.io/PySDM/PySDM/builder.html\"\n click BUILDER_ADD_DYN \"https://open-atmos.github.io/PySDM/PySDM/builder.html\"\n click ENV_INIT \"https://open-atmos.github.io/PySDM/PySDM/environments/index.html\"\n click ENV \"https://open-atmos.github.io/PySDM/PySDM/environments/index.html\"\n click KERNEL_INIT \"https://open-atmos.github.io/PySDM/PySDM/dynamics/collisions/collision_kernels/index.html\"\n click KERNEL \"https://open-atmos.github.io/PySDM/PySDM/dynamics/collisions/collision_kernels/index.html\"\n click EXP_INIT \"https://open-atmos.github.io/PySDM/PySDM/initialisation/spectra/index.html\"\n click IS \"https://open-atmos.github.io/PySDM/PySDM/initialisation/spectra/index.html\"\n click CM_INIT \"https://open-atmos.github.io/PySDM/PySDM/initialisation/sampling/spectral_sampling.html\"\n click CM_INSTANCE \"https://open-atmos.github.io/PySDM/PySDM/initialisation/sampling/spectral_sampling.html\"\n click SAMPLE \"https://open-atmos.github.io/PySDM/PySDM/initialisation/sampling/spectral_sampling.html\"\n click PARTICULATOR_INSTANCE \"https://open-atmos.github.io/PySDM/PySDM/particulator.html\"\n click BACKEND_INSTANCE \"https://open-atmos.github.io/PySDM/PySDM/backends/numba.html\"\n click BUILDER_BUILD \"https://open-atmos.github.io/PySDM/PySDM/builder.html\"\n click NP_LOGSPACE \"https://numpy.org/doc/stable/reference/generated/numpy.logspace.html\"\n click SPECTRUM_INIT \"https://open-atmos.github.io/PySDM/PySDM/products/size_spectral/particle_volume_versus_radius_logarithm_spectrum.html\"\n click SPECTRUM \"https://open-atmos.github.io/PySDM/PySDM/products/size_spectral/particle_volume_versus_radius_logarithm_spectrum.html\"\n```\n\n## Hello-world condensation example in Python, Julia and Matlab\n\nIn the following example, a condensation-only setup is used with the adiabatic \n[`Parcel`](https://open-atmos.github.io/PySDM/PySDM/environments/parcel.html) environment.\nAn initial [`Lognormal`](https://open-atmos.github.io/PySDM/PySDM/initialisation/spectra.html#PySDM.initialisation.spectra.Lognormal)\nspectrum of dry aerosol particles is first initialised to equilibrium wet size for the given\ninitial humidity. \nSubsequent particle growth due to [`Condensation`](https://open-atmos.github.io/PySDM/PySDM/dynamics/condensation.html) of water vapour (coupled with the release of latent heat)\ncauses a subset of particles to activate into cloud droplets.\nResults of the simulation are plotted against vertical \n[`ParcelDisplacement`](https://open-atmos.github.io/PySDM/PySDM/products/housekeeping/parcel_displacement.html)\nand depict the evolution of \n[`PeakSupersaturation`](https://open-atmos.github.io/PySDM/PySDM/products/condensation/peak_supersaturation.html), \n[`EffectiveRadius`](https://open-atmos.github.io/PySDM/PySDM/products/size_spectral/effective_radius.html), \n[`ParticleConcentration`](https://open-atmos.github.io/PySDM/PySDM/products/size_spectral/particle_concentration.html#PySDM.products.particles_concentration.ParticleConcentration) \nand the \n[`WaterMixingRatio `](https://open-atmos.github.io/PySDM/PySDM/products/size_spectral/water_mixing_ratio.html).\n\n<details>\n<summary>Julia (click to expand)</summary>\n\n```Julia\nusing PyCall\nusing Plots; plotlyjs()\nsi = pyimport(\"PySDM.physics\").si\nspectral_sampling = pyimport(\"PySDM.initialisation.sampling\").spectral_sampling\ndiscretise_multiplicities = pyimport(\"PySDM.initialisation\").discretise_multiplicities\nLognormal = pyimport(\"PySDM.initialisation.spectra\").Lognormal\nequilibrate_wet_radii = pyimport(\"PySDM.initialisation\").equilibrate_wet_radii\nCPU = pyimport(\"PySDM.backends\").CPU\nAmbientThermodynamics = pyimport(\"PySDM.dynamics\").AmbientThermodynamics\nCondensation = pyimport(\"PySDM.dynamics\").Condensation\nParcel = pyimport(\"PySDM.environments\").Parcel\nBuilder = pyimport(\"PySDM\").Builder\nFormulae = pyimport(\"PySDM\").Formulae\nproducts = pyimport(\"PySDM.products\")\n\nenv = Parcel(\n dt=.25 * si.s,\n mass_of_dry_air=1e3 * si.kg,\n p0=1122 * si.hPa,\n initial_water_vapour_mixing_ratio=20 * si.g / si.kg,\n T0=300 * si.K,\n w= 2.5 * si.m / si.s\n)\nspectrum = Lognormal(norm_factor=1e4/si.mg, m_mode=50*si.nm, s_geom=1.4)\nkappa = .5 * si.dimensionless\ncloud_range = (.5 * si.um, 25 * si.um)\noutput_interval = 4\noutput_points = 40\nn_sd = 256\n\nformulae = Formulae()\nbuilder = Builder(backend=CPU(formulae), n_sd=n_sd, environment=env)\nbuilder.add_dynamic(AmbientThermodynamics())\nbuilder.add_dynamic(Condensation())\n\nr_dry, specific_concentration = spectral_sampling.Logarithmic(spectrum).sample(n_sd)\nv_dry = formulae.trivia.volume(radius=r_dry)\nr_wet = equilibrate_wet_radii(r_dry=r_dry, environment=env, kappa_times_dry_volume=kappa * v_dry)\n\nattributes = Dict()\nattributes[\"multiplicity\"] = discretise_multiplicities(specific_concentration * env.mass_of_dry_air)\nattributes[\"dry volume\"] = v_dry\nattributes[\"kappa times dry volume\"] = kappa * v_dry\nattributes[\"volume\"] = formulae.trivia.volume(radius=r_wet) \n\nparticulator = builder.build(attributes, products=[\n products.PeakSupersaturation(name=\"S_max\", unit=\"%\"),\n products.EffectiveRadius(name=\"r_eff\", unit=\"um\", radius_range=cloud_range),\n products.ParticleConcentration(name=\"n_c_cm3\", unit=\"cm^-3\", radius_range=cloud_range),\n products.WaterMixingRatio(name=\"liquid water mixing ratio\", unit=\"g/kg\", radius_range=cloud_range),\n products.ParcelDisplacement(name=\"z\")\n])\n\ncell_id=1\noutput = Dict()\nfor (_, product) in particulator.products\n output[product.name] = Array{Float32}(undef, output_points+1)\n output[product.name][1] = product.get()[cell_id]\nend \n\nfor step = 2:output_points+1\n particulator.run(steps=output_interval)\n for (_, product) in particulator.products\n output[product.name][step] = product.get()[cell_id]\n end \nend \n\nplots = []\nylbl = particulator.products[\"z\"].unit\nfor (_, product) in particulator.products\n if product.name != \"z\"\n append!(plots, [plot(output[product.name], output[\"z\"], ylabel=ylbl, xlabel=product.unit, title=product.name)])\n end\n global ylbl = \"\"\nend\nplot(plots..., layout=(1, length(output)-1))\nsavefig(\"parcel.svg\")\n```\n</details>\n<details>\n<summary>Matlab (click to expand)</summary>\n\n```Matlab\nsi = py.importlib.import_module('PySDM.physics').si;\nspectral_sampling = py.importlib.import_module('PySDM.initialisation.sampling').spectral_sampling;\ndiscretise_multiplicities = py.importlib.import_module('PySDM.initialisation').discretise_multiplicities;\nLognormal = py.importlib.import_module('PySDM.initialisation.spectra').Lognormal;\nequilibrate_wet_radii = py.importlib.import_module('PySDM.initialisation').equilibrate_wet_radii;\nCPU = py.importlib.import_module('PySDM.backends').CPU;\nAmbientThermodynamics = py.importlib.import_module('PySDM.dynamics').AmbientThermodynamics;\nCondensation = py.importlib.import_module('PySDM.dynamics').Condensation;\nParcel = py.importlib.import_module('PySDM.environments').Parcel;\nBuilder = py.importlib.import_module('PySDM').Builder;\nFormulae = py.importlib.import_module('PySDM').Formulae;\nproducts = py.importlib.import_module('PySDM.products');\n\nenv = Parcel(pyargs( ...\n 'dt', .25 * si.s, ...\n 'mass_of_dry_air', 1e3 * si.kg, ...\n 'p0', 1122 * si.hPa, ...\n 'initial_water_vapour_mixing_ratio', 20 * si.g / si.kg, ...\n 'T0', 300 * si.K, ...\n 'w', 2.5 * si.m / si.s ...\n));\nspectrum = Lognormal(pyargs('norm_factor', 1e4/si.mg, 'm_mode', 50 * si.nm, 's_geom', 1.4));\nkappa = .5;\ncloud_range = py.tuple({.5 * si.um, 25 * si.um});\noutput_interval = 4;\noutput_points = 40;\nn_sd = 256;\n\nformulae = Formulae();\nbuilder = Builder(pyargs('backend', CPU(formulae), 'n_sd', int32(n_sd), 'environment', env));\nbuilder.add_dynamic(AmbientThermodynamics());\nbuilder.add_dynamic(Condensation());\n\ntmp = spectral_sampling.Logarithmic(spectrum).sample(int32(n_sd));\nr_dry = tmp{1};\nv_dry = formulae.trivia.volume(pyargs('radius', r_dry));\nspecific_concentration = tmp{2};\nr_wet = equilibrate_wet_radii(pyargs(...\n 'r_dry', r_dry, ...\n 'environment', env, ...\n 'kappa_times_dry_volume', kappa * v_dry...\n));\n\nattributes = py.dict(pyargs( ...\n 'multiplicity', discretise_multiplicities(specific_concentration * env.mass_of_dry_air), ...\n 'dry volume', v_dry, ...\n 'kappa times dry volume', kappa * v_dry, ... \n 'volume', formulae.trivia.volume(pyargs('radius', r_wet)) ...\n));\n\nparticulator = builder.build(attributes, py.list({ ...\n products.PeakSupersaturation(pyargs('name', 'S_max', 'unit', '%')), ...\n products.EffectiveRadius(pyargs('name', 'r_eff', 'unit', 'um', 'radius_range', cloud_range)), ...\n products.ParticleConcentration(pyargs('name', 'n_c_cm3', 'unit', 'cm^-3', 'radius_range', cloud_range)), ...\n products.WaterMixingRatio(pyargs('name', 'liquid water mixing ratio', 'unit', 'g/kg', 'radius_range', cloud_range)) ...\n products.ParcelDisplacement(pyargs('name', 'z')) ...\n}));\n\ncell_id = int32(0);\noutput_size = [output_points+1, length(py.list(particulator.products.keys()))];\noutput_types = repelem({'double'}, output_size(2));\noutput_names = [cellfun(@string, cell(py.list(particulator.products.keys())))];\noutput = table(...\n 'Size', output_size, ...\n 'VariableTypes', output_types, ...\n 'VariableNames', output_names ...\n);\nfor pykey = py.list(keys(particulator.products))\n get = py.getattr(particulator.products{pykey{1}}.get(), '__getitem__');\n key = string(pykey{1});\n output{1, key} = get(cell_id);\nend\n\nfor i=2:output_points+1\n particulator.run(pyargs('steps', int32(output_interval)));\n for pykey = py.list(keys(particulator.products))\n get = py.getattr(particulator.products{pykey{1}}.get(), '__getitem__');\n key = string(pykey{1});\n output{i, key} = get(cell_id);\n end\nend\n\ni=1;\nfor pykey = py.list(keys(particulator.products))\n product = particulator.products{pykey{1}};\n if string(product.name) ~= \"z\"\n subplot(1, width(output)-1, i);\n plot(output{:, string(pykey{1})}, output.z, '-o');\n title(string(product.name), 'Interpreter', 'none');\n xlabel(string(product.unit));\n end\n if i == 1\n ylabel(string(particulator.products{\"z\"}.unit));\n end\n i=i+1;\nend\nsaveas(gcf, \"parcel.png\");\n```\n</details>\n<details open>\n<summary>Python (click to expand)</summary>\n\n```Python\nfrom matplotlib import pyplot\nfrom PySDM.physics import si\nfrom PySDM.initialisation import discretise_multiplicities, equilibrate_wet_radii\nfrom PySDM.initialisation.spectra import Lognormal\nfrom PySDM.initialisation.sampling import spectral_sampling\nfrom PySDM.backends import CPU\nfrom PySDM.dynamics import AmbientThermodynamics, Condensation\nfrom PySDM.environments import Parcel\nfrom PySDM import Builder, Formulae, products\n\nenv = Parcel(\n dt=.25 * si.s,\n mass_of_dry_air=1e3 * si.kg,\n p0=1122 * si.hPa,\n initial_water_vapour_mixing_ratio=20 * si.g / si.kg,\n T0=300 * si.K,\n w=2.5 * si.m / si.s\n)\nspectrum = Lognormal(norm_factor=1e4 / si.mg, m_mode=50 * si.nm, s_geom=1.5)\nkappa = .5 * si.dimensionless\ncloud_range = (.5 * si.um, 25 * si.um)\noutput_interval = 4\noutput_points = 40\nn_sd = 256\n\nformulae = Formulae()\nbuilder = Builder(backend=CPU(formulae), n_sd=n_sd, environment=env)\nbuilder.add_dynamic(AmbientThermodynamics())\nbuilder.add_dynamic(Condensation())\n\nr_dry, specific_concentration = spectral_sampling.Logarithmic(spectrum).sample(n_sd)\nv_dry = formulae.trivia.volume(radius=r_dry)\nr_wet = equilibrate_wet_radii(r_dry=r_dry, environment=env, kappa_times_dry_volume=kappa * v_dry)\n\nattributes = {\n 'multiplicity': discretise_multiplicities(specific_concentration * env.mass_of_dry_air),\n 'dry volume': v_dry,\n 'kappa times dry volume': kappa * v_dry,\n 'volume': formulae.trivia.volume(radius=r_wet)\n}\n\nparticulator = builder.build(attributes, products=[\n products.PeakSupersaturation(name='S_max', unit='%'),\n products.EffectiveRadius(name='r_eff', unit='um', radius_range=cloud_range),\n products.ParticleConcentration(name='n_c_cm3', unit='cm^-3', radius_range=cloud_range),\n products.WaterMixingRatio(name='liquid water mixing ratio', unit='g/kg', radius_range=cloud_range),\n products.ParcelDisplacement(name='z')\n])\n\ncell_id = 0\noutput = {product.name: [product.get()[cell_id]] for product in particulator.products.values()}\n\nfor step in range(output_points):\n particulator.run(steps=output_interval)\n for product in particulator.products.values():\n output[product.name].append(product.get()[cell_id])\n\nfig, axs = pyplot.subplots(1, len(particulator.products) - 1, sharey=\"all\")\nfor i, (key, product) in enumerate(particulator.products.items()):\n if key != 'z':\n axs[i].plot(output[key], output['z'], marker='.')\n axs[i].set_title(product.name)\n axs[i].set_xlabel(product.unit)\n axs[i].grid()\naxs[0].set_ylabel(particulator.products['z'].unit)\npyplot.savefig('parcel.svg')\n```\n</details>\n\nThe resultant plot (generated with the Matlab code) looks as follows:\n\n\n\n## Contributing, reporting issues, seeking support \n\n#### Our technologicial stack: \n[](https://www.python.org/)\n[](https://numba.pydata.org)\n[](https://llvm.org)\n[](https://pypi.org/project/ThrustRTC/)\n[](https://numpy.org/)\n[](https://pytest.org/) \n[](https://colab.research.google.com/)\n[](https://codecov.io/)\n[](https://pypi.org/)\n[](https://github.com/features/actions)\n[](https://jupyter.org/)\n[](https:///)\n\nSubmitting new code to the project, please preferably use [GitHub pull requests](https://github.com/open-atmos/PySDM/pulls) - it helps to keep record of code authorship, \ntrack and archive the code review workflow and allows to benefit\nfrom the continuous integration setup which automates execution of tests \nwith the newly added code. \n\nCode contributions are assumed to imply transfer of copyright.\nShould there be a need to make an exception, please indicate it when creating\na pull request or contributing code in any other way. In any case, \nthe license of the contributed code must be compatible with GPL v3.\n\nDeveloping the code, we follow [The Way of Python](https://www.python.org/dev/peps/pep-0020/) and \nthe [KISS principle](https://en.wikipedia.org/wiki/KISS_principle).\nThe codebase has greatly benefited from [PyCharm code inspections](https://www.jetbrains.com/help/pycharm/code-inspection.html)\nand [Pylint](https://pylint.org), [Black](https://black.readthedocs.io/en/stable/) and [isort](https://pycqa.github.io/isort/)\ncode analysis (which are all part of the CI workflows).\n\nWe also use [pre-commit hooks](https://pre-commit.com). \nIn our case, the hooks modify files and re-format them.\nThe pre-commit hooks can be run locally, and then the resultant changes need to be staged before committing.\nTo set up the hooks locally, install pre-commit via `pip install pre-commit` and\nset up the git hooks via `pre-commit install` (this needs to be done every time you clone the project).\nTo run all pre-commit hooks, run `pre-commit run --all-files`.\nThe `.pre-commit-config.yaml` file can be modified in case new hooks are to be added or\n existing ones need to be altered. \n\nFurther hints addressed at PySDM developers are maintained in the [open-atmos/python-dev-hints Wiki](https://github.com/open-atmos/python-dev-hints/wiki).\n\nIssues regarding any incorrect, unintuitive or undocumented bahaviour of\nPySDM are best to be reported on the [GitHub issue tracker](https://github.com/open-atmos/PySDM/issues/new).\nFeature requests are recorded in the \"Ideas...\" [PySDM wiki page](https://github.com/open-atmos/PySDM/wiki/Ideas-for-new-features-and-examples).\n\nWe encourage to use the [GitHub Discussions](https://github.com/open-atmos/PySDM/discussions) feature\n(rather than the issue tracker) for seeking support in understanding, using and extending PySDM code.\n\nWe look forward to your contributions and feedback.\n\n## Credits:\n\nThe development and maintenance of PySDM is led by [Sylwester Arabas](https://github.com/slayoo/).\n[Piotr Bartman](https://github.com/piotrbartman/) had been the architect and main developer \nof technological solutions in PySDM. \nPySDM includes contributions from researchers \nfrom [Jagiellonian University](https://en.uj.edu.pl/en) departments of computer science, physics and chemistry;\nfrom [Caltech's Climate Modelling Alliance](https://clima.caltech.edu/),\nfrom [University of Warsaw](https://en.uw.edu.pl/) (dept. physics), and\nfrom [AGH University of Krakow](https://agh.edu.pl/en) (dept. physics \\& applied computer science) where release maintenance takes place currently.\n\nDevelopment of PySDM had been initially supported by the EU through a grant of the \n[Foundation for Polish Science](https://www.fnp.org.pl/)) (grant no. POIR.04.04.00-00-5E1C/18) \nrealised at the [Jagiellonian University](https://en.uj.edu.pl/en).\nThe immersion freezing support in PySDM was developed with support from the\nUS Department of Energy [Atmospheric System Research](https://asr.science.energy.gov/) programme\nthrough a grant (no. DE-SC0021034) realised at the \n[University of Illinois at Urbana-Champaign](https://illinois.edu/).\nDevelopment of isotopic fractionation representation and mixed-phase support is carried out with support from\nthe [Polish National Science Centre](https://ncn.gov.pl/en) (grant no. 2020/39/D/ST10/01220).\n\ncopyright: [Jagiellonian University](https://en.uj.edu.pl/en) (2019-2023) & [AGH University of Krakow](https://agh.edu.pl/en) (2023-...) \nlicence: [GPL v3](https://www.gnu.org/licenses/gpl-3.0.html)\n\n## Related resources and open-source projects\n\n### SDM patents (some expired, some withdrawn):\n- https://patents.google.com/patent/US7756693B2\n- https://patents.google.com/patent/EP1847939A3\n- https://patents.google.com/patent/JP4742387B2\n- https://patents.google.com/patent/CN101059821B\n\n### Other SDM implementations:\n- SCALE-SDM (Fortran): \n https://github.com/Shima-Lab/SCALE-SDM_BOMEX_Sato2018/blob/master/contrib/SDM/sdm_coalescence.f90\n- Pencil Code (Fortran): \n https://github.com/pencil-code/pencil-code/blob/master/src/particles_coagulation.f90\n- PALM LES (Fortran): \n https://palm.muk.uni-hannover.de/trac/browser/palm/trunk/SOURCE/lagrangian_particle_model_mod.f90\n- libcloudph++ (C++): \n https://github.com/igfuw/libcloudphxx/blob/master/src/impl/particles_impl_coal.ipp\n- LCM1D (Python) \n https://github.com/SimonUnterstrasser/ColumnModel\n- superdroplet (Cython/Numba/C++11/Fortran 2008/Julia) \n https://github.com/darothen/superdroplet\n- NTLP (FORTRAN) \n https://github.com/Folca/NTLP/blob/SuperDroplet/les.F\n- CLEO (C++)\n https://yoctoyotta1024.github.io/CLEO/\n\n### non-SDM probabilistic particle-based coagulation solvers\n\n- PartMC (Fortran): \n https://github.com/compdyn/partmc\n\n### Python models with discrete-particle (moving-sectional) representation of particle size spectrum\n\n- pyrcel: https://github.com/darothen/pyrcel\n- PyBox: https://github.com/loftytopping/PyBox\n- py-cloud-parcel-model: https://github.com/emmasimp/py-cloud-parcel-model\n\n\n",
"bugtrack_url": null,
"license": "GPL-3.0",
"summary": "Pythonic particle-based (super-droplet) warm-rain/aqueous-chemistry cloud microphysics package with box, parcel & 1D/2D prescribed-flow examples in Python, Julia and Matlab",
"version": "2.52",
"project_urls": {
"Documentation": "https://open-atmos.github.io/PySDM",
"Homepage": "https://github.com/open-atmos/PySDM",
"Source": "https://github.com/open-atmos/PySDM",
"Tracker": "https://github.com/open-atmos/PySDM/issues"
},
"split_keywords": [
"physics-simulation",
" monte-carlo-simulation",
" gpu-computing",
" atmospheric-modelling",
" particle-system",
" numba",
" thrust",
" nvrtc",
" pint",
" atmospheric-physics"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "a6d78df033c36bd4df512c588fe57487a432946718bf49e66437e45f944c79e5",
"md5": "1fb570932ef570882802677e56b494b1",
"sha256": "331fbfeadf143dbbd55f2d156d8dd2aaf1b968cf6a151fc91f4ee7cb987338af"
},
"downloads": -1,
"filename": "PySDM-2.52-py3-none-any.whl",
"has_sig": false,
"md5_digest": "1fb570932ef570882802677e56b494b1",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": null,
"size": 308089,
"upload_time": "2024-04-20T20:38:33",
"upload_time_iso_8601": "2024-04-20T20:38:33.106663Z",
"url": "https://files.pythonhosted.org/packages/a6/d7/8df033c36bd4df512c588fe57487a432946718bf49e66437e45f944c79e5/PySDM-2.52-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "723b18f2304825f395f4aaad6d6414fab3a702384fc29d9fb499d9bd86f8641b",
"md5": "583eb3b475ab10ad9e072a865b586ad9",
"sha256": "cdac86186fc5f854cb92ad396ce29dae16dec4e6b653f8070830ce5a79b5e32c"
},
"downloads": -1,
"filename": "PySDM-2.52.tar.gz",
"has_sig": false,
"md5_digest": "583eb3b475ab10ad9e072a865b586ad9",
"packagetype": "sdist",
"python_version": "source",
"requires_python": null,
"size": 3847392,
"upload_time": "2024-04-20T20:38:38",
"upload_time_iso_8601": "2024-04-20T20:38:38.025710Z",
"url": "https://files.pythonhosted.org/packages/72/3b/18f2304825f395f4aaad6d6414fab3a702384fc29d9fb499d9bd86f8641b/PySDM-2.52.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2024-04-20 20:38:38",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "open-atmos",
"github_project": "PySDM",
"travis_ci": false,
"coveralls": false,
"github_actions": true,
"appveyor": true,
"lcname": "pysdm"
}
JSON
