/PySDM

Pythonic particle-based (super-droplet) warm-rain/aqueous-chemistry cloud microphysics package with box, parcel & 1D/2D prescribed-flow examples

Primary LanguagePythonGNU General Public License v3.0GPL-3.0

Python 3 LLVM CUDA Linux OK macOS OK Windows OK Jupyter
Travis Build Status Github Actions Build Status Appveyor Build status Dependabot Coverage Status
Maintenance GitHub issues GitHub issues GitHub issues GitHub issues
OpenHub EU Funding Copyright License: GPL v3

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 core is a Pythonic high-performance implementation of the Super-Droplet Method (SDM) Monte-Carlo algorithm for representing collisional growth (Shima et al. 2009), hence the name.

PySDM has two alternative parallel number-crunching backends available: multi-threaded CPU backend based on Numba and GPU-resident backend built on top of ThrustRTC. The Numba backend named CPU is the default, and features multi-threaded parallelism for multi-core CPUs. It uses the just-in-time compilation technique based on the LLVM infrastructure. The ThrustRTC backend named GPU offers GPU-resident operation of PySDM leveraging the SIMT parallelisation model. Using the GPU backend requires nVidia hardware and CUDA driver.

For an overview paper on PySDM v1 (and the preferred item to cite if using PySDM), see Bartman et al. 2021 arXiv e-print (submitted to JOSS). For a list of talks and other materials on PySDM, see the project wiki.

Dependencies and Installation

PySDM dependencies are: Numpy, Numba, SciPy, Pint, chempy, ThrustRTC and CURandRTC.

To install PySDM using pip, one may use: pip install git+https://github.com/atmos-cloud-sim-uj/PySDM.git.

For development purposes, we suggest cloning the repository and installing it using pip -e. Test-time dependencies are listed in the requirements.txt file.

Besides the PySDM repository (where this README file is hosted), there are three related repositories:

The examples have additional dependencies listed in PySDM_examples package setup.py file.

Running the examples requires the the PySDM_examples package to be installed (or within PYTHONPATH). Since the examples package includes Jupyter notebooks, the suggested install and launch steps are:

git clone https://github.com/atmos-cloud-sim-uj/PySDM-examples.git
cd PySDM-examples
pip install -e .
jupyter-notebook

PySDM examples (Jupyter notebooks reproducing results from literature):

0D box-model coalescence-only examples (work with both CPU and GPU backends):

  • Shima et al. 2009 Fig. 2 Binder Open In Colab
    (Box model, coalescence only, test case employing Golovin analytical solution)
  • Berry 1967 Figs. 5, 8 & 10 Binder Open In Colab
    (Box model, coalescence only, test cases for realistic kernels)

0D parcel-model condensation only examples (CPU backend only, stay tuned...)

  • Arabas & Shima 2017 Fig. 5 Binder Open In Colab
    (Adiabatic parcel, monodisperse size spectrum activation/deactivation test case)
  • Yang et al. 2018 Fig. 2: Binder Open In Colab
    (Adiabatic parcel, polydisperse size spectrum activation/deactivation test case)

0D parcel-model condensation/aqueous-chemistry example (CPU backend only, stay tuned...)

1D kinematic (prescribed-flow, single-column)

  • Shipway & Hill 2012 Binder Open In Colab
    (Fig. 1 with thermodynamics & condensation only - no particle displacement)

2D kinematic (prescribed-flow) Sc-mimicking aerosol collisional processing (warm-rain) examples (CPU backend only, stay tuned...)

  • Arabas et al. 2015 Figs. 8 & 9: Binder Open In Colab
    (interactive web-GUI with product selection, parameter sliders and netCDF/plot export buttons)
  • Bartman et al. 2021 (in preparation): Binder Open In Colab
    (default-settings based script generating a netCDF file and loading it subsequently to create the animation below)

animation

Hello-world example

In order to depict the PySDM API with a practical example, the following listings provide a sample code roughly reproducing the Figure 2 from Shima et al. 2009 paper. It is a coalescence-only set-up in which the initial particle size spectrum is exponential and is deterministically sampled to match the condition of each super-droplet having equal initial multiplicity:

from PySDM.physics import si
from PySDM.initialisation.spectral_sampling import ConstantMultiplicity
from PySDM.initialisation.spectra import Exponential

n_sd = 2**15
initial_spectrum = Exponential(norm_factor=8.39e12, scale=1.19e5 * si.um**3)
attributes = {}
attributes['volume'], attributes['n'] = ConstantMultiplicity(spectrum=initial_spectrum).sample(n_sd)

The key element of the PySDM interface is the Core class which instances are used to manage the system state and control the simulation. Instantiation of the Core class is handled by the Builder as exemplified below:

from PySDM import Builder
from PySDM.environments import Box
from PySDM.dynamics import Coalescence
from PySDM.dynamics.coalescence.kernels import Golovin
from PySDM.backends import CPU
from PySDM.products.state import ParticlesVolumeSpectrum

builder = Builder(n_sd=n_sd, backend=CPU)
builder.set_environment(Box(dt=1 * si.s, dv=1e6 * si.m**3))
builder.add_dynamic(Coalescence(kernel=Golovin(b=1.5e3 / si.s)))
products = [ParticlesVolumeSpectrum()]
particles = builder.build(attributes, products)

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 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 Monte-Carlo algorithm (Super Droplet Method) is registered as the only dynamic in the system (other available dynamics representing condensational growth and particle displacement). Finally, the build() method is used to obtain an instance of Core which can then be used to control time-stepping and access simulation state.

The run(nt) 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:

from PySDM.physics.constants import rho_w
from matplotlib import pyplot
import numpy as np

radius_bins_edges = np.logspace(np.log10(10 * si.um), np.log10(5e3 * si.um), num=32)

for step in [0, 1200, 2400, 3600]:
    particles.run(step - particles.n_steps)
    pyplot.step(x=radius_bins_edges[:-1] / si.um,
                y=particles.products['dv/dlnr'].get(radius_bins_edges) * rho_w / 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.svg')

The resultant plot looks as follows:

plot

Package structure and API

Credits:

Development of PySDM is supported by the EU through a grant of the Foundation for Polish Science (POIR.04.04.00-00-5E1C/18).

copyright: Jagiellonian University
licence: GPL v3

Related resources and open-source projects

SDM patents (some expired, some withdrawn):

Other SDM implementations:

non-SDM probabilistic particle-based coagulation solvers

Python models with discrete-particle (moving-sectional) representation of particle size spectrum