PythonicDISORT

What the maker publishes

Additional calculations

The package includes a subroutine for computing actinic fluxes and supports integration with respect to optical depth. source

Boundary conditions

Optional inputs include bidirectional reflectance Fourier modes, boundary sources, and isotropic internal sources. source

Dependencies

The listed runtime requirements are Python 3.8 or later, NumPy 1.8.0 or later, and SciPy 1.8.0 or later. source

Differentiation

When autograd is installed and compatibility is enabled, it can compute tau derivatives of output functions, with slightly lower efficiency. source

Documentation and audience

The documentation recommends that new users read the non-optional parts of sections 1 and 2, which cover parameter choices and modules and outputs. source

Implementation

The project describes itself as coded entirely in Python 3 and based on Stamnes’ FORTRAN DISORT. source

Implementation

It is implemented entirely in Python 3 as a reimplementation of DISORT, rather than as a wrapper. source

Installation

The package can be installed from PyPI or Conda-Forge, or by cloning its repository. source

Installation

The package can be installed from PyPI or Conda-Forge, or by cloning its repository and running pip install ., and requires Python 3.8 or later. source

Integrations

The installation instructions list PyPI and Conda-Forge as package distribution channels; no other integrations are stated on that page. source

License

The GitHub repository lists an MIT license. source

License

PyPI lists the package license as MIT License. source

Notable limit

Only the first NFourier bidirectional reflectance Fourier modes are used; additional modes are ignored. source

Numerical limitation

The documentation says computational cost scales approximately with the cube of the number of quadrature points and warns that around 300 or more streams can cause numerical instability and inaccurate intensity at some angles. source

Optional dependencies

Autograd can optionally compute gradients of output functions; pytest is optional for running tests, and Jupyter, Notebook, and Matplotlib are listed for the comprehensive notebook. source

Optional integrations

Autograd is optional for derivatives; Jupyter, Notebook, and Matplotlib are listed as optional requirements for the comprehensive notebook. source

Outputs

Its solver computes fluxes and can optionally compute intensity. source

Performance limitation

The project says it is optimized for speed but naturally slower than similar Fortran algorithms. source

Platform compatibility

The project says it should be system agnostic given its minimal dependencies and pure Python code, and that the repository was built and tested on Windows 11. source

Platform compatibility

The project says it should be system agnostic because it has minimal dependencies and pure Python code, and says the repository was built and tested on Windows 11. source

Purpose

PythonicDISORT solves the one-dimensional radiative transfer equation for single- or multi-layer plane-parallel atmospheres. source

Purpose

PythonicDISORT solves the one-dimensional radiative transfer equation for a plane-parallel, horizontally homogeneous atmosphere. source

Radiative transfer features

The documented solver supports optional delta-M scaling and Nakajima-Tanaka intensity corrections. source

Solver features

Features include a multilayer solver, delta-M scaling, Nakajima-Tanaka corrections, flux-only mode, direct beam and isotropic internal sources, diffuse-flux boundary sources, surface reflection using a bidirectional reflectance distribution function, and polar-angle interpolation. source

Checked against the maker's pages on 12 Oct 2026.