Brian Rose, University at Albany
This is a stand-alone Python wrapper for the NCAR CAM3 radiation scheme.
The primary use-case is to serve as the under-the-hood radiation driver for climlab, but it can be used as a stand-alone model if you are familiar with the Fortran source code. This is a lightweight wrapper that emulates the Fortran interface as closely as possible.
Pre-built binaries for many platforms are available from conda-forge.
To install in the current environment:
conda install climlab-cam3-radiation --channel conda-forge
or create a self-contained environment:
conda create --name my_env python=3.10 climlab-cam3-radiation --channel conda-forge
conda activate my_env
See below for instructions on how to build from source.
You can import the Fortran driver into a Python session with
import climlab_cam3_radiation as cam3
Please see the directory climlab_cam3_radiation/tests/
directory in this repository
for working examples that set up all the necessary input arrays and call the driver.
Here are instructions to create a build environment (including Fortran compiler) with conda/mamba and build using f2py. It should be possible to build using other Fortran compilers, but I haven't tested this.
To build (example for Apple M1 machine, see ./ci/
for other environment files):
mamba create --name cam3_build_env python=3.10 --channel conda-forge
mamba env update --file ./ci/requirements-macos-arm64.yml
conda activate cam3_build_env
python -m pip install . --no-deps -vv
To run tests, do this from any directory other than the climlab-cam3-radiation repo:
pytest -v --pyargs climlab_cam3_radiation
Version 0.2 is the first public release (April 2022). The Python wrapper code has been extracted from climlab v0.7.13.