climlab-rrtmg

Brian Rose, University at Albany

About

This is a stand-alone Python wrapper for the RRTMG radiation modules.

The primary use-case is to drive the RRTMG radiation processes in climlab, but it can be used as a stand-alone radiation model if you are familiar with the RRTMG Fortran interface. This is a lightweight wrapper that emulates the Fortran interface as closely as possible.

Currently we are wrapping RRTMG_LW v4.85 and RRTMG_SW v4.0. The Fortran source code is included in this repository. The latest versions 5.0 of the RRTMG source code are available on GitHub here

This wrapper includes the following modifications:

• RRTMG_LW is modified to allow reporting of OLR components in spectral bands, as illustrated using climlab here. This modification is strictly diagnostic does not change any other behavior of RRTMG_LW.
• RRTMG_SW is modified to allow gridpoint-specific values of the input parameter adjes which is the adjustment to total solar irradiance.

The parameter adjes is used within RRTMG_SW to account for time-of-year adjustments to Earth-Sun distance. Latest versions of climlab compute non-uniform values of irradiance_factor to compensate for different diurnal averages of the solar zenith angle. climlab's irradiance_factor is mapped to adjes when this RRTMG driver is called from climlab.

Installation

Pre-built binaries for many platforms are available from conda-forge.

To install in the current environment:

conda install climlab-rrtmg --channel conda-forge

or create a self-contained environment:

conda create --name my_env python=3.10 climlab-rrtmg --channel conda-forge
conda activate my_env

See below for instructions on how to build from source.

Example usage

You can import the modules into a Python session with

from climlab_rrtmg import rrtmg_lw, rrtmg_sw

The main RRTMG drivers are exposed through

rrtmg_lw.climlab_rrtmg_lw()

and

rrtmg_sw.climlab_rrtmg_sw()

Please see the directory climlab_rrtmg/tests/ directory in this repository for working examples that set up all the necessary input arrays and call the drivers.

Building from source

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:

conda env create --file ./ci/requirements.yml
conda activate rrtmg_build_env
python -m pip install . --no-deps -vv

To run tests, do this from any directory other than the climlab-rrtmg repo:

pytest -v --pyargs climlab_rrtmg

Version history

• Version 0.3 (released January 2024) includes the modification for grid-variable adjes described above. It is designed to work with climlab v0.9 and higher.
• Version 0.2 is the first public release (April 2022). The Python wrapper code has been extracted from climlab v0.7.13.