C++ interface of RTE+RRTMGP

This is a C++ interface to the Radiative Transfer for Energetics (RTE) and Rapid Radiative Transfer Model for GCM applications Parallel (RRTMGP).

The original code is found at https://github.com/earth-system-radiation/rte-rrtmgp.

Contacts: Robert Pincus and Eli Mlawer email: [email protected]

This C++ interface can be downloaded from https://github.com/earth-system-radiation/rte-rrtmgp-cpp

Contact: Chiel van Heerwaarden email: [email protected]

Use and duplication is permitted under the terms of the BSD 3-clause license, see http://opensource.org/licenses/BSD-3-Clause

The source code of the testing executable in the src_test and include_test directory is released under the GPLv3 license, see https://www.gnu.org/licenses/gpl-3.0.en.html

In order to check out the code including the rte-rrtmgp submodule, use:

git clone --recurse-submodules https://github.com/earth-system-radiation/rte-rrtmgp-cpp

In case you had already checked out the repository, use:

git submodule update --init

Basic instructions

Building the source creates an executable test_rte_rrtmgp. Three test cases are provided in directories rfmip, allsky, and rcemip. In order to run those cases follow the instructions in the README.md of those respective directories.

In general, in order to run a test case, make sure the following files are present in the directory from which test_rte_rrtmgp is triggered:

1. Input file rte_rrtmgp_input.nc with atmospheric profiles of pressure, temperature, and gases.
2. Long wave coefficients file from original RTE+RRTMGP repository (in rrtmgp/data) as coefficients_lw.nc
3. Short wave coefficients file from original RTE+RRTMGP repository (in rrtmgp/data) as coefficients_sw.nc

Instructions for neural network-based gas optics parametrization

Obtain repository https://github.com/MennoVeerman/machinelearning-gasoptics to generate training data for neural networks, to train neural networks and to generate testing data. Then run with ./test_rte_rrtmgp --nn-gas-optics