Supplement to "Data-Driven Quadrature for Longwave and Shortwave Absorption by Major Greenhouse Gases"

Readme This dataset contains the python scripts and the data that were used for the article "Data-Driven Quadrature for Longwave and Shortwave Absorption by Major Greenhouse Gases." * atmdata contains the atmospheric data that were used for the simulation of Section 3.3. * scattering_data contains the scattering data that were used for the simulation of Section 3.3. * results contains the results of the simulations of Section 3.3 namely the fluxes for the different schemes and the reference solution. * scripts contains the Python scripts that were used for the simulations of Section 3.3 and for the conversion of the ERA5 data to the ARTS format, as well as the scripts used to run RRTMGP, konrad, and process the data (compute the errors) presented in the remaining figures. The radiative transfer simulations were performed using the ARTS radiative transfer model [text](https://atmtools.github.io/arts-docs-2.6/index.html). To run the scripts, you need to have installed pyarts. pyarts is the Python interface for the Atmospheric Radiative Transfer Simulator (ARTS). * DDQ configurations stores the wavenumbers (cm$^{-1}$) and accompanying quadrature weights of the quadrature scheme for present-day longwave (DDQ_LW_present) and shortwave (DDQ_SW_present) calculations, as well as sets that are trained on variability in the concentrations of major greenhouse gases (DDQ_LW and DDQ_SW). Additional optimizations are available in the subfolder `Additional configurations`. Lookup tables in the ARTS format for the 4 main configurations are available in the subfolder `LUTs`. * Tutorial.ipynb describes the full process of optimizing a quadrature scheme, including creating a training dataset of line-by-line fluxes with the `FluxSimulator` ARTS module, writing the cost functions, setting up the training data and running the optimization procedure via Python package `datadrivenquadrature`. The required packages and their versions are detailed in environment.yml.