TROPOMI CSF Software

To reproduce the results presented by Sadavarte, et al., 2021 , we provide an open source software package GitHub TROPOMI_CSF available on GitHub. The repository includes a README.md file with detailed instructions for installation and use. The software is released under the MIT License. However, GitHub TROPOMI_CSF depends on several third-party libraries, including Cartopy, cdsapi, Matplotlib , netCDF4, NumPy, GDAL/OSGeo, and Shapely, and users are responsible for ensuring compliance with the respective licenses of all dependencies.

Readers are encouraged to install the project locally to verify the reported results and to conduct their own analyses.

The Algorithm section below presents a pseudocode reconstruction of the CSF algorithm as described in the page 16575 of Sadavarte, et al., 2021

The Data section describes the datasets used by the software and explains the key choices made in processing both ERA5 and TROPOMI inputs. This includes decisions regarding the temporal alignment of ERA5 fields, grid point approximations, handling of missing orbits, selection of processing versions, and the application of data quality filtering.

Following the data discussion, we describe the implementation of the algorithm in sections corresponding to its major components:

• Boundary layer pressure averaged wind,
• Background calculation,
• Downwind box calculation,
• Transects calculation,
• Calculations Not Implemented,
• Algorithm difficulties, and
• Emissions and IEF.

All steps are illustrated using TROPOMI orbit 09956.

The final section, Computational stability, addresses computational stability by examining how variations in methodological choices affect the calculated results, using TROPOMI orbit 09956 as a representative example.