Tropospheric bromine monoxide radical vertical profiles and their differential slant column densities (DSCD) dataset based on forward simulations with an atmospheric radiative transfer model

This dataset was generated through simulation using the atmospheric radiative transfer model SCIATRAN v2.2, which includes three typical forms of BrO vertical profiles (exponential, Boltzmann, and Gaussian) and their corresponding differential slant column concentration (DSCD) data under different aerosol and geometric observation parameters. The profile generation is based on a parameter range obtained from literature research, covering different boundary layer heights, peak positions, and distribution widths, to simulate various distribution scenarios of BrO in the real atmosphere. Each set of profile data is combined with parameters such as aerosol extinction coefficient vertical profile, solar zenith angle (SZA), observed zenith angle (VZA), relative azimuth angle (RAA), etc., and input into the SCIATRAN model. By numerically solving the radiative transfer equation (RTE), the corresponding DSCD value under geometric conditions is calculated. The dataset contains approximately 50 million sets of samples, each containing a set of input parameters (aerosol extinction coefficient vertical profile, SZA, RAA, and a set of DSCD sequences) and an output data (BrO vertical profile). The data is stored in Parquet format, with each file containing multiple batches of data records, including column labels such as SZA, RAA, DSCD, Aerosol, BrO, etc. There were no missing values in the data, and all samples underwent integrity verification. Due to the fact that the data is generated through simulation, its errors mainly come from the numerical calculation errors of the radiative transfer model and the uncertainty of the parameterization scheme. It is suitable for machine learning model training and atmospheric composition inversion algorithm validation. This dataset is suitable for fields such as atmospheric remote sensing, machine learning inversion algorithm development, and BrO vertical distribution research. It is recommended to use Pandas, PyArrow, or PySpark in Python to read Parquet files. Relevant tools can be obtained through Anaconda or PyPI official channels.