Synthetic Sentinel-5/UVNS measurements

This dataset contains synthetic Sentinel-5/UVNS measurements in the NIR, SWIR1, and SWIR3 bands, generated with RemoTAP. This data was used for neural network training and evaluation in our paper "Accelerating greenhouse gas retrieval with neural network-based forward models". It is structured as follows: simple_s5_spectrometer_only_1e6_fixed_LHS_sza75_vza35_lowerAOT_matched_retrieval_era5: training/validation/test dataset covering a large range of parameter values. simple_s5_spectrometer_only_1e3_fixed_LHS_sza70_vza30_lowerAOT_matched_retrieval_era5_limited_angstrom: final retrieval test dataset with restricted parameter ranges representative of actual satellite observations.