The Copernicus Sentinel-5 Precursor (Sentinel-5P or S5P) satellite mission is one of the European Space Agency's (ESA) new mission family - Sentinels, and it is a joint initiative between the Kingdom of the Netherlands and the ESA. The sole payload on Sentinel-5P is the TROPOspheric Monitoring Instrument (TROPOMI), which is a nadir-viewing 108 degree Field-of-View push-broom grating hyperspectral spectrometer, covering the wavelength of ultraviolet-visible (UV-VIS, 270nm to 495nm), near infrared (NIR, 675nm to 775nm), and shortwave infrared (SWIR, 2305nm-2385nm). Sentinel-5P is the first of the Atmospheric Composition Sentinels and provides measurements of ozone, NO2, SO2, CH4, CO, formaldehyde, aerosols and clouds at high spatial, temporal and spectral resolutions. Spatial resolution is 5.5 km along-track and 3.5 km across-track at nadir. S5P is in a sun-synchronous orbit with an ascending node of ~13:30 which provides daily global coverage of the sun-lit Earth. Data are archived in enhanced netCDF-4 format. Beginning August 6th in 2019, S5P implemented along-track high spatial resolution (~5.5km at nadir). The near-real-time (NRT) data are generally available within three hours of measurement as required by the Land Atmosphere NRT Capability Earth Observing System (LANCE). Data granules are in five-minute intervals of an orbit, rather than a full orbit as provided for the standard products. NRT data are only archived for five days and not considered to be science-quality data. The data are intended for rapid turnaround assessments, for rigorous scientific analyses users should use the standard Version 3 product.

The primary variables for the S5P cloud product are: cloud pressure, cloud altitude, cloud-top temperature, cloud fraction, and cloud phase. The TROPOMI Cloud products are retrieved by two algorithms. The S5P_CLOUD_OCRA is based on the Optical Cloud Recognition Algorithm which generates the fractional cloud cover. The S5P_CLOUD_ROCINN (the Retrieval Of Cloud Information through Neural Networks algorithm) is built on measurements in and around the O2 A-band, which produces the cloud optical thickness and the cloud top height. The Version 3 cloud algorithm implements a new scan angle correction for the Optical Cloud Recognition Algorithm (OCRA) based on Rayleigh scattering, and uses a new a-priori based on a neural network.