A174-0005
The Operational TROPOMI/Sentinel-5 Precursor Cloud Products Version 2

Tuesday, 15 December 2020
Poster
Athina Argyrouli1,2, Ronny Lutz3, Fabian Romahn4, Diego G Loyola3 and Richard Siddans5, (1)German Aerospace Center (DLR), Remote Sensing Technology Institute, Weßling, Germany, (2)Technical University of Munich (TUM), Department of Civil, Geo and Environmental Engineering, Chair of Remote Sensing Technology, Munich, Germany, (3)German Aerospace Center (DLR), Remote Sensing Technology Institute (IMF), Oberpfaffenhofen, Germany, (4)German Aerospace Center (DLR), The Remote Sensing Technology Institute (IMF), Oberpfaffenhofen, Germany, (5)STFC Rutherford Appleton Laboratory, Didcot, United Kingdom
Abstract:
TROPOMI on board of Sentinel-5 Precursor (S5P) provides a continuous daily distribution of several cloud properties required as input for trace-gas retrievals. The operational TROPOMI cloud retrieval is a two-step algorithm where the OCRA (Optical Cloud Recognition Algorithm) computes a radiometric cloud fraction using a broad-band UV/VIS color space approach and ROCINN (Retrieval of Cloud Information using Neural Networks) retrieves the cloud height, cloud optical thickness and cloud albedo from NIR measurements in and around the oxygen A-band (~760nm). Lately, S5P CLOUD version 2 has become publicly available with significant improvements against the initial version 1. A good cloud retrieval performance over snow/ice conditions is challenging because the spectral appearance in UVN of the highly reflective surfaces is similar to those of the clouds. In the initial operational product version 1 this effect could lead to cloud height retrievals at the ground level. This behavior has been largely improved in S5P CLOUD version 2. The radiometric cloud fraction retrieval has been improved by replacing clear-sky background reflectance maps which were previously based on OMI measurements and are now directly retrieved from the TROPOMI measurements themselves. This resulted in a decreased scan angle dependency and more reliable retrieval at small cloud fractions. The loose formation of the Suomi-NPP (SNPP) satellite with Sentinel-5 Precursor (S5P) enabled the direct comparison of the TROPOMI/S5P cloud properties against VIIRS/SNPP products. Prior validation against the VIIRS products showed that TROPOMI cloud version 1 is biased low for the cloud top height property and biased high for the cloud optical thickness. The latter bias is decreased in version 2. A more sophisticated cloud model in ROCINN is called Clouds-as-Layers (CAL) and considers the cloud as a homogeneous cluster of scattering liquid water spherical particles using Mie theory. The CAL model seems to perform very accurately over ocean where the low-level stratocumulus clouds are dominant but over land the cloud top heights are underestimated. The derivation of the cloud phase through the temperature information at the cloud top enables the future extension of the CAL model from liquid to ice where the used ice cloud parameterization considers the cloud as a cluster of a variety of general severely rough habits. An additional improvement of the ROCINN cloud retrieval is achieved by the replacement of the MERIS surface albedo climatology of a coarse resolution (0.25o x 0.25 o) with retrieved surface properties from TROPOMI measurements themselves at the native TROPOMI resolution (5.5 x 3.5 km2) and at the same NIR spectral window. In order to better account for rapidly varying surface conditions, these global S5P surface albedo maps are updated on a daily basis.
<< Previous Abstract | Next Abstract >>