Atmosphere
About 70% of the Earth is covered by clouds, therefore clouds are often present in satellite observations. Cloud properties can be retrieved from satellite observation. Cloudy pixels are often screened before deriving atmospheric and surface properties. In the satellite products, cloud is typically assumed as a horizontal homogeneous layer. However, in reality cloud is a three dimensional (3D) object: clouds cast shadows on the ground surface or on lower clouds; clouds look brighter on the sun illuminated side and darker on the shadow side. The impacts of 3D cloud features on aerosol retrievals have been studied using high resolution satellite imagery data and lidar measurements. Clouds are also important in the trace gas retrievals. The research on the 3D cloud effects on trace gas retrievals is a new topic because the pixel size of the satellite spectrometers like GOME-2 is too big (40 km x 80 km) to see the 3D cloud effects. Since the launch of Sentinel-5p (S5p) in 2017, the trace gases are retrieved at a pixel size of 3.6 km x 5.6 km. We have seen the cloud shadows on the S5P images, which indicates the present of 3D cloud features in the S5p products. The objectives of the project are to analyze the impacts of the 3D clouds on trace gas retrievals, detect the cloud shadows, and derive aerosol and surface albedo products. Aerosol properties and surface albedo are important input parameters in the trace gas retrievals. Aerosol optical thickness (AOT) and surface albedo will be retrieved for selected scenes using the cloud shadow pixels. This is a complimentary method for the general used method nowadays. The algorithm will be demonstrated using Sentinel-2, Sentinel-5P, GF-1/6. The retrieved AOT will be validated from ground-based measurements and compared with Sentinel-3 aerosol products.We will detect cloud shadows from S5p and compared with collocated VIIRS data. The high resolution imagery of VIIRS will provide more accurate detection of cloud shadows and cloud edges on the S5p data. From selected scenes we will study the variation of trace gas column densities with the distance to the clouds. We will use the 3D radiative transfer components of the Earth Clouds and Aerosol Radiation Explorer (EarthCARE) simulator (ECSIM) together with 3D high resolution cloud fields generated using Large-Eddy Simulation (LES) model to simulate S5p/TROPOMI measurements. The simulations will help us to understand the shadow and the 3D cloud effects on the TROPOMI cloud, Absorbing Aerosol Index (AAI), AOT, and nitrogen dioxide (NO2) products. Ultimately, we plan to correct the impact of 3D clouds (shadows) on the Sentinel-5p/4/5 products. The project will use Sentinel-2/3/5p, GF-1/6 products and can be applied to S4/S5 after they are in orbit.The deliverable are reports, publications, and demonstration products and data analysis results. The KNMI team is partly supported by the User Support Programme Space Research of Dutch Research Council and KNMI internal funding. The IAP/CAS team is supported by IAP internal funding.The topic is Atmosphere. Subtopic is related to air quality but also related to greenhouse gases because the greenhouse gas products from satellite observations will also be impacted by 3D clouds and shadows.
