Day 4

Detailed paper information

Back to list

Paper title Evaluation of Solar Flux Estimates over Cloudy Oceans for the EarthCARE BBR Instrument Using a New Semi-Physical Angular Distribution Model
  1. Nils Madenach Freie Universität Berlin Speaker
  2. Florian Tornow Columbia University, Columbia University Center for Climate System Research and NASA GISS, New York, NY
  3. Carlos Domenech GMV
  4. Rene Preusker Freie Universität Berlin (FUB)
  5. Jürgen Fischer Freie Universität Berlin (FUB)
Form of presentation Poster
  • A1. Atmosphere
    • A1.09 EarthCARE: Preparing for the Scientific Mission Exploitation to Quantify the Impact of Clouds and Aerosols on Radiation
Abstract text The Earth Clouds, Aerosols and Radiation Explorer (EarthCARE) has the scientific goal to achieve agreement of +- 10 W/m² for average SW/LW fluxes simulated using radiative transfer models acting on the retrieved profiles of cloud and aerosol properties and values inferred from collocated measurements made by the broadband radiometer (BBR).

The fluxes are estimated from BBR measurements at a single sun-observer geometry of the satellite using angular distribution models (ADMs). ADMs for SW radiances are created for different scene types and constructed from Clouds and the Earth’s Radiant Energy System (CERES) data using a feed-forward back-propagation artificial neural network (ANN) technique (Domenech et al., 2011) .
To further improve the solar flux estimates, a new method has been developed to possibly supplement the ANN technique (Tornow et al., 2020). The semi-physical log-linear approach incorporates cloud effective radius (Reff) and cloud topped water vapor as additional parameters which can significantly influence the TOA solar flux through changes in scattering direction and absorption respectively. A comparison with the state-of-the-art solar flux retrievals obtained from CERES and GERB instruments showed significant flux differences for cloudy scenes over ocean, which has been attributed to extremes in Reff and cloud topped water vapor (Tornow et al., 2021).

In the study presented here, the new method is evaluated and compared with the ANN technique. Since EarthCare is not yet in orbit, simulated EarthCare frames (1/8 of the orbit) are used. The frames were created by the EarthCare team using the Global Environmental Multiscale (GEM) model from Environment Canada and Climate Change and ESA instrument models.
Situations with large differences are analysed and interpreted in more detail. Further it is discussed in which situations the ANN technique could be complemented by the new method.