Deep Convective Cloud


Improve the method of in-flight radiometric calibration of optical sensors based upon the deep convective clouds (DCC), which is currently used operationaly at CNES. The aim was two-fold:

  • improve the accuracy of the inter-bands calibration, which is currently about +/- 1%
  • try to use DCC targets for absolute calibration of, at least, one of the bands of the sensor to be calibrated.


We built a new radiative model of DCC from a database of DCC optical properties. First, we created this database gathering data from A-train instruments and geostationary sensors. In a second step, we have performed sensitivity study of the shape and magnitude of TOA signal in UV, VIS, NIR and SWIR spectral domains to DCC properties. For that, we have used the radiative transfer tools of the ARTDECO package, already used to simulate Level 1B images of future 3MI and MetImage sensors on EPS-SG platform. The combined analysis of the geophysical varibility of micro- and macro-physical parameters of the DCC from the database and their impact on the shape and magnitude of the TOA signal from the sensitivity study has led to a forecasting algorithm of TOA radiances over the DCC. Finally, this algorithm has been implemented in a python software that was used for a performance analysis on Sentinel-3/OLCI and Sentinel-2/MSI sensors, which has led to recommendations for potential improvement of the method.

Funder: CNES
Duration: 2018-2019
Contact at Hygeos: Dominique Jolivet