Machine-learning-based reconstruction of grey-scaled MSG-SEVIRI Dust RGB images

Reconstructions of grey-scaled images of the MSG-SEVIRI Dust RGB product at 9, 12, and 15 UTC in 2021 and 2022 and the trained artificial neural networks used to obtain the reconstructions. The reconstructions were performed to restore the full extent of dust plumes, which are partially obscured by clouds. The dataset further contains reconstructed dust aerosol optical depth reconstructions, which were used to gauge the performance.

The dataset accompanies the following publication:

Kanngießer and Fiedler, 2024, "Seeing" beneath the clouds - machine-learning-based reconstruction of North African dust plumes, AGU Advances, In Press.