IceCloudNet is a deep learning model that maps between geostationary data and vertically resolved DARDAR and DARDAR-Nice data. IceCloudNet is able to adequately reconstruct the vertical cloud structure and predict ice water content and ice crystal number concentration of clouds containing ice with high precision.
The data set produced by IceCloudNet combines the spatio-temporal coverage and resolution of Meteosat SEVIRI with the vertical resolution of DARDAR-Nice, increasing the availability of vertically resolved cirrus and mixed-phase cloud profiles by over six orders of magnitude compared to the DARDAR-Nice data set.
IceCloudNet data enables many possiblities for new research on cloud formation and development. For instance, by tracking any long-lasting cloud system such as mesoscale convective systems and tropical cyclones in all spatial and temporal dimensions. Additionally, IceCloudNet data can be utilized as an observational constraint for the validation of high-resolution climate models.
This research was supported by grants from the European Union’s Horizon 2020 research and innovation program iMIRACLI under Marie Skłodowska-Curie grant agreement No 860100.