mlsanityData to 7 Oct 2026

IceCloudNet: 3D reconstruction of cloud ice from Meteosat SEVIRI

arXiv 2410.04135 ↗First posted 5 Oct 2024physics.ao-ph cs.AI cs.CVcited by 3

Not on the map. Why?

Abstract (arXiv)

IceCloudNet is a novel method based on machine learning able to predict high-quality vertically resolved cloud ice water contents (IWC) and ice crystal number concentrations (N_textrmice). The predictions come at the spatio-temporal coverage and resolution of geostationary satellite observations (SEVIRI) and the vertical resolution of active satellite retrievals (DARDAR). IceCloudNet consists of a ConvNeXt-based U-Net and a 3D PatchGAN discriminator model and is trained by predicting DARDAR profiles from co-located SEVIRI images. Despite the sparse availability of DARDAR data due to its narrow overpass, IceCloudNet is able to predict cloud occurrence, spatial structure, and microphysical properties with high precision. The model has been applied to ten years of SEVIRI data, producing a dataset of vertically resolved IWC and N_textrmice of clouds containing ice with a 3 kmx3 kmx240 mx15 minute resolution in a spatial domain of 30degW to 30degE and 30degS to 30degN. The produced dataset increases the availability of vertical cloud profiles, for the period when DARDAR is available, by more than six orders of magnitude and moreover, IceCloudNet is able to produce vertical cloud profiles beyond the lifetime of the recently ended satellite missions underlying DARDAR.

Key facts. Cited by 3 papers in the map (data through 7 Oct 2026). Builds on 12 of its 57 references that are in the map. First posted on arXiv 5 Oct 2024.

Builds on

12 of 57 references.

References · 12

  1. 2017Attention Is All You Needcited by 59,932
  2. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  3. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  4. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,604
  5. 2016Image-to-Image Translation with Conditional Adversarial Networkscited by 4,510
  6. 2020Taming Transformers for High-Resolution Image Synthesiscited by 3,229
  7. 2016Deep Networks with Stochastic Depthcited by 973
  8. 2017Wasserstein GANcited by 635
  9. 2015Autoencoding beyond pixels using a learned similarity metriccited by 407
  10. 2022Cold Diffusion: Inverting Arbitrary Image Transforms Without Noisecited by 275
  11. 2023Artificial intelligence to advance Earth observation: : A review of models, recent trends, and pathways forwardcited by 28
  12. 2023IceCloudNet: Cirrus and mixed-phase cloud prediction from SEVIRI input learned from sparse supervisioncited by 2

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 2; 2026: 1.

3 papers.

  1. 2025Global 3D Reconstruction of Clouds & Tropical Cyclonescited by 1
  2. 2025SGMAGNet: A Baseline Model for 3D Cloud Phase Structure Reconstruction on a New Passive Active Satellite Benchmarkcited by 0
  3. 2026C3DIR: A Deep Learning 3-Dimensional Cloud Property Retrieval Scheme for Passive Satellite Imagerscited by 0