IceCloudNet: Cirrus and mixed-phase cloud prediction from SEVIRI input learned from sparse supervision

TL;DR

This paper introduces IceCloudNet, a neural network trained on satellite data to predict ice cloud properties, aiming to improve climate models and reduce uncertainties related to cirrus and mixed-phase clouds.

Contribution

The work presents a new deep learning approach for regime-dependent ice microphysical property prediction using geostationary satellite data, enhancing observational constraints.

Findings

Achieved regime-dependent ice cloud predictions from satellite data

Enabled improved understanding of ice cloud processes

Provided a tool to reduce climate model uncertainties

Abstract

Clouds containing ice particles play a crucial role in the climate system. Yet they remain a source of great uncertainty in climate models and future climate projections. In this work, we create a new observational constraint of regime-dependent ice microphysical properties at the spatio-temporal coverage of geostationary satellite instruments and the quality of active satellite retrievals. We achieve this by training a convolutional neural network on three years of SEVIRI and DARDAR data sets. This work will enable novel research to improve ice cloud process understanding and hence, reduce uncertainties in a changing climate and help assess geoengineering methods for cirrus clouds.