3D Cloud reconstruction through geospatially-aware Masked Autoencoders

TL;DR

This paper presents a novel self-supervised learning approach using geospatially-aware Masked Autoencoders to reconstruct 3D cloud structures from satellite imagery and radar data, significantly improving accuracy over existing methods.

Contribution

It introduces a geospatially-aware SSL framework for 3D cloud reconstruction that outperforms current state-of-the-art techniques like U-Nets.

Findings

SSL methods outperform traditional supervised models

Geospatial encoding enhances reconstruction accuracy

Model demonstrates potential for real-time climate data applications

Abstract

Clouds play a key role in Earth's radiation balance with complex effects that introduce large uncertainties into climate models. Real-time 3D cloud data is essential for improving climate predictions. This study leverages geostationary imagery from MSG/SEVIRI and radar reflectivity measurements of cloud profiles from CloudSat/CPR to reconstruct 3D cloud structures. We first apply self-supervised learning (SSL) methods-Masked Autoencoders (MAE) and geospatially-aware SatMAE on unlabelled MSG images, and then fine-tune our models on matched image-profile pairs. Our approach outperforms state-of-the-art methods like U-Nets, and our geospatial encoding further improves prediction results, demonstrating the potential of SSL for cloud reconstruction.