Foundation AI Models for Aerosol Optical Depth Estimation from PACE Satellite Data

2026-05-01

AI Summary

This study addresses the limitations of traditional aerosol optical depth (AOD) retrieval methods, which rely heavily on physical models and auxiliary data, as well as existing data-driven approaches that struggle to effectively integrate spatial–spectral information from hyperspectral imagery, often yielding spatially discontinuous and noise-sensitive results. To overcome these challenges, this work proposes ViTCG, a novel framework that introduces foundational AI models into AOD retrieval for the first time. Built upon the Vision Transformer architecture, ViTCG incorporates an innovative channel grouping mechanism to jointly model spatial context and spectral signatures directly from PACE satellite hyperspectral radiance data in an end-to-end manner. Experimental results demonstrate that the proposed method significantly outperforms state-of-the-art models—including Prithvi—on PACE data, achieving a 62% reduction in mean squared error and producing highly spatially coherent AOD fields.

Abstract

Aerosol Optical Depth (AOD) retrieval is essential for Earth observation, supporting applications from air quality monitoring to climate studies. Conventional physics-based AOD retrieval methods formulate the problem as a pixel-wise inversion, relying on radiative transfer modeling, memory-intensive look-up tables, and auxiliary meteorological data. While recent data-driven approaches have shown promise, many fail to exploit the spatial-spectral coherence of hyperspectral imagery, leading to spatially inconsistent and noise-sensitive retrievals. We present the first study exploring Foundation AI models for AOD retrieval and propose ViTCG, a Vision Transformer with Channel-wise Grouping-based spatial regression framework that reduces retrieval bias and error. ViTCG uses hyperspectral top-of-atmosphere radiance as input and jointly models spatial context and spectral information. Validation with PACE radiance observations demonstrates a 62% reduction in mean squared error compared to state-of-the-art foundation models, including Prithvi, and produces spatially coherent AOD fields.

Problem

• Aerosol Optical Depth
• hyperspectral imagery
• spatial-spectral coherence
• satellite data
• retrieval inconsistency

Innovation

• Foundation AI models
• Aerosol Optical Depth
• Vision Transformer
• hyperspectral imagery
• spatial-spectral coherence