A satellite-data fusion algorithm maps atmospheric aerosols in three dimensions beyond narrow lidar tracks. By combining vertical profiles with geostationary observations, the method produces regional aerosol extinction fields and boundary-layer heights at high resolution. The advance could strengthen research on pollution transport, atmospheric stability, aerosol–boundary layer interactions, and climate processes.
Aerosol optical depth describes the light-extinguishing effect of particles within an atmospheric column, but not their vertical distribution. That missing dimension matters because aerosol layers affect radiative forcing, atmospheric stability, pollutant accumulation, and long-range transport. Spaceborne lidar provides accurate vertical profiles, yet its narrow swath and low revisit frequency leave gaps. Passive geostationary satellites offer broad coverage but cannot resolve vertical structure. Existing passive retrievals and model-based products may also have coarse vertical or horizontal resolution and depend on assumptions about aerosol type. Based on these challenges, in-depth research is needed on high-resolution fusion of active and passive aerosol observations.
A team led by Zhejiang University, with collaborators from the China Meteorological Administration, the Chinese Academy of Sciences, the Satellite Environment Center, and industrial partners, developed a new active–passive remote sensing fusion approach. Published (DOI: 10.34133/remotesensing.1051) on June 26, 2026, in Journal of Remote Sensing , the study addresses the longstanding difficulty of generating continuous regional aerosol profiles that retain lidar-scale vertical detail while extending across the broad viewing areas of geostationary satellites for atmospheric and climate research across diverse atmospheric environments worldwide.
The active–passive fusion algorithm for three-dimensional aerosol extinction coefficient mapping (APEX) integrates aerosol extinction coefficient (AEX) profiles with aerosol optical depth (AOD) and multispectral satellite measurements. It produces three-dimensional aerosol extinction coefficient (3D-AEX) fields at 5-km horizontal and 48-m vertical resolution. Unlike conventional interpolation based mainly on geographical distance, APEX adaptively selects neighboring profiles and weights them using both spatial proximity and spectral similarity. This approach preserves comparable aerosol structures over longer reconstruction distances and reduces interference from dissimilar observations. The fields support planetary boundary layer height (PBLH) estimation at 5-km resolution—about five times finer horizontally than the European Centre for Medium-Range Weather Forecasts reanalysis (ERA5).
In dead-zone reconstruction tests, the researchers removed observed lidar profiles and estimated them from surrounding active and passive measurements. Across the reported tests, the coefficient of determination reached 0.7718 within 50 km and 0.6155 within 200 km. The geometric-based distance threshold adjustment improved reconstruction accuracy relative to fixed-threshold selection, especially over longer distances. Combined distance and spectral weighting also outperformed either weighting strategy alone. Independent validation used 199 cases from six Micro-Pulse Lidar Network stations spanning coastal, urban, industrial, mountainous, and high-latitude environments. APEX achieved an average root mean squared error of 0.0298 km⁻¹, while 77.37% of reconstructed values fell within the ground-measurement error bands. Compared with spectral radiance matching, it reduced average error by 0.0288 km⁻¹ and increased agreement by 16.53 percentage points overall. A Bay of Bengal demonstration further accurately captured spatial changes in regional aerosol layers and boundary-layer height.
The researchers emphasized that APEX turns sparse vertical measurements into continuous regional fields without sacrificing fine atmospheric structure. They noted that this capability can provide stronger data for examining aerosol–boundary layer interactions and improving regional climate analysis. With broader satellite coverage, the framework may eventually support three-dimensional aerosol monitoring across much larger areas.
The team aligned profiles from the Aerosol and Carbon Dioxide Detection Lidar aboard the Atmospheric Environment Monitoring Satellite with AOD and multispectral observations from the Advanced Himawari Imager on Himawari-8 and the Advanced Baseline Imager on Geostationary Operational Environmental Satellite-16. A geometric-based distance threshold adjustment selected profiles, while distance and spectral-similarity weights prioritized observations. Iterative cost-function optimization balanced these profiles against satellite AOD. The calculation was repeated across each grid and evaluated through dead-zone reconstruction and comparisons with ground-based lidar.
Future work will address coverage gaps caused by clouds and the lack of passive AOD and visible-spectrum data at night. The researchers propose incorporating nightlight observations, spatiotemporal interpolation, Bayesian methods, generative models, and deep learning to improve continuity and robustness. As Fengyun-4, Meteosat Third Generation, EarthCARE, and other satellite systems expand coordinated observations, APEX could move toward global applications. Potential uses include tracking pollution transport, studying aerosol–boundary layer feedbacks, improving regional climate models, and assessing aerosol-related climate effects under carbon-neutrality goals.
