Global Assessment and Error Budget of EarthCARE Column Aerosol Optical Depth: Physical Limits and Quantitative Mitigation

Abstract

The Earth Cloud Aerosol and Radiation Explorer (EarthCARE) provides critical spaceborne high spectral resolution LiDAR observations for global aerosol profiling. However, integrating these vertically resolved aerosol extinction coefficient profiles into column Aerosol Optical Depth (AOD) introduces systematic biases from physical observation boundaries, which currently lack global validation and quantitative attribution. This study presents a global assessment and error budget of EarthCARE 355 nm column AOD using solar and lunar AErosol Robotic NETwork measurements from August 2024 to March 2026. Baseline validations demonstrate robust spatiotemporal tracking capabilities (daytime R = 0.873, nighttime R = 0.859) and physically consistent independently retrieved LiDAR ratio distributions across coarse and fine mode aerosol regimes. Nevertheless, cross-framework comparisons show that complex absorbing mixtures are frequently assigned to fine mode dominant EarthCARE categories. Concurrently, a pervasive systematic underestimation persists globally (daytime MB = −0.060, nighttime MB = −0.073). Mechanistically, this underestimation is associated with three distinct physical limits: near-surface clutter truncation (0–400 m), free-tropospheric signal-to-noise ratio detection limits, and attenuation-induced sampling biases under extreme aerosol loading (AOD > 1.0). Applying a constant boundary-layer geometric extrapolation mitigates the near-surface truncation, reducing the daytime MB by ∼40% (to −0.036) and recovering the linear regression slope from 0.67 to 0.77 while preserving high R. Further, compensating for free-tropospheric background detection limit reduces the ubiquitous underestimation by an additional ∼17% (+0.010). This physically based error-budget framework quantifies the major recoverable contributions and characterizes the factors associated with the remaining bias, providing a robust basis for optimizing future spaceborne LiDAR retrievals.