Method for retrieving range-resolved aerosol microphysical properties from polarization lidar measurements.

Area of Science:

• Atmospheric Science
• Remote Sensing
• Climate Change Research

Background:

• Aerosol microphysical properties like volume concentration (VC) and effective radius (ER) are crucial for climate change studies.
• Current remote sensing methods struggle to provide range-resolved aerosol VC and ER, limiting climate impact assessments.
• Column-integrated data from sun-photometers are the primary available remote sensing product.

Purpose of the Study:

• To develop and validate a novel retrieval method for range-resolved aerosol VC and ER.
• To leverage polarization lidar and AERONET sun-photometer data for improved aerosol property characterization.
• To provide a practical approach for obtaining full-day, height-resolved aerosol data.

Main Methods:

• Utilized partial least squares regression (PLSR) and deep neural networks (DNN).
• Combined data from polarization lidar and collocated AERONET sun-photometer observations.
• Validated results with Aerodynamic Particle Sizer (APS) measurements.

Main Results:

• The DNN method achieved high accuracy for aerosol VC (R²=0.89) and ER (R²=0.77) retrieval.
• Lidar-derived height-resolved VC and ER showed good agreement with APS observations.
• Significant diurnal and seasonal variations in aerosol VC and ER were observed at SACOL.

Conclusions:

• Polarization lidar measurements can reliably derive range-resolved aerosol VC and ER using the proposed DNN method.
• This approach offers a practical way to obtain full-day, height-resolved aerosol data, even with clouds.
• The method is applicable to existing lidar networks and spaceborne missions for enhanced climate effect evaluation.