Lidar AOD inversion and aerosol extinction profile correction method based on GA-BP neural network

Area of Science:

• Atmospheric Science
• Remote Sensing Technology
• Computational Methods

Background:

• Accurate retrieval of aerosol optical thickness (AOD) and aerosol extinction coefficient profile (AECP) using lidar is crucial for understanding atmospheric properties.
• Selecting the appropriate aerosol extinction-backscattering ratio (AE-BR) is a critical challenge in lidar inversion accuracy.

Purpose of the Study:

• To develop and validate a novel inversion algorithm for AOD and AECP using a genetic BP (GA-BP) neural network.
• To improve the accuracy of lidar-based atmospheric aerosol property retrieval.

Main Methods:

• Utilized simultaneous measurements from CE318 sun photometer and lidar.
• Developed a GA-BP neural network to establish the relationship between lidar echo signals and AOD.
• Implemented real-time dynamic correction of AE-BR based on GA-BP output for AECP inversion.

Main Results:

• Achieved accurate AOD inversion with a mean absolute error of 0.0156.
• Determined the optimal AE-BR range to be 20-50 sr.
• Demonstrated real-time dynamic correction of AECP, showing strong generalization ability.

Conclusions:

• The GA-BP neural network method provides accurate AOD and AECP retrieval.
• Dynamic adjustment of AE-BR significantly enhances AECP inversion accuracy.
• The proposed method is applicable and effective for practical lidar-based atmospheric aerosol studies.