Machine Learning Approach for Retrieving Vertical PM Concentrations via Coherent Doppler Lidar

Researchers at the Ocean University of China have developed a novel approach for retrieving vertical particulate matter (PM) concentrations using a single coherent Doppler lidar (CDL) combined with machine learning (ML) models. The study presents a promising pathway for three-dimensional PM observation with atmospheric dynamics information, offering the potential to enhance air quality monitoring through expanded CDL networks and ML applications.

Key Takeaways:

• The study presents a novel approach for retrieving vertical PM concentrations using a single CDL combined with ML models, demonstrating good performance in capturing and characterizing vertical PM distribution.
• The models were trained using in-situ PM2.5 and PM10 data as true values, along with input features including particle extinction coefficient, signal-to-noise ratio, wind speed, wind direction from CDL observations, as well as temperature and relative humidity from ERA5 reanalysis data.
• Observations from Qingdao, China, during winter and spring (2020-2024) were used for model development and evaluation, with the models showing good performance in both PM2.5 and PM10 test sets.
• Case studies of haze and dust events demonstrated the capability of this method in capturing vertical PM distribution, with the PM layers of typical haze and dust events primarily concentrated within 1 km and 1.2 km in altitude, respectively.
• The study found that the enhancement of wind speed and negative vertical velocity in the later phase of PM events may result in the dissipation and deposition of the PM.
• The research has been peer-reviewed and published in the journal Atmospheric Environment.

Statistics:

• R2, RMSE, MAE of the PM2.5 test set comparison: 0.787, 18.11 μg/m³, 11.23 μg/m³
• R2, RMSE, MAE of the PM10 test set comparison: 0.803 and 29.98 μg/m³, 18.93 μg/m³
• Height of PM layers in typical haze and dust events: 1 km and 1.2 km in altitude, respectively

Sources:

• Atmosphere Environment, 2025; 359.
• Ocean University of China, Qingdao, People's Republic of China.
• Pergamon-elsevier Science Ltd, The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, England.