A Lagrangian Time-Series Machine Learning Framework for Predicting Concentrations and Exploring Drivers of Atmospheric Aerosols: Model Development and Application to Cloud Condensation Nuclei in Marine Boundary Layer

Key Points

• This research aims to develop a machine learning framework to predict atmospheric aerosol concentrations and explore their environmental drivers.
• Developed a Lagrangian time-series machine learning model.
• Applied the model specifically to cloud condensation nuclei in the marine boundary layer.
• Investigated environmental factors influencing aerosol concentration.
• Successfully predicted aerosol concentrations using the new framework.
• Identified key environmental drivers affecting cloud condensation nuclei.
• Demonstrated significant advantages in prediction accuracy compared to traditional methods.

Abstract