Aerosol particles are key components of the earth system; important in dictating radiative balance, human health, and other areas of key societal concern. Understanding their formation, evolution and impacts relies on developments from multiple disciplines covering both experimental laboratory work, field studies and numerical modelling. In this general session all topics of Aerosol Chemistry and Physics are covered. Contributions from aerosol laboratory, field, remote sensing and model studies are all highly encouraged.
Alongside general contributions, this year we also propose a focus on the increasing uptake and use of tools under the banner of Data Science and AI. Adoption of machine learning can be found across all areas of science, including atmospheric aerosol research. We see increasing use of a range of tools, from computational image tools for improved detection and classification through to development of hybrid numerical models. However we must also recognize the importance of the broader data science ecosystem, from evolving regulations, standards and increased emphasis on data discoverability and provenance. With this in mind we welcome submissions that full under a broad range of atmospheric aerosol applications. This could include, but are not limited to:
- Improved classification of aerosol types from spectral, time-series through to image datasets
- Identification of new processes as a result of ML adoption
- Highlighting new challenges in adopting ML in aerosol research
- Improved understanding of process and parameter sensitivity
- Development of Digital Twins
- Development of hybrid process-ML based aerosol models
- Increased resolution and/or computational efficiency of numerical methods
- New ‘Machine Learning’ ready data repositories
- Use of foundation models and generative AI in atmospheric aerosol research
Alongside general contributions, this year we also propose a focus on the increasing uptake and use of tools under the banner of Data Science and AI. Adoption of machine learning can be found across all areas of science, including atmospheric aerosol research. We see increasing use of a range of tools, from computational image tools for improved detection and classification through to development of hybrid numerical models. However we must also recognize the importance of the broader data science ecosystem, from evolving regulations, standards and increased emphasis on data discoverability and provenance. With this in mind we welcome submissions that full under a broad range of atmospheric aerosol applications. This could include, but are not limited to:
- Improved classification of aerosol types from spectral, time-series through to image datasets
- Identification of new processes as a result of ML adoption
- Highlighting new challenges in adopting ML in aerosol research
- Improved understanding of process and parameter sensitivity
- Development of Digital Twins
- Development of hybrid process-ML based aerosol models
- Increased resolution and/or computational efficiency of numerical methods
- New ‘Machine Learning’ ready data repositories
- Use of foundation models and generative AI in atmospheric aerosol research