Atmospheric Chemistry and Data Science – The Final Frontier?
LOCATION
Convenors: Paul Griffiths, University of Cambridge and Ryan Hossaini, Lancaster University
Atmospheric science is being transformed by the growing use of innovative data science and machine learning methods. In this ACSG meeting we explored how such methods have been successfully applied in atmospheric chemistry research to provide new insights into processes affecting air quality and climate. Focused on application as well as method, our national and international speakers covered topics that included (1) downscaling of atmospheric models with machine learning for improved air quality exposure studies, (2) how causal discovery algorithms offer a novel approach for climate model evaluation, and (3) how the development of ‘digital twins’ might operationally be used to improve air quality.
Agenda
8th Dec 2022 11:00 - 16:00
Time | Title | Speaker |
---|---|---|
11:00 | Digital Twins of urban air quality: opportunities and challenges. | David Owen Topping |
11:30 | Downscaling an atmospheric chemistry transport model ozone surface using machine learning | Lily Gouldsbrough |
12:00 | Using Automated Image Detection to Find Pollution Sources from Satellite Data | Douglas Finch |
12:30 | Machine learning for atmospheric chemistry: why, where, how? | Peer Johannes Nowack |
14:00 | Deep Learning for Air Quality Research: Concepts, Issues, and Benchmarks | Martin Schultz |
14:30 | An Online-Learned Machine Learning Chemical Solver for Stable, Fast, and Long-Term Global Simulations of Atmospheric Chemistry | Makoto Kelp |
15:00 | Gaussian process emulation of the global methane budget: a sensitivity analysis | Angharad Caroline Stell |
15:30 | Combining chemistry-climate model ensembles and observations with Bayesian machine learning. | Matt Amos |
Registration
REGISTRATION IS CLOSED
Convenors: Paul Griffiths, University of Cambridge and Ryan Hossaini, Lancaster University
Atmospheric science is being transformed by the growing use of innovative data science and machine learning methods. In this ACSG meeting we explored how such methods have been successfully applied in atmospheric chemistry research to provide new insights into processes affecting air quality and climate. Focused on application as well as method, our national and international speakers covered topics that included (1) downscaling of atmospheric models with machine learning for improved air quality exposure studies, (2) how causal discovery algorithms offer a novel approach for climate model evaluation, and (3) how the development of ‘digital twins’ might operationally be used to improve air quality.
Agenda
8th Dec 2022 11:00 - 16:00
Time | Title | Speaker |
---|---|---|
11:00 | Digital Twins of urban air quality: opportunities and challenges. | David Owen Topping |
11:30 | Downscaling an atmospheric chemistry transport model ozone surface using machine learning | Lily Gouldsbrough |
12:00 | Using Automated Image Detection to Find Pollution Sources from Satellite Data | Douglas Finch |
12:30 | Machine learning for atmospheric chemistry: why, where, how? | Peer Johannes Nowack |
14:00 | Deep Learning for Air Quality Research: Concepts, Issues, and Benchmarks | Martin Schultz |
14:30 | An Online-Learned Machine Learning Chemical Solver for Stable, Fast, and Long-Term Global Simulations of Atmospheric Chemistry | Makoto Kelp |
15:00 | Gaussian process emulation of the global methane budget: a sensitivity analysis | Angharad Caroline Stell |
15:30 | Combining chemistry-climate model ensembles and observations with Bayesian machine learning. | Matt Amos |