The Radical Solution: Machine Learning and the Hydroxyl Radical for Air Quality and Climate
Qindan Zhu (MIT)
Colloquium
Wednesday, February 7, 2024, 3:30 pm
Abstract
The hydroxyl radical (OH) lies at the nexus of climate and air quality as the primary oxidant for
both reactive greenhouse gases and many hazardous air pollutants. Lacking direct observations,
interannual trends of OH either in urban areas or at continental-to-global scale are not well
understood due to the short lifetime and high spatial heterogeneity of OH. This talk will describe
two examples from my research predicting OH using a synthesis of chemistry/climate model,
satellite observations and machine learning (ML). At the urban scale, I will present OH
predictions for 49 North American cities between 2005 and 2014 and show how these OH
predictions can be used to inform the best control strategy of ozone pollution in each city. I will
also present the ML predicted OH trends at continental-to-global scale. As the primary loss
pathway for methane is its reaction with OH, I will utilize these OH predictions to interpret
observed methane trends. Finally, I will provide an overview of OH sensitivity to factors
modulated by climate and discuss how OH is expected to respond to future climate change.
Bio
Qindan is a NOAA Climate & Global Change Postdoc Fellow working with Arlene Fiore at
MIT, focusing on fully coupled chemistry-climate models and chemistry-climate interaction. She
earned her bachelor's degree with dual majors in environmental science and mathematics from
Peking University in 2017 and completed her PhD in Earth and Planetary Science at the
University of California, Berkeley, under the guidance of Ron Cohen. Her research encompasses
satellite observations of air pollutants, airborne flux measurements, weather sensitive emissions
and hydroxyl radical chemistry. A special emphasis in her work is the synthesis of machine
learning, models, and observations to advance the understanding of the interplay between air
quality and climate change.