CAREER: Transforming Modern Understanding of Atmospheric Chemistry and the Biosphere through Physics-Informed Machine Learning

This CAREER project will apply machine learning tools to develop new computational methods to improve predictability in atmospheric chemistry. The project is focused on gaining a better understanding of the controlling factors behind the chemical composition of the atmosphere, especially those related to the plant biosphere. This effort is expected to transform the current understanding of biosphere-atmosphere interactions and provide an important base for numerous future atmospheric observational and modeling studies.

This work supports an investigation into the impacts of the uncertainties in the biosphere-atmosphere interactions on atmospheric composition using a combination of traditional computational methods alongside novel machine learning techniques, all grounded in a suite of observations. The plant biosphere contributes many trace species to the atmosphere, as well as exerting secondary effects and feedbacks within the Earth System.

Biogenic emissions are responsible for most reactive carbon in the atmosphere, and control much of the atmospheric reactivity worldwide. The science objectives of this project are to: (1) quantify uncertainty in modern understanding of the influence of the plant biosphere on atmospheric chemistry and composition; and (2) investigate the impact of canopy-scale processes on regional-to-global aerosol and reactive trace gas abundances.

The project will support the participation of both graduate and undergraduate students in the research. It also includes an outreach program focused on the impact of the plant biosphere on air quality.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.