Epscor research fellows: NSF: understanding low cloud evolution and their interactions with aerosols using observations, machine learning, and modeling

The National Science Foundation Office of Integrative Activities awarded the Desert Research Institute $299,567 on February 1, 2027, under the Integrative Activities program (CFDA 47.083) to support research on low cloud evolution and aerosol interactions. The award funds a fellowship for an assistant research professor and training for a graduate student at Desert Research Institute, with collaboration from Dr. Sarah Doherty at the University of Washington. The research investigates how marine low clouds evolve as they move across the subtropical eastern Pacific Ocean and how atmospheric aerosols affect their development and ability to reflect incoming sunlight. The project integrates satellite observations, Lagrangian trajectory analysis, machine learning, and large-eddy simulation modeling to identify environmental conditions and physical mechanisms controlling cloud evolution pathways. Deliverables include a comprehensive library of Lagrangian trajectories characterizing cloud evolution across aerosol and meteorological conditions, multiple machine learning models employing explainable artificial intelligence techniques to identify controlling atmospheric variables, and simulations of selected trajectories using large-eddy simulation. Work is performed in Seattle, Washington, with a period of performance from February 1, 2027, through January 31, 2029. The project aims to improve understanding of uncertainty in earth system prediction, strengthen institutional collaboration, train atmospheric scientists, and produce datasets and computational tools for the broader scientific community.

Abstract

This research infrastructure improvement (RII) epscor research fellows project provides a fellowship to an assistant research professor and training for a graduate student at the desert research institute. This work is conducted in collaboration with DR. Sarah doherty at the university of washington. Through the fellowship, the PI will investigate how marine low clouds evolve as they move across the subtropical eastern pacific ocean and how atmospheric aerosols affect their development and ability to reflect incoming sunlight. By combining atmospheric science, satellite remote sensing, machine learning, and physical modeling, the project will identify the environmental conditions and physical mechanisms responsible for different cloud evolution pathways and aerosol?cloud interactions. The project will improve understanding of an important source of uncertainty in earth system prediction, strengthen collaboration between research institutions, train the next generation of atmospheric scientists, and produce datasets and computational tools that will benefit the broader scientific community. The project will investigate stratocumulus-to-cumulus transitions and aerosol?cloud interactions over the northeast and southeast pacific by integrating satellite observations, lagrangian trajectory analysis, machine learning (ML), and large-eddy simulation (LES). A comprehensive library of lagrangian trajectories will be developed to characterize cloud evolution across a broad range of aerosol and meteorological conditions. Multiple ML models, together with explainable artificial intelligence techniques, will identify the atmospheric variables that most strongly control cloud properties and evaluate predictive skill against observations. Selected trajectories will then be simulated using the university of washington realistic LES under both standard and perturbed aerosol conditions to investigate aerosol impacts on cloud evolution and to compare the performance of ML and physically based numerical modeling. The fellowship will strengthen research capacity at the desert research institute through student training, enhanced collaboration with the university of washington, development of publicly available datasets, and expanded expertise in integrating observations, ML, and realistic LES. This project is supported by the epscor research infrastructure improvement program: epscor research fellows (ERF). The ERF program supports early- and mid-career investigators in eligible jurisdictions to develop collaborations at the nation?s private, government or academic research institutions. 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.- subawards are not planned for this award.