Caltech_Logo
Logo_Lockup-sep
c-dg_wordmark

Our Research

Climate Modeling

Numerical climate modeling is computationally demanding, often relying on legacy Fortran code. Inherited inefficiencies from past decades, coupled with uncertainties in parameterization schemes, contribute to limitations and uncertainties in climate projections. Our researchers and software engineers collaborate closely to leverage cutting-edge technologies and modern software engineering practices. By developing and supporting public code in Julia, we directly contribute to the wider scientific software community, aiming to build a more accurate, usable, and extensible climate model.

illustrated figure for Climate Modeling page section
Credit: Schneider, T., et al., 2017
illustrated figure for Large-Scale Atmospheric Dynamics page section

Large-Scale Atmospheric Dynamics

We explore fundamental questions about large-scale atmospheric dynamics, including tropical circulations, midlatitude storm tracks, and their interactions. Using a variety of tools—from idealized climate models to coupled earth system models and observations—we aim to uncover the fundamental physical laws governing Earth’s climate. These insights serve as a basis for understanding future changes in extreme temperatures, global rainfall patterns, and other critical climate impacts.

Statistics and Machine Learning for Climate Science

Earth-observation satellites collect terabytes of data daily about land, ocean, and atmosphere. Recent advances in computing power have enabled scientists to extract insights from decades of data. Our team pioneers novel machine learning and data assimilation techniques to harness Earth observations, enriching climate models and enhancing prediction capabilities.

illustrated figure accompanying Statistics and Machine Learning page section
Credit: Schneider, T., et al., 2017
Illustrated figure for Clouds, Convection and Cloud Microphysics page section
Credit: Pressel, K., et al., 2015

Clouds, Convection and Cloud Microphysics

Clouds are a major source of uncertainty in future climate projections. Our objective is to enhance our understanding of turbulence, convection, and clouds and apply this knowledge to refine or redesign parameterizations in climate models. Researchers in our group use idealized and comprehensive models, combined with observational data, to advance our understanding of cloud processes across scales, from the microscale to the kilometer scale.