News Teaching Publications Group Research
UMD

M

Welcome! We focus on Predictability and Applied Research for the Earth-system with Training and Optimization (PARETO). As such, we use machine learning (e.g., neural networks) and numerical modeling systems (e.g., CESM) to answer pressing questions and address challenges in modeling the Earth system. Examples of problems that we are tackling include:

☁ extending our understanding of Earth system predictability,
☁ parameterizing subgrid scale processes in Earth system models, and
☁ uncovering multi-scale and causal patterns in the climate system.

Our research also strives to incorporate open-source software and data, accessible communication, and multi-discipline collaboration (particularly with computer science).

Our group name, PARETO, is inspired by the Pareto frontier, a fundamental concept in optimization theory. The concept was named after the Italian economist Vilfredo Pareto (1848-1923) and delineates trade-offs between competing objectives.

group-photo

Fall 2024 group photo. Pictured from left to right: Luke Wichrowski, Kyle Hall, Sandy Kerr, Sunny Sharma, Emily Wisinski, Dean Calhoun, Jon Starfeldt, Jhayron S. Perez-Carrasquilla, Assistant Professor Maria J. Molina, and Visiting Postdoctoral Fellow Manuel Titos. Learn more about our group [here].

If you are interested in joining our group as a graduate researcher, please note that all interested applicants must apply online to be considered.


Recent News

☁ [October] Maria J. Molina is joining AGU's Journal of Advances in Modeling Earth Systems (JAMES) as an Associate Editor.

☁ [October] Maria J. Molina was featured on the AMS blog, The Front Page, for Hispanic Heritage Month. [Read More]

☁ [September] Jonathan Starfeldt and Maria J. Molina are presenting this week at the 6th NOAA AI Workshop, which focuses on heat resilience this year. Read more [here]!

More news available [here].

Recent Publications

☁ Schreck, J. S., D. J. Gagne II, C. Becker, W. E. Chapman, K. Elmore, G. Gantos, E. Kim, D. Kimpara, T. Martin, M. J. Molina, V. M. Pryzbylo, J. Radford, B. Saavedra, J. Willson, and C. Wirz (In Press). Evidential Deep Learning: Enhancing Predictive Uncertainty Estimation for Earth System Science Applications. Artificial Intelligence for the Earth Systems. [Link]

☁ Beadling, R. L., P. Lin, J. Krasting, W. Ellinger, A. Coomans, J. Milward, K. Turner, X. Xu, T. Martin, and M. J. Molina (2024). From the surface to the stratosphere: large-scale atmospheric response to Antarctic meltwater. Geophysical Research Letters. [Link]

☁ *Chan, M. A., M. J. Molina, and C. A. Metzler (Accepted). Estimating Epistemic and Aleatoric Uncertainty with a Single Model. Thirty-Eighth Annual Conference on Neural Information Processing Systems (NeurIPS 2024); Main Track. [Preprint]

More publications available [here].


Any opinions, findings and conclusions or recommendations expressed herein do not necessarily reflect the views of the University of Maryland.

MRG