Climate Dynamics & Data Science
We are a research group in the Department of Climate, Meteorology, and Atmospheric Sciences and the Department of Earth Sciences and Environmental Change at the University of Illinois Urbana-Champaign, led by Cristi Proistosescu.
Our group studies the dynamics of Earth's climate system and how it responds to natural and anthropogenic forcing. We combine physical theory, numerical model simulations, and observational data using modern data science methods to understand climate variability and change across timescales — from the deep paleoclimate record to future projections of warming and extreme events.
Contact
Cristian Proistosescu
Department of Atmospheric Sciences & Department of Geology
University of Illinois Urbana-Champaign
cristi [at] illinois.edu
Research
Our group works on understanding how Earth's climate responds to forcing — and why that response is so hard to pin down. We combine theory, numerical model experiments, modern observations, and paleoclimate proxies to study radiative feedbacks, climate sensitivity, and the role of sea-surface temperature patterns in shaping both. A recurring theme is the interplay between forced and unforced variability, and how confounding the two leads to biased estimates of future warming.
We draw on both classic statistical methods and modern machine learning and AI to bridge analytical models, numerical experiments, and observations.
Current research spans coupled ocean-atmosphere dynamics, the physics of heat waves, paleoclimate variability, climate model evaluation, and the economics of climate risk. We are broadly interested in how uncertainty in the climate system propagates into uncertainty in impacts and policy.
People












Publications
-
arXiv:2603.24488, in review at GRL
-
arXiv:2507.15767, in review at JClim
-
Atmospheric Chemistry and Physics, 26(6), 4289-4311, 2026
-
Proceedings of the National Academy of Sciences, 123(4), e2511370123, 2026
-
Journal of Climate, 38(24), 7395-7413, 2025
-
Geophysical Research Letters, 52(3), e2024GL111626, 2025
-
Journal of Geophysical Research: Machine Learning and Computation, 2(3), e2025JH000774, 2025
-
Journal of Climate, 38(13), 3037-3053, 2025
-
Paleoceanography and Paleoclimatology, 40(10), e2024PA004991, 2025
-
Journal of Climate, 38(2), 513-529, 2025
-
Climatic Change, 177(5), 72, 2024
-
Science Advances, 10(16), eadk9461, 2024
-
Proceedings of the National Academy of Sciences, 121(12), e2312093121, 2024
-
Journal of Advances in Modeling Earth Systems, 16(2), e2023MS003700, 2024
-
Earth's Future, 12(10), e2024EF004844, 2024
-
Journal of Climate, 36(20), 7005-7023, 2023
-
Nature Climate Change, 12(6), 547-552, 2022
-
Proceedings of the National Academy of Sciences, 119(28), e2204761119, 2022
-
Geophysical Research Letters, 49(17), e2022GL100011, 2022
-
Geophysical Research Letters, 48(24), e2021GL095778, 2021
-
Journal of Climate, 34(21), 8777-8792, 2021
-
Journal of Climate, 34(21), 8717-8738, 2021
-
Reviews of geophysics, 58(4), e2019RG000678, 2020
-
Journal of Climate, 33(18), 7755-7775, 2020
-
Geophysical Research Letters, 47(7), e2019GL086588, 2020
-
Geophysical Research Letters, 47(5), e2019GL086705, 2020
-
Nature Climate Change, 10(2), 124-129, 2020
-
One Earth, 2(6), 515-517, 2020
-
Journal of Climate, 32(17), 5471-5491, 2019
-
Geophysical Research Letters, 46(1), 346-355, 2019
-
Geophysical Research Letters, 46(3), 1690-1701, 2019
-
Nature, 563(7729), E6-E9, 2018
-
Nature Climate Change, 8(12), 1076-1081, 2018
-
Geophysical Research Letters, 45(10), 5082-5094, 2018
-
Science Advances, 3(7), e1602821, 2017
-
Science, 352(6292), 1405, 2016
-
Geophysical Research Letters, 43(10), 5425-5434, 2016
-
Earth and Planetary Science Letters, 325, 100-107, 2012
Teaching
Applied AI for Earth Sciences
A graduate-level course on modern AI/ML methods in earth sciences. It cover fundamental ML concepts and architectures (e.g. regularization, convolutional neural nets, transformers), Python infrastructure (e.g. PyTorch) and modern applications to topics such as remote sensing, climate modeling and weather forecasting, and environmental hazard modeling.
The course is taught in a modern AI-assisted programming and analysis framework, teaching students how to effectively integrate generative AI tools (e.g., Large Language Models like CoPilot, and Claude) domain knowledge and conceptual understanding.
Climate Dynamics
A graduate-level class on climate dynamics. Investigates dynamical and physical processes that govern Earth’s past, present, and future climates. Emphasizes fundamental physical principles that determine present climate, and both natural and anthropogenic climate changes across spatial and temporal scales. Observations and climate models are used to examine past changes and potential future impacts.
Climate and Global Change
An introductory undergraduate course on climate change. Introduces climate change focusing, in turn, on mechanisms, impacts, and solutions The goal of the course is to provide you with a good understanding of Earth’s Changing Climate. By the end of the course, you will understand three major aspects:
(1) what physical and socio-economic factors drive climate change
(2) how climate change impacts natural, social, and economic systems
(3) what we can do about it!
Notes
A collection of personal notes and thoughts on climate dynamics and data science.