2026 SUMMER LECTURES IN CLIMATE DATA SCIENCE
May 28, 2026 - July 30, 2026
- 12:00 – 1:30 pm (EDT)
- IN-PERSON at the Columbia Engineering Innovation Hub (2276 12th Ave, New York, NY)
- VIRTUAL ATTENDANCE available
Click to see past Lectures in Climate Data Science from Spring 2026, Fall 2025, Summer 2025, Spring 2025, Fall 2024, Summer 2024, Spring 2024, Fall 2023, Spring 2023, and Fall 2022.
THURSDAY || MAY 28, 2026
LILY XU
Columbia University
“High-Stakes Decisions from Low-Quality Data: AI for Biodiversity Decisions”
The proliferation of big data has rapidly advanced global-scale monitoring of ecosystems and biodiversity. However, these advances in measuring species distributions and changing land cover are often followed by traditional decision-making processes that are reactive and heuristic-based. This talk will explore opportunities to leverage AI for not just monitoring nature, but also for making decisions — to design effective, timely interventions under resource constraints. I’ll present technical advances in machine learning, reinforcement learning, and causal inference, addressing research questions that emerged from on-the-ground challenges in wildlife conservation. Such interventions are the underpinnings of resource allocation, adaptive management, payment design, and other critical solutions for nature. I’ll also discuss ongoing work on invasive species management and optimizing biodiversity monitoring.
THURSDAY || June 11, 2026
COURTNEY COGBURN
Columbia University
Climate, Science + Justice: Why Broad Participation and Knowledge Transfer Makes Better Science
Climate data science doesn’t happen in a vacuum — the questions we ask, the data we trust, and the communities our models serve are all shaped by who is in the room. This session makes the case for broadening participation and knowledge transfer not as an add-on to rigorous research but as a condition of it: richer questions, more robust methods, and science that reaches the people a changing climate affects most. Drawing on the uneven distribution of climate burden and the makeup of the field itself, we’ll consider how a scientific community’s composition shapes what it is able to know. Through short reflective exercises, participants will map their own disciplines, assumptions, and spheres of influence as emerging climate scientists.
THURSDAY || June 18, 2026
CHAD SMALL
University of Washington
“The Art (and Science) of Science Communication for Earth Scientists
Science communication – clearly and comprehensively delivering scientific information to non-technical audiences – is becoming increasingly important. Whether it’s for helping people understand the world around them, or building public trust in the scientific process, science communication is being featured more prevalently in researchers’ toolkits. For geoscientists, science communication is becoming essential as anthropogenic climate change tips the scales toward more environmental crises. But if we’re supposed to increase scientific literacy on both general and policymaking levels, there needs to be an understanding of which parts of our work are most salient to those outside our field. In other words, we need to know which parts of our research tell the stories that need to be heard the most. This interactive workshop will unpack how we should communicate research findings to different audiences. There will be a special focus on how geoscientists should respond when asked for comments by the media, and also how we can craft our own independent research-based stories.
THURSDAY || June 25, 2026
OLIVER WATT-MEYER
Allen Institute for AI
AI for Fast and Flexible Climate Modeling
AI-driven weather forecast models are now more accurate and faster than the best
physics-based systems. Extending these advances to seamless weather–climate prediction poses broader challenges, but progress is rapid. Several models trained on ERA5 capture historical variability and trends. The key question is whether such systems can generalize to project future climate reliably. Purely data-driven extrapolation remains elusive, but emulators of physics-based models trained across multiple climates are emerging as a promising route. These can generate ensembles of ocean-coupled simulations that are statistically consistent with their reference models but at orders-of-magnitude lower cost.
I present results from the open-source Ai2 Climate Emulator (ACE). ACE emulates daily weather variability and climate at 100 km resolution, running ~1500 years/day on a single GPU—about 100× faster than comparable physics-based models. It can be trained on ERA5 or AMIP-style forcings, paired with AI downscaling to km-scale weather, or coupled to slab-ocean models to capture climate change responses. Most recently, when coupled to Samudra, a full-depth ocean emulator, ACE reproduced stable coupled climate states and realistic El Niño–Southern Oscillation variability. Together, these advances point to AI emulators enabling the rapid generation of climate prediction information for a wide range of user-specific needs.
WEDNESDAY || July 1, 2026
JOSHUA FISHER
Chapman University
The Fate of the Terrestrial Biosphere
One of the largest uncertainties of the future of the Earth is how the terrestrial biosphere will counteract or exacerbate the rise in CO2. Whether or not ecosystems absorb or emit carbon depends on water, temperature, nutrients, CO2 fertilization, and other factors. Here, Dr. Joshua Fisher will discuss how satellite and airborne remote sensing in conjunction with climate models are used to understand how Earth’s terrestrial water, carbon, and nutrient cycles are linked and impact the Earth system as a whole, highlighting new insights into the behavior and understanding of the terrestrial biosphere in a changing climate.
THURSDAY || July 9, 2026
KARA LAMB
Columbia University
Automating Parameterization Development for Earth System Models with Agentic AI
Earth System Models (ESMs) encode our knowledge about the physical world, enabling short-term weather and long-term climate prediction. Because these models cannot explicitly resolve all relevant physical processes, they rely on simplified parameterizations to represent sub-grid-scale processes such as convection, cloud microphysics, and turbulence. While recent advances in machine learning (ML) have enabled impressive emulation of model components and even full model surrogates, physically grounded parameterizations remain essential: they provide interpretability, scientific insight, and robustness for out-of-distribution prediction. However, it often takes years to translate new physical understanding into stable, validated parameterizations integrated into ESMs. Current workflows are fragmented, highly manual, and dependent on expert intuition, creating a major bottleneck in the pace of model development.
Recent advances in agentic AI offer a promising alternative that could substantially accelerate parameterization development and integration, while enabling more systematic exploration of design choices. In this talk, I will discuss progress toward an agentic AI framework that can synthesize scientific literature, analyze simulation data, generate parameterizations, tune parameters, and evaluate performance within a JAX-based Earth system model. The framework combines both physics-based and data-driven approaches to develop parameterizations from theory and high-fidelity simulations while maintaining physical consistency and computational efficiency. I will focus on applications to atmospheric cloud microphysics and discuss how agentic workflows can accelerate scientific discovery, reduce development time, and enable more systematic exploration of parameterizations in next-generation Earth System Models.
THURSDAY || July 16, 2026
DON EDMONSON
Columbia University, College of Physicians + Surgeons
From Climate Projections to Clinical Risk: Data Science for Identifying Heat-Vulnerable People, Places, and Mechanisms
Extreme heat is already increasing cardiovascular morbidity and mortality, but current warning systems rely heavily on outdoor weather metrics that may poorly reflect the heat people actually experience indoors. This lecture will describe a climate-health data science program that links outdoor climate data, indoor microclimate monitoring, neighborhood heat vulnerability, building characteristics, electronic health records, ambulatory ECG, actigraphy, and daily behavioral data to identify heat-vulnerable people, places, and mechanisms. Focusing on adults with hypertension and diabetes in New York City, I will discuss how rising heat may produce acute cardiovascular risk through autonomic strain, impaired thermoregulation, sleep disruption, and unequal access to cooling resources. The talk will highlight methodological challenges central to climate data science: exposure misclassification, multiscale data integration, within-person modeling of acute heat responses, and translation of risk models into clinical and public health decision tools. The goal is to show how climate data science can help move from general warnings about heat toward actionable, mechanism-informed strategies for protecting the patients and neighborhoods most at risk.
THURSDAY || July 23, 2026
ANN LEE
Carnegie Mellon University
Trustworthy Predictive Distributions for the Physical Sciences
THURSDAY || July 30, 2026
SUMMER 2026 REU FINAL PRESENTATIONS