Applications of explainable machine learning for climate variability and change
Machine learning techniques, such as neural networks, are rapidly gaining traction across the sciences. In weather and climate, this surge reflects the growing role of AI in improving forecasts, accelerating model development, and making better use of the massive volumes of high-resolution observational and simulation data we now produce. However, there is also some hesitancy for considering the use of these algorithms due to concerns about their reliability, reproducibility, and interpretability. In this seminar, I will show examples of how relatively simple classification problems can be combined with explainability methods to improve our understanding of climate prediction and projection. Overall, we find that explainable neural networks are highly skillful in identifying patterns of forced signals within climate model large ensembles and observations. This is especially useful for disentangling regional responses to anthropogenic climate change versus natural variability, such as in detection and attribution applications. This same framework can be easily adapted for a wide variety of problems in the environmental sciences.
Bio: Zachary Labe ’15 is a climate scientist at Climate Central. His research interests explore the intersection of climate risk, climate impacts, future scenarios, extreme events, early warning predictions, and data science methods like AI/ML. In addition to this work, Labe is passionate about improving science communication through accessible data visualizations.
Prior to his time at Climate Central, Labe was a research physical scientist at NOAA Geophysical Fluid Dynamics Laboratory. He earned my Ph.D. in 2020 from the Department of Earth System Science at the University of California, Irvine, where he explored the links between Arctic climate change – particularly sea ice loss – and its influence on the large-scale atmospheric circulation and extreme weather. After his Ph.D., Labe completed postdocs at Colorado State University’s Department of Atmospheric Science and later at Princeton University’s Atmospheric and Oceanic Sciences Program. He first got into climate science through my B.Sc. in atmospheric sciences at Cornell University, where he was president of the American Meteorological Society Chapter.